[Advances in Ecological Research] Advances in Ecological Research Volume 13 Volume 13 || Throughfall and Stemflow in the Forest Nutrient Cycle

  • Published on
    07-Apr-2017

  • View
    214

  • Download
    2

Transcript

Throughfall and Stemflow in the Forest Nutrient Cycle G. G. PARKER I. Summary . . 58 11. Introduction . . . 58 111. Definitions . . 61 A. Hydrological . . 61 B. Chemical . . . 62 IV. Magnitude and Importance of the Fluxes . . 65 A. Throughfall . . 65 B. Stemflow . . 69 C. Seasonality. . . 72 D. Recycling Rates . . . 73 E. Variability . . . 74 V. Factors Affecting Throughfall Quality. . . 75 A. Amount of Precipitation . . 76 B. Latitude . . 71 C. Proximity to Sources . . 78 D. Forest Type . . 80 E. Stand Age . . . 81 F. Site Fertility . . 82 G. Insect Consumption . . 84 H . Other Factors . . 84 VI. Processes of Throughfall Enhancement . . 85 A. Incident Precipitation . . . 86 B. Evaporative Concentration . . . 86 C . Inter-Event Deposition . . 87 D. Leaching . . . 90 VII. Partitioning Net Throughfall . . 92 A. Direct Approaches . . 92 VIII. Elements in Throughfall . . . 98 A. Carbon . . 98 B. Hydrogen . . . 98 C . Potassium . . . 99 D . Calcium . . 99 E. Magnesium . . 100 E. Foliar Uptake . . 91 B. Indirect Approaches . . 94 58 G. ti. PARKER F. Sodium . . 100 G. Phosphorus . . 100 H. Nitrogen . . 101 J . Sulphur . . 102 1. Chloride . . . 101 K . Silicon, Iron and Aluminium . . 103 L. Heavy Metals . . 103 IX. Recommendations for Future Research . . 104 X. Conclusions . . 105 XI. Acknowledgements , . . 106 References . 101 Appendix . . 121 I. SUMMARY The quality of precipitation falling on forests is altered during a brief but significant interaction with the surfaces of plants, resulting in the transfer of additional mineral matter to the forest floor. While incident precipitation is the largest nutrient input to many forests, throughfall fluxes are substantially greater, ranging from 1.27 (for NO,-N) to 11.2 times as high (for K). These alterations in nutrient concentrations involve numerous processes and combine materials originating both within (biotic) and outside of the forest ecosystem (atmospheric). The resulting flux is a pathway of rapid mineral cycling, a route for transferring metabolic biproducts and substances of allelochemic and pedogenic importance and a method for washing the air filtering surfaces of the forest canopy. Throughfall and stemflow are major pathways in nutrient recycling. The annual nutrient return to the forest soil for the elements K, Na and S is predominantly via throughfall and stemflow and little due to litterfall. Stemflow transfers only 5-20% of the total in precipitation-borne solutes, yet it is the major nutrient input to restricted areas of the forest floor. In this review 1 discuss a number of factors influencing throughfall and stemflow quality and their variation. Particular attention is paid to the effect of the canopy in altering precipitation quality, and to the numerous processes involved. Foliar leaching is concluded to be the major process controlling throughfall and stemflow enhancement for nearly all elements, though foliar uptake and canopy filtration of dusts, aerosols and gases is important in particular situations. 11. INTRQDUCTION Precipitation is an important source of nutrient input to forested ecosystems (e.g. Nye, 1961; Miller. 1963; Likens ef d., 1977; Swank and Henderson, THROUGHFALL, STEMFLOW IN FOREST NUTRITION 59 1976), especially where rock weathering is slow (Gorham, 1953; Jordan et a/., 1980). Since nutrient transfers in throughfall and stemflow are usually substantially larger than those in incident precipitation, an increasing amount of interest has been directed at this flux (e.g. Will, 1955; Yawney et a/., 1970; Szabo, 1977; Killingbeck and Wali, 1978; Prebble and Stirk, 1980; Khanna and Ulrich, 1981). Estimation of the fluxes of elements in incident pre- cipitation, throughfall and stemflow is now a routine part of nutrient budget studies in forests (e.g. Corlin, 1971; Tsutsumi, 1971; Likens et a/., 1977; Schlesinger, 1978; Kelly, 1979; Clesceri and Vasudevan, 1980; Sollins et a/., 1980). The alteration of the composition of water in contact with plant tissues has been recognized since de Saussure in 1804 (Tukey, 1970a). That throughfall may be important in plant nutrition and soil fertility is a comparatively recent idea, stemming from Ingham ( 1950a.b) and Tamm ( 1 950). The material added to incident precipitation due to the action of the canopy, was found to derive both from within (Tamm, 1950,1951)and from upon (Ingham, 1950a,b)plant surfaces. Will (1955) first showed the importance of forest rainwater relative to the nutrient flux in litterfall for Na, K and Mg. Madgwick and Ovington (1959) were the first to report that different forest types have unique effects in changing the concentration of precipitation. Studies of the chemical composition of throughfall and stemflow are now commonplace and increasing in number; most of the reports on throughfall and stemflow quality reported here have been published within the past 5 years. In the past interest was directed at (1) factors affecting the hydrological partitioning of precipitation into interception, throughfall and stemflow; (2) the quantification of nutrient fluxes under various climates and canopies, and (3) the examination of factors controlling throughfall composition, its seasonality and heterogeneity. Recently, much attention is being focused on the external and internal sources of throughfall enrichment (Ulrich et a/., 1978; Lakhani and Miller, 1980; Miller and Miller, 1980; Parker et a/., 1980). The composition of throughfall and stemflow has been studied in a number of geographical locations (Fig. l), especially populated western hemisphere regions. Some of the most extensive forests remain unstudied. High latitude forests have received little attention (Kazimirov and Morozova, 1973; Johnson, 1975; and Van Cleve, in Cole and Rapp, 1981) and there are only two descriptions of throughfall composition from the Amazon basin (Nort- cliff and Thornes, 1978; Jordan et a/., 1980). In this review the processes of throughfall (precipitation which falls through the canopy) and stemflow (precipitation which flows down the tree stem) are considered in the context of the nutrient economy of forests. I examine the quality, magnitude and timing of these fluxes relative to other major nutrient pathways, with consideration of their biological, geographical, hydrological, and meteorological influences. Particular attention is paid to the problem of estimating the sources of throughfall enhancement (recycling and input). Fig. 1. Studies of throughfall and stemflow chemistry around the world relative to the distribution of forests. Shaded areas are forested. Adapted from Dansereau (1957); Strahler (1973); Dassman (1976) and Neil, York Ti/tie.s (1978). THROUGHFALL, STEMFLOW IN FOREST NUTRITION 61 This review will not concern itselfwith aspects ofthroughfall quality that d o not pertain to the biological cycle of nutrient elements. Thus I shall avoid discussion of the information-rich allelochemic compounds whose effects are surely disproportionate to their weights (e.g. Whittaker and Feeney, 1971). leachates which can mobilize soil iron. aluminium and manganese (e.g. Malcolm and McCracken, 1968) and the exudates which may drip from the canopy in dry periods (Browne, 1932; Durant, 1932; Carrier, 1958: Wyatt- Smith, 1958). 111. DEFINITIONS A. Hydrological Since throughfall and stemflow are associated with precipitation events, the transport of nutrients contained in throughfall and stemflow depends on the magnitude, timing and form of the precipitation. Thus a reliable estimate of throughfall nutrient flux demands a good forest hydrological budget. Some essential concepts in throughfall hydrology are defined and discussed in the following section. Water falling on the forest is called incident precipitation (gross pre- cipitation (Leonard, 1966), wetfall or grossfall (Lindberg ct d., 1979)). Most commonly it impinges upon the canopy vertically (direct precipitation) but often it is laterally advected and intercepted by the canopy (indirect or occult precipitation, (Kittredge, 1948), which is important for mist, fog or cloud water. Direct precipitation is measured either above the canopy (e.g. Attiwill, 1966; Lindberg et a/., 1979) or, more commonly, in open areas adjacent to the forests. Indirect inputs are more difficult to measure (Falconer and Falconer. 1979; Falconer and Kadlecek, 1980). Precipitation which passes through the canopy and falls to the ground is called throughfall (net or effective rainfall (Helvey and Patric, 1965), canopy drip, crown runoff (Bache, 1977), drainage water (Brosset, 1976), leafwash (Cole and Rapp, 1981), rainwash (Nye, 1961; Whittaker and Feeney, 197 1 ) or pluviolessivage (Denaeyer-DeSmet, 1966; Rapp, 1969; Lemee, 1974)) and may exceed incident precipitation in quantity at some locations (Zinke, 1962; Banaszak, 1975; Prebble and Stirk, 1980). Throughfall includes incident precipitation which penetrates canopy gaps. unless the individual gaps are large and frequent (e.g. > lo",, of the area). An additional portion of precipitation reaches the ground by running down the branches and trunk, depositing at the base of the tree. This portion is called stemflow (trunk or stem runoff). The sum ofthroughfall and stemflow is called total forest water (net precipitation or net rainfall, in Zinke. 1966). Incident precipitation that does not appear on the forest floor by either of these routes is called the interception loss (Kittredge, 1948). This includes 62 G. G. PARKER ( I ) stemflow which never reaches the ground, (2) water evaporated from the canopy during the course of a storm (Leonard, 1966) and (3) the amount of water held in the canopy and evaporated after the storm, the canopy saturation (Zinke, 1966). The latter portion includes the canopy storage capacity (Zinke, 1966), the maximum amount of water the canopy can retain at a given time, which ranges from 1-5mm and appears to be relatively constant for a given forest and season, though affected by amount and intensity of precipitation (e.g. Leonard, 1966). As it cascades through the canopy, a unit of water interacts with numerous plant surfaces. For example, nearly all stemflow encounters leaf surfaces first (Jordan, 1978) and throughfall may include water dripping from twigs and branches. However, the hydrological classification depends on the last surface encountered. If that surface is the tree trunk then the precipitation is termed stemflow, otherwise it is throughfall. Where measurements are made at several levels in a multi-storied canopy, interception and throughfall can be further subdivided according to the number of strata defined (Patterson, 1975); even the litter layer may be considered as a canopy (Helvey and Patric, 1965). Stemflow is not similarly partitioned. The compartments of the forest hydrological budget are summarized by the equation (after Helvey and Patric, 1965): R = I + T + S where R , I, T, S are incident precipitation, interception loss, throughfall and stemflow respectively, all values given in linear dimensions (mm). Only these terms will be employed in this treatment. These hydrological conventions are diagrammed in Fig. 2. The quantity and distribution of throughfall and stemflow depends on microscale features of canopy structure, such as, crown density (Anderson et al., 1969; Paivanen, 1974; Nicholson et al., 1980), closeness of the foliar elements, distance from the nearest bole (Prebble and Stirk, 1980), or to open spaces in the canopy (Tamm, 1951). Trees with drooping branches can produce much throughfall at the canopy perimeter. Quantity of stemflow is certainly affected by bark smoothness (Voigt 1960a, Voigt and Zwolinski, 1964), stem diameter (Kittredge, 1948) and branch angle (Mina, 1965; Hutchinson and Roberts, 198 I ) . Large-crowned emergent trees with smooth bark and raised branches produce the most stemflow. B. Chemical In a forest the flux of nutrients in precipitation is closely related to the amount of precipitation. However, there are important differences between nutrient concentration and water quantity: nutrient fluxes are quite variable and not directly obtainable from the water budget. Variability in nutrient fluxes is treated in Section IV. E. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 63 Fig. 2. Compartments and fluxes in throughfall hydrology. I use the terminology proposed by Carlisle rt a/ . (1966b) and extended by Gosz el a/. ( 1 976) to describe nutrient transfer by various pathways. The flux of nutrients onto the forest by either direct or indirect precipitation is termed incident precipitation (INC), which is further called bulk precipitation if the collector is open to dry deposition between storms (Whitehead and Feth, 1964) or wet precipitation otherwise (Galloway and Parker, 1980). For a given nutrient the total flux of precipitation borne nutrients to the forest floor is the sum of the materials falling through the canopy (throughfall, designated as THF) and travelling down along the trunk (stemflow, STF), both ofwhich are conventionally expressed as elemental mass deposited per unit area per unit time. Such fluxes are termed depositions (e.g. Richter and Granat, 1978; Galloway and Parker, 1980). When the appropriate nutrient deposition in incident rainwater is subtracted from that in either throughfall or stemflow (Carlisle c>r a/., 1966b) the remainder is called net throughfall (NTF) or net stemflow (NSF), either of which can be negative. The total effect of the canopy on these depositions is obtained by subtracting the incident precipitation deposition from the sum of the throughfall and stemflow depositions. This quantity has no accepted name (Eaton rt al. (1973) call it "net stemflow and throughfall") and will be designated in this treatment as net forest water (NFW). These conventions are summarized in equation form below (Table 1). D;, D, and D, are the amounts (depth) of water sampled in incident precipitation, 64 G. G. PARKER throughfall and stemflow; C,, C, and C', are the respective solute con- centrations of a given nutrient (concentrations applying to depositions for more than one event should be volume-weighted). When amounts of water and concentrations are in compatible units one may calculate the deposition of nutrient elements (given by capitalized symbols) according to the following lo rm ulae . Table I Definitions and equations relating precipitation amounts and nutrient concentrations to nutrient depositions. ~ ~~ ~. ~ ~- D ~ f i t 1rd I vrt? I S amount of precipitation ionic concentration gross deposition net deposition C'ull~ulrtl~cl,pll.~-l~.~ ~~~~~ ~ Precipitation pathway incident precipitation through fa I I stenif ow . ~- ~. ~ ~~ ~~~ _ _ ~- D , D, D , c, c-t C, INC = D,C', THF=D,C', SF = D,C, N T F = D,(C', C,) NSF= D,(C, C,) Several enrichment factors may be defined. The ratio of an element's concentration in throughfall (THF) to that in incident rainwater (INC) is called the concentration ratio (CR) or the net concentration ratio (NCR), for net throughfall. The deposition and net deposition ratios (DR, NDR) are analogously obtained. These are summarized below: Concmtrut ion Rut ios CR = C,/C'; NC R = (c', ~ C;)/C'i Deposit ion Rutios DR = D,c',/D;C, NDR = D,(C', C;)IDjCi Note that the concentration ratio usually exceeds the deposition ratio since the volumes of water differ, by the amount intercepted by the canopy. That is, C R [ D , / D , ] = DR. Net depositions should be adjusted for the appropriate fraction of incident precipitation only (Carlisle i>t d., 1966b). Therefore. net stemflow equals the total stemflow deposition minus the fraction of incident deposition that arrives as stemflow ( D s C j ) , usually less than 157;, of the nutrient deposition in incident precipitation. For throughfall, the effect of the canopy on nutrient flux is not strictly the simple difference between total throughfall and incident THROUGHFALL, STEMFLOW I N FOREST NUTRITION 65 deposition since not all incident precipitation produces throughfall, some being stemflow or interception loss. When defined in this manner, the net effect of the forest is the sum of net stemflow and net throughfall. NFW = D,C, + D,,C'., - N T F + N S F DiC, =TH F + STF ~~ INC The sum of net throughfall and net stemflow is thus a slight underestimate of the net forest water flux. by an amount equal to (0; ~ D, ~~ D,)C;, the mass of nutrients contained in intercepted water. The interception loss is eventually evaporated from the canopy, but the nutrients contained in that fraction of incident precipitation either revolatilize, precipitate onto the canopy or appear in throughfall and stemflow. This small amount (10-207,, of INC, less than 5';,, of T H F for most elements) is commonly ignored. IV. MAGNITUDE AND IMPORTANCE OF THE FLUXES Throughfall and stemflow are important flux pathways in the internal nutrient dynamics of the forest. These fluxes must be considered in the estimation of ( I ) annual plant uptake, for they include material absorbed from the soil (Remezov, 1961; Duvigneaud and Denaeyer-De Smet, 1970, 1975; Schlesinger 1978; Cole and Rapp, 1981) and ( 2 ) turnover times for nutrient pools in both forest canopy and forest floor (Gosz et al., 1976; Henderson e f al., 1977; Szabo, 1977). Furthermore throughfall and stemflow contain some materials of atmospheric origin, which must be considered in constructing input/output budgets. The extent to which input and recycled materials may be confounded is deferred to Sections VI and VI1. A. Throughfall 1. Drgrre of' Enhancement The amounts of solutes delivered to the forest floor in throughfall is almost without exception, much greater than the amounts received in incident precipitation. In throughfall the combined weight of all solutes can amount to several hundred kilograms per hectare annually or higher. The marked enhancement of throughfall concentrations relative to those in incident precipitation is clear from Table 2. The distribution of reported concentrations is skewed especially for sea salt elements Na, Mg and C1 and also for N03-N and S04-S (in industrial regions). The influence of canopies in altering the concentration or nutrient mass of Table 2 Volume weighted mean concentrations of throughfall and incident precipitation (mg element per litre) and their standard deviations. Calculated from the Appendix Table and other reports. N, NH4-N NO,-N P T K Ca Mg Na C1 so,-s Incident precipitation 0.98 0.36 0-31 0.12 0.52 0.82 0.40 1.27 1.01 1.43 0.92" 0.25 0.36 0.19 0.58 0.94 1.02 2.50 0.66 1.10 ~~ - Throughfallh 1.57 0.72 047 0.3 1 3.72 2.58 1.39 4.97 4.15 3.90 1.47" 0.76 0.67 0.82 2.99 2.03 3.43 7.46 3.12 4.84 Standard deviations. Includes stemflow in some cases THROUGHFALL, STEMFLOW IN FOREST NUTRITION incident precipitation may also be indicated by the concentration or deposition ratio, respectively. Mean deposition ratios of commonly analysed compounds for reported studies are given in Table 3 (below). Precipitation amounts are always reduced (DR < 1). Elemental depositions are generally increased: up to 2-fold for nitrogen species, more for sulphate, chloride, basic cations and phosphorus. Potassium nearly always exhibits the most dramatic relative increases (> 1 I-fold). 67 2. Relative to Other Nutrient Fall Pathways Net throughfall often contains a major fraction of the total amount of nutrients falling to the forest floor via precipitation and litterfall. For S, K, Na and Mg, net throughfall is the largest pathway of nutrient fall. The net throughfall contribution to the total yearly nutrient fall ranges from 40-65% for potassium, 15-35% for magnesium, 10-202, for phosphorus and calcium, and 0-1 5% for nitrogen. Where measured, sodium, manganese and sulphur in net throughfall contribute 34.0% (1 1 years of data), 20.2% (n = 3) and 35.5% (n =4), respectively. Net throughfall carbon supplies about 5% of nutrient fall (n = 2 ) . Differences between elements in pathways of nutrient fall may be more clearly illustrated by ternary diagrams (Fig. 3). Each side of the equilateral triangle is an axis for the percentage of nutrients falling due to either litterfall (LF), incident (INC) or net throughfall deposition (NTF). Such a pre- sentation is possible for 3 component systems because in an equilateral triangle the sum of the perpendiculars from any point is constant. In each region of this diagram are different modes of nutrient movement to the forest floor. Near the top of the triangle is clearly input (in incident precipitation) and outside the left side of the triangle is foliar uptake, where net throughfall is negative. Towards the triangle base, pathways important in recycling dominate: to the left-hand side, decomposition of litterfall and to the right, leaching. There is much disagreement regarding the source of materials in the right central region of the diagram, which will be more fully addressed in Sections VI and VII. Such diagrams are presented to clarify the various pathways for the elements K, Mg, Ca, S, P, N, CI and Mn in Fig. 4 (each datum represents the partitioning of mineral input to the forest floor for a separate year's data as calculated from the literature). Each element appears to fall to the forest floor via a unique combination of pathways, since points for each element are clustered in different areas of the graphs. Nitrogen inputs to the soil derive largely from litterfall and some incident precipitation; the canopy adds little soluble nitrogen to precipitation (in fact, it often removes some). Calcium and phosphorus also fall via litter, with some net throughfall and incident precipitation contributions. Manganese derives from litterfall and some net throughfall, with nearly nothing from incident precipitation. Magnesium is Table 3 Mean deposition ratios (DR). Calculated from Appendix p. 121" SO,-S H2O ~- NH,-N NO,-N NT' P K Ca Mg Na CI ~~ R 1.63 1.27 1.90 3.94 11.2 2.86 4.00 2.41 3.03 2.26 .76 S.D. 1.28 .89 1.48 3.30 22.1 2.25 4.18 1.71 1.49 1.39 . I 1 n' 31 31 46 60 94 91 84 42 19 33 60 *Arithmetic means and standard deviations. Includes Total Kjeldahl Nitrogen, ' Number of years of data averaged THROUGHFALL, STEMFLOW IN FOREST NUTRITION Inc tdent precipitation 69 ~ L i t terfol l Net t hroughfoll Fig. 3. Explanatory ternary diagram for the pathways of nutrient fall to the forest floor. Each perpendicular of the triangle is an axis corresponding to a pathway of nutrient fall, either incident precipitation, net throughfall or litterfall. Every point in the diagram is a unique combination of these pathways. similar to manganese, but with greater net throughfall contribution. Net throughfall accounts for most of potassium inputs, though litterfall can be important. Sodium and sulphur inputs are due to incident precipitation and net throughfall, litterfall adding very little. The points within the clusters for each element do not appear to separate according to stand type or latitude (which affect the total flux but not the partitioning of nutrient fall to the forest floor). Differences in proximity to sources of dry deposition can, however, affect the partitioning of nutrient fall (Rapp, 1969; Baker ct al., 1977) as discussed Section V. C. B. Stemflow I . Relative to Throughfull Stemflow transports a smaller amount of material to the forest floor than throughfall. Its percentage contribution to the flux of water borne nutrients is between 1 and 20';/,, averaging about 12%, depending on the element and the type of stand. Figure 5 is compiled from numerous studies reporting the contribution of nutrients in stemflow as a percentage of the total in stemflow 70 G . G. PARKER Fig. 4. The pathways of nutrient fall to the forest floor. Percentages of total fall in incident precipitation (INC), net throughfall (NTF) and litterfall (LF) in world forests for nitrogcn, phosphorus, potassium. calcium, magnesium. sodium sulphur and manganese. From literature reports. and throughfall for precipitation amounts and commonly measured nutrient elements. In stands where stemflow is a significant part of the annual forest water flux, it provides large nutrient depositions of calcium, sulphate, sodium and manganese (e.g. Mayer and Ulrich, 1972), and smaller depositions of potassium, magnesium, chloride, phosphate and organic matter and much less of most nitrogen species. Nutrient deposition in stemflow appears to be most important in forests of smooth-barked species (e.g. beech, aspen, poplar) and not so for rough-barked ones (e.g. spruce) (Voigt and Zwolinski, 1964; Mayer and Ulrich, 1972; Ulrich cjt al., 1978). THROUGHFALL, STEMFLOW IN FOREST NUTRITION 71 Though it is more infrequent than throughfall (Bol\en er a!., 1968; Stark 1973), stemflow is almost always the largest pathway of water (e.g. Voigt 1960a) and nutrient input (e.g. Stark, 1973) to the stemflow zone of the forest floor. Elemental concentrations in stemflow are distincly higher than those in throughfall (e.g. Iwatsubo and Tsutsumi, 1967), by up to an order of magnitude. Stemflow has a pH characteristically much lower than that of throughfall (Pozdnyakov, 1956; Carlisle et ul., 1967; Mahendrappa, \974; Nicholson et al., 1980) and has high concentrations of Ca, K, S, and Mg and particulate organic matter (e.g. Mina, 1967; Mahendrappa, 1974). 2. Spatial Distribution On the forest floor, the stemflow input is restricted to small, irregularly-shaped regions around the boles of individual trees (Helvey and Patric, 1965; Gersper and Hollowaychuck, 1971). Estimates of the width of the inundated area around the bole vary from 0.3 to 5 m (Voigt, 1960a; Carlisle et ul., 1967; Bollen rt al., 1968; Gersper and Hollowaychuck, 1970, 1971; Abee and Lavender, 1972; and Stark, 1973; Szabo, 1977). The area of the stemflow zones on the forest floor probably totals no more than the stand basal area, in mz ha- ' (Abee and Lavender, 1972). The limited distribution of stemflow is important N" "I 0 3 S Mn mTT7W 7 - 7 T - 1 1 ' - 1 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Sternflow as a percentage of total forest water Fig. 5. Histograms of the percentage stemflow contribution to nutrient deposition in total forest water (SF/(THF + SF)). From literature reports. 72 G. G. PARKER for both the trunk epiflora (Nye, 1961; Rasmussen and Johnsen, 1976; Nieboer c'f a/., 1978; Pike, 1978) and the soil about the bole (Mina, 1965; Gersper and Hollowaychuck, 1970, 1971; Patterson, 1975). For example various soil parameters may decline with distance away from the bole, especially on the side of the tree receiving most stemflow (Gersper and Hollowaychuck, 1971), including soil moisture (Lunt, 1934; Kaul and Billings, 1965)- acidity and concentrations of nutrients and radioactive cesium (Mina, 1967; Gersper and Hollowaychuck, 1971; Patterson, 1975). However, soil in the stemflow zone may not necessarily differ from the rest of the forest floor in nutrient concentrations, as found by Bollen et a/. (1968) and Tarrant ct ( I / . (1968). Throughfall and stemflow differ in the surfaces encountered, their area, texture, composition and associated biota and the amount of time in contact with those surfaces (Mahendrappa and Ogden, 1973; Gersper and Holloway- chuck. 1971). In deciduous forests both fluxes show similar seasonality (Eaton et d., 1973). Carlisle et a/. (1967) found significant correlations between monthly concentrations in throughfall and stemflow for major nutrients. Stemflow may be thought of as a concentrated version of throughfall (McColl, 1970). C. Seasonality Nutrient deposition in throughfall is commonly more seasonal than that in incident precipitation (Miller, 1963), though less seasonal than in stemflow (Carlisle c'f d. , 1966b; Eaton e f al., 1973; Ulrich et al., 1978). In temperate deciduous forests the nutrient fall in throughfall is more even throughout the year than that in litterfall (Miller. 1963; Carlisle et a/., 1966a,b) which can be strongly seasonal (Bray and Gorham, 1964). Throughfall and stemflow depositions exhibit greater seasonality than does incident rainwater. particularly in temperate hardwoods (Miller, 1963; Carlisle et d., 1966a, 1967; Ulrich et a/., 1978). In eastern Tennessee, Henderson ct N/. (1977) found that nutrient depositions in throughfall under hardwood stands exceeded those under coniferous ones during the growing season. The relationship was reversed during the dormant season. The annual fluctuation in throughfall deposition is primarily due to seasonality in metabolic processes (growth and translocation), though wet and dry pre- cipitation inputs can also vary seasonally (Ulrich et a/., 1978; Galloway and Parker, 1979). Concentrations of most nutrients in throughfall, and pre- sumably, their leaching rates, are lowest at the beginning of the growing season (or of the lifespan of the leaf), becoming progressively higher until foliar abscission, when they are generally highest (Carlisle et a/., 1966b; Tukey, 1970a: Ulrich et a/., 1978). Nutrients, such as nitrogen, which are partially withdrawn from foliage prior to abscission, do not show increasing leaching rates during litterfall (e.g. Miller, 1963; Carlisle et al., 1966b). THROUGHFALL, STEMFLOW IN FOREST NUTRITION 73 The supply of water borne nutrients, however, may continue the year round, requiring only sufficient precipitation. Temporal variability in the supply of nutrients to the forest floor has not been detailed for evergreen or tropical forests. On a short-term basis the relative importance ofthese pathways depends on the seasonality of the fluxes of organic matter and water. the transport media for nutrients. The bottom of Fig. 6 gives the monthly percentage of throughfall in total nutrient fall (throughfall and litterfall) for all elements measured (lower line) and all elements excluding carbon (upper line) in a mixed oak forest in Lancashire. England ( q f i ~ ~ . Carlisle P / ul., 1966a.b). The upper portion traces the monthly fluxes of nutrient carriers. litterfall and precipitation. The recycling pathways of the seven elements measured spans the throughfall litterfall continuum. Sodium and potassium transfer pre- dominately via throughfall all year long, calcium and magnesium for most of the year. Even for phosphorus, nitrogen and carbon, there are periods in which throughfall dominates. The autumn drop in throughfall importance in all elements is, of course, due to the litterfall pulse. Though litterfall may supply more material annually, litterfall nutrients are only slowly released from organic matter (e.g. Gosz et a/., 1976) while throughfall nutrients are nearly all dissolved. Nutrient concentrations in throughfall are higher than those in water extracts of undecomposed litter (Mahendrappa and Odgen, 1973) and throughfall solutes are available for plant uptake (foliar leachates are readily reabsorbed by plant roots, Tukey et id., 1958). D. Recycling Rates Consideration of the throughfall pathway affects the estimates of nutrient cycling rates. Reiners (1972) and Gosz et ul. (1976) have shown that turnover time estimates for nutrients in the forest floor are strongly reduced when nutrient fluxes in throughfall and stemflow are considered along with those in litterfall. Reduced residence times were observed for all elements but were most pronounced for Na, S, and K. For potassium in the Hubbard Brook forest, revised residence times suggest that the forest floor pool ofthis nutrient cycles a t a rate equivalent to 1.43 replacements per year (Gosz et a/., 1976). Residence times of canopy bound materials are also reduced when net forest water is considered along with litterfall. The ratio of canopy content to annual net throughfall flux (the time required for canopy replacement via net throughfall) has been cited as an index of elemental mobility (Attiwill, 1966). A ranking of these indices by element would indicate relative leachability (Henderson et a/., 1977; EdtOn et ul., 1973; Szabo, 1977), if net throughfall were due to leachates only. Ratios of nutrient fluxes recycled to fluxes input have been used as crude indices of cycling efficiency (Jordan and Kline. 1972). Such an index, whose 74 G . G. PARKER r c c P E E C 0 t 0 a F a c 40 3 0 2 0 10 0 -Corbon P 2000 I500 1000 500 0 f P - 0 I 0 Y I 1 1 1 1 1 1 1 1 1 1 ~ J J A S O N D J F M A M 80 1 0 70 2 5 0 o 4 0 c % 3 0 10 2ol 0 0 3 I , , I J J A S O N D J F M A M Fig. 6. In the lower diagram is the ;< throughfall in the monthly total of nutrients falling to the soil in throughfall and litterfall. The line marked -Carbon is calculated from the sum of nutrient deposition in N , P. K , Ca, Mg, and Na; the line marked +Carbon also includes carbon. In the upper diagram are the monthly carrier fluxes of litterfall (kg ha -) and precipitation depth (cm). Adapted from Carlisle et a/ . (1966a,b). value appears to be characteristic of the element (Ulrich, 1971), has been employed as a measure of nutrient cycle stability (Jordan et al., 1972). Elements with high throughfall concentration often have high cycling indices, even when the index is the ratio of material recycled to material flowing through the system (Finn, 1976, 1980). E. Variability Throughfall and stemflow nutrient depositions characteristically exhibit much variability in time and space (e.g. Kimmins, 1973; Bringmark, 1980), which is composed, in turn, of hydrological and chemical variability. For THROUGHFALL, STEMFLOW IN FOREST NUTRITION 75 measurements in a given stand, this variability can be further partitioned into variances due to sampler location and sampling period (Wilm, 1946). The resulting variability is itself characteristic of the stand; for example, mixed forests have greater variability in throughfall depositions and concentrations than do single species stands (Kimmins, 1973; Torrenueva, 1975; Parker, 1982). In general, variability in concentration is greater than for precipitation depth (Parker, 1982). Incident depositions are less variable (Kimmins, 1973), but stemflow concentrations and depositions are more variable still (Tor- reneuva, 1975). Concentrations vary with place and time (e.g. Kimmins, 1973; Bringmark, 1980), even within a forest during a single storm, where relative variability (standard deviation as a percentage of mean) in nutrient con- centrations can range to over 60% depending on the element and the forest type (Parker, 1982). Even annual variability can be substantial, as can be seen from Fig. 7 (p. 76), which gives the annual sulphateesulphur depositions in throughfall, stemflow and incident precipitation measured by Ulrich et a/ . (1978) for eight years in Solling, Germany, the longest record of throughfall measurements I have found. Stemflow deposition of sulphur appears to vary more than does throughfall, which in turn, varies more than incident precipitation. Net throughfall is most variable. Clearly, a single year's measurement of throughfall or stemflow quality may not be representative of long-term trends. Additional study may show that this variability is resolvable into a distinct spatial pattern in amounts of nutrients contributed by throughfall, as Ford and Deans (1978) and Kellman (1979) have shown for patterns of throughfall depth. The stemflow zone is clearly such a pattern (see Section IV. B. 2). It is likely that a spatial pattern of throughfall composition is associated with the locations of trees, given the known differences in leaching and decomposition rates of leaves between species (Gosz et a/., 1975) and the heterogeneity in forest composition. It has been observed that certain collector positions have consistently higher throughfall concentrations (Lindberg et al., 1979; Parker, 1982; Haines, personal communication) but no study of heterogeneity in throughfall quality over small spatial scales has come to my attention. V. FACTORS AFFECTING THE QUALITY OF THROUGHFALL In the following sections the effects of a number of large-scale factors on throughfall behaviour are assessed with data gathered from a multitude of sources. The data vary in quality due to diversity in forest systems and the sampling and analytical methods of the researchers. Some studies involved numerous collectors sampled by event over an entire year, others a composite 76 00 70 c 7 6 0 m I v) y" 50 0 C c 4 0 U 3 L 3 0 ? - m L $ 2 0 v) 10 0 G. G. PARKER Sternflow Throuqhfoll Incident precipitation 68/69 69/70 70/71 71/72 72/73 73/74 74/75 75/76 Fig. 7. Sulphate-sulphur in throughtiill. stemflow and incident precipitation under Fagus .s.v/vaticu in Solling, Germany, 1968-1976. Atter Ulrich et al. (1978). of several combined samples. Estimates were made from many years' data in several cases. Often subannual values were extrapolated and occasionally the whole year is represented by only a few samples. Most of the deposition values reported for throughfall and incident precipitation are listed in the Appendix table. In order to compare effects of forest canopies on changing precipitation concentrations, studies reporting only ( I ) concentration or (2) stemflow deposition and most of those which neglected incident deposition have been omitted. Depositions are given as the elemental weight (e.g. NO3 in kg N ha- ' yr- '). A. Amount of Precipitation The amount and timing of precipitation is the dominant control on the magnitude of material removed from both the atmosphere and the canopy (for total throughfall: Madgwick and Ovington, 1959; Attiwill, 1966; Reiners, 1972; Bernhard-Renversat, 1975; Szabo, 1977; Verry and Timmons, 1977; Lindberg et al., 1979; Parker rt ul., 1980; for incident precipitation: Larson and Hettick, 1956; Attiwill, 1966; Wolaver and Lieth, 1972; Parker et al., THROUGHFALL, STEMFLOW IN FOREST NUTRITION 77 1980). Both precipitation and throughfall concentrations show a strong negative relation with precipitation depth. This relation has been modelled with linear (Larson and Hettick, 1956), exponential (Madgwick and Ovin- gton, 1959; Lindberg et al., 1979; Parker, 1982). power (Attiwill, 1966) or log-log functions (Szabo, 1977). Depositions on the other hand, are positively related to precipitation amounts, either linearly or exponentially (Parker, 1982). Figure 8 (below) gives such a relation for net throughfall magnesium and incident precipitation amount for two Virginia piedmont forests. Magnesium release from both of these canopies is proportional to pre- cipitation applied, which suggests a leaching process. 8 0 6.0 B. Latitude Latitude influences nutrient depositions in throughfall through the length of the growing season, the rates of precipitation and litterfall (Basilevich and Rodin, 1966; Rodin and Basilevich 1967), particularly where latitude is correlated with seasonal variation. Not surprisingly, nutrient fluxes in throughfall and litterfall are much higher in tropical and subtropical forests than in temperate or boreal ones (Rodin and Basilevich, 1967; Herrera et a/., 1978; Cole and Rapp, 198 I). However, mean throughfall concentrations do - ~- r = 64 Magnesium 8 0 6 0 4 0 2 0 0 A 0 20 40 60 Incident precipitation depth mm storm Fig. 8. Net deposition of magnesium in single storms (mg Mg m- storm- I ) in two forests in the Piedmont of Virginia. Parker (1982). 78 G . G . PARKER not appear to differ with latitude, possibly because of great within-latitude variation. C. Proximity to Sources Where stands are within range of active aerosol, dust or gas sources (natural and anthropogenic), both incident precipitation and throughfall deposition can be affected. Studies reporting the effect of source proximity here involved either strong or extremely characteristic sources. Marine salts (Miller, 1963; Attiwill, 1966; Art et al., 1974; Miller et al., 1976; Potts, 1978; Miller, 1979; Miller and Miller, 1980) and continental dust (Parent, 1972; Hart and Parent, 1974) have been considered more often than sources of anthropogenic origin (Haughbotn, 1973; Horntvedt and Joranger, 1974; Abrahamson et al., 1976; Baker et al., 1977; Lindberg et d., 1979; Parker et a/., 1980). Even so, effects of sources are often difficult to demonstrate (Parker et al., 1980), since dry deposition load is not the only factor varying with distance from a source. Also, mineral elements leached from canopy tissues may have originated in previous dry deposition, especially for elements that are efficiently recycled. I . Muritime Influences Many studied stands are close enough to oceans to be affected by sea salt depositions (Ingham, 1950a; Madgwick and Ovington, 1959; Miller, 1963; Attiwill, 1966; Carlisle et al., 1966a; Rapp, 1969; Nihlgard, 1970; Art et a/., 1974; Clements and Colon, 1975; Golley et al., 1969). Junge and Werby (1958) and Wolaver and Lieth (1972) have described a clear oceanic effect on incident rainwater concentrations for Na, C1, Mg and SO,; Miller (1963), Attiwill ( 1966), Rapp ( 1969), Art et al. ( 1974). and Graustein (1980) for various ions in net throughfall. Rapp (1969) found a pronounced maritime gradient of net throughfall sodium concentrations in four forests within 37km of the Mediterranean. An effect of proximity to the ocean may even be observed across a single stand. Potts (1978) found a strong gradient of throughfall sodium, and to a lesser extent magnesium, with nearness to the windward edge of a maritime stand of Sitka spruce (Picea sitchensis) in Wales. The additional salts, which levelled out in deposition amount after three canopy heights from the stand edge, are due to dry deposition (see Fig. 9). As with incident precipitation (Wolaver and Lieth, 1972), the marine influence upon through- fall is limited to the near vicinity ( < 100 km) of the ocean. 2. An t h ropogen ic Sources Several compounds of anthropogenic origin are often more concentrated in throughfall near sources. For example, sulphate-sulphur can be more enriched near sulphur sources than farther away (Haugbotn, 1973; Baker et a/., 1977; Turner and Lambert, 1980); the highest throughfall sulphate THROUGHFALL, STEMFLOW IN FOREST NUTRITION Distance in multiples of canopy height L r m $ 5 0 - + 200 - - % - 0 I Y c 0 Lo cm 150- .- c .- a c 100- 0 u f 0 0 79 Distance from forest edge (m) Fig. 9. Gross throughfall deposition of sea salt cations with distance from the windward edge of a maritime sitka spruce (Pic.cu.sirc,hmsis) stand in Wales. After Potts ( 1978). deposition recorded, of 1 1 1.2 kg S ha -' yr - l , comes from a stand next to a factory (Haugbotn, 1973). Unfortunately some anthropogenic sources are regional in extent (Likens and Bormann, 1974; Galloway and Parker, 1979) and must be identified indirectly, via trajectory analyses or other means (e.g. Miller et al., 1979). Throughfall sulphur in industrialized regions (Haugbotn, 1973; Knabe and Gunther, 1976; Baker et al., 1977; Lindberg rt al., 1979) is generally higher than in remote areas (Hesse, 1957; Graustein, 1980; Jordan et al., 1980). Heavy metal dry deposition is clearly highest around sources (e.g. Hughes et al., 1980), particularly to leaves (e.g. Little and Wiffen, 1977). The proximity effects on throughfall concentration of heavy metals have not been in- vestigated and may prove difficult to demonstrate because of the insoluble nature of many heavy metal compounds (e.g. Lindberg rt af., 1979). 80 G. G. PARKER D. Forest Type Throughfall quality varies according to the nature of the intercepting canopy, whether individual tree or entire forest (Tamm, 1951; Tamm in Stenlid, 1958; Madgwick and Ovington, 1959; Voigt, 1960b; Denaeyer-DeSmet, 1966; Parent, 1972; Eaton rt al., 1973; Hart and Parent, 1974; Patterson, 1975; Tsutsumi, 1977). Stemflow concentrations also show species effects (Kaul and Billings, 1965; Mahendrappa and Ogden, 1973). I . Hardwoods and Conifers Investigations of nutrient cycling differences between temperate forests most often compare hardwood and conifer stands (Mina, 1965; Tarrant eta/. , 1968; Rapp, 1969; Nihlgard, 1970; Best and Monk, 1975; Reiners, 1972; Wellsera/., 1975; Heinrichs and Mayer, 1977; Henderson ct al., 1977; Verry and Timmons, 1977). In such comparisons the hardwood stands have much grcatcr annual nutrient depositions in throughfall than pines, despite a shorter growing season (Iwatsubo and Tsutsumi, 1967; Tarrant rt a/., 1968; Rapp, 1969; Best and Monk, 1975; Wells rt a/., 1975; Comerford and White, 1977; Verry and Timmons, 1977). For example, Tarrant et al., (1968) have shown that the throughfall fluxes in total dissolved solids and various nitrogen forms decrease in the order aspen > mixed >conifer for stands in coastal Oregon. Red spruce (Picra ruhrns) stands, however, show greater nutrient deposition in throughfall compared to adjacent hardwoods (Mina, 1965; Nihlgard, 1970; Heinrichs and Mayer, 1977). On the other hand, throughfall depositions in an aspen forest were found to be greater than in a nearby black spruce (Picea marina) bog (Verry and Timmons, 1977). Throughfall under conifers is commonly more acidic than for hardwoods in the same area (Nihlgard, 1970; Akhtyrtsev and Sviridova, 1975; Patterson, 1975; Torreneuva, 1975; Wells et al., 1975; Clesceri and Vasudevan, 1980; Cronan et al., 1980; Miller and Miller, 1980: Parker, 1982). This parallels the differences in soil pH under these canopy types (e.g. Spurr and Barnes, 1973). Both throughfall and soil solution acidity are likely to be affected by organic acids leached from living and dead tissue. The free acidity of conifer throughfall can be greater than in incident precipitation (Tarrant rt al., 1968; Stark, 1973; Hart and Parent, 1974; Feller, 1977), even where the incident precipitation is anthropogenically acidified (Nihlgard, 1970). Coniferous forests appear to reflect better the proximity of local dry deposition sources: Na is better represented in seaside conifer throughfall than in adjacent hardwood stands (Madgwick and Ovington, 1959; Rapp, 1969; Nihlgard, 1970). Coniferous throughfall sulphur is similarly enhanced in industrialized areas compared to adjacent hardwoods (Baker rt al., 1977; Heinrichs and Mayer, 1977). THROUGHFALL, STEMFLOW IN FOREST NUTRITION 81 2. Other Features Edaphic and physiological factors may play a role: only small differences between stand types were found in some studies (Tarrant eta/. , 1968; Reiners, 1972; Henderson et al., 1977; Rolfe et a/., 1978; Weaver and Brown, 1979). Net throughfall depositions were not significantly different between conifers and hardwoods in 14 seaside stands in England (Madgwick and Ovington, 1959). Throughfall and stemflow fluxes under stands of Fagus syylvatica in Denmark were much higher for forests on mull soils than for those on mor, especially for Ca, Mg, Mn and P (Astrup and Biilow-Olsen, 1979). Killing- beck and Wali (1978) have described the effect of aspect on throughfall depositions for a gallery forest in North Dakota. Villecourt and Roose (1978) report that net throughfall calcium and phosphorus were nearly identical for savannah and nearby tropical gallery forests in the Ivory Coast. The presence and composition of a forest understory can affect total throughfall de- positions, even when overstory species are the same (Comerford and White, 1977; Astrup and Biilow-Olsen, 1979). Abee and Lavender (1972) found strong differences in net throughfall between six old growth Douglas fir stands along an altitudinal gradient in Washington. E. Stand Age The question whether forest nutrient cycles become more (Odum, 1969, 1971) or less efficient (Vitousek and Reiners, 1975) during stand aging has never been systematically investigated. To be sure, some aspects of nutrient cycling during stand development have been documented. Changes over time in nutrient accumulation and fluxes in wood, foliage and forest floor have been reported (Ovington 1959, 1962; Switzer and Nelson 1972; Kazimirov and Morozova 1973; Cole and Rapp, 1981). Throughfall, on the other hand, has received little attention in the context of stand age. The only report of changes of throughfall composition with stand age comes from Lemee (1974) who studied adjacent stands of Fagus sylvatica ranging from 4 to over 200 years of age in the vicinity of Paris, France. Annual net throughfall depositions of K, Ca, Mg, NH,-N, N03-N and PT rose in a remarkably linear fashion with stand age (Fig. 10, overleaf). Also, the relative magnitudes of net throughfall and incident precipitation nutrient deposition change with age. For the older stands, bulk precipitation input of nutrients is less than in net throughfall. Total mineral nitrogen (N03-N + NH,-N) in net throughfall nearly equals that in precipitation for the 200+ year-old stand and K in net throughfall exceeds that of incident precipitation for all stands. Net throughfall depositions of P, Ca and Mg exceed those in precipitation at various points within the age sequence. 82 30 - s - 0 I 0 Y C 20 U 8 - - 0 0 3 c e 5 c 10 i 0 G . G . PARKER t 0 50 100 150 > 200 Age of stond Yrs Fig. 10. Net throughfall deposition of (NO,+NH,) N, P, K, Ca and Mg in an age sequence of Fugus sylvuricu stands. The horizontal bar gives the range of tree ages within one stand. After Lemee (1974). F. Site Fertility I . Native Trophic Status The importance of the throughfall pathway necessarily depends on the trophic context of the forest. Oligotrophic systems, such as bogs (Gorham, 1953), swamps (Holmes and Brinson, 1977; Schlesinger, 1978; Brinson et af., 1980), and coastal areas (Etherington, 1967 and Westman, 1978), some tropical forests (Jordan et al., 1980), and plants with low nutrient stocks will require efficient cycling of materials. Rootless epiphytic plants without access to litterfall return, such as lichens (Lang et al., 1976; Pike, 1978), mosses (Tamm, 1950; Rasmussen and Johnsen, 1976), algae (Witkamp, 1970) and bromeliads THROUGHFALL, STEMFLOW IN FOREST NUTRITION 83 (Tukey, 1970b) are particularly dependent on throughfall and stemflow nutrients. Schlesinger and Marks (1977) have found that the epiphyte Spanish moss (Tillandsia usneoides) preferentially colonizes tree species which have highest throughfall concentrations, among other factors. Gorham (1953) and Schlesinger (1978) term such systems ombrotrophic to emphasize their dependence on precipitation. 2. Fertilization Increases in the nutrient status of the stand either enhances plant vigor and uptake rates or increases the pool of material available for translocation. Where available soil nutrients have been artificially increased, throughfall element deposition increases with increasing amounts of fertilizer (Mahen- drappa and Ogden, 1973; Wells et al., 1975; Miller et al., 1976; Yawney e ta / . , 1978; Khanna and Ulrich, 1981). Soil amendments appear to influence throughfall depositions by raising the foliar pools of leachable nutrients since foliar concentrations also rise with fertilization. Mahendrappa and Ogden (1973), Wells et al. (1973, Miller et al. (1976) and Khanna and Ulrich (1981) found that increases of throughfall and litterfall depositions corresponded to changes in foliar concentrations of these elements. Several authors (e.g. Tukey, 1970a; Eaton et al., 1973; Henderson et al., 1977) have suggested that foliage concentration of a nutrient is a useful predictor of throughfall enrichment, particularly for major nutrients N, P and K. Miller et a/. (1976) present a cubic relation for nitrogen net throughfall deposition and the concentration of the previous years foliage. Throughfall nitrogen under peatland Scots pine showed a strong amendment effect three years after heavy fertilization (Paivanen, 1974). Foliar concentrations of nonfertilized elements may also increase and, in turn, increase depositions in litterfall and throughfall (Paivanen, 1974; Wells et a/., 1975). Stemflow depositions may (Yawney eta / . , 1978) or may not (Mahendrappa and Odgen, 1973) reflect fertilizer additions. Nitrogen fixation constitutes a form of fertilization. Tarrant rt al. (1968) found greater net throughfall nitrogen under nodulated red alders than under nearby coniferous and mixed alder-conifer stands of similar age. Through fall nitrogen under single trees (Bollen et al., 1968) and stands of red alder (Turner et al., 1976) was quite enriched, having a deposition ratio of 3.18 for nitrogen. While interactions between site fertility and amount of precipitation have not been investigated, it appears that nutrient depositions in throughfall and stemflow are closely related to the trophic status of the stand. In two extremely oligotrophic rainforests in the northern Amazon, Jordan et al. (1980) observed negative net throughfall depositions for N, K and S and in one site, Ca and P as well. Also, Mina (1965) reports that stemflow is most concentrated in rich sites. Throughfall enhancement might prove to be an index of stand trophic status. 84 G. G. PARKER G. Insect Consumption Insect consumption in the canopy may cause elevated concentrations of throughfall (Carlisle et al., 1966b, 1967; Krause, 1977; Crossley and Seastedt, 198 1 ; Haines, personal communication). The principle mechanism for this effect is probably increased leaching from damaged leaves (Tukey and Morgan, 1963; Kimmins, 1972), though increased fall of litter (Kimmins, 1972; Tiedemann et al., 1980), frass and frass leachates (Kimmins, 1972), insect exuviae and exudates such as aphid honeydew (Carlisle, 1965; Carlisle et a/., 1966a,b), may also contribute. The presence of insects is central to this phenomenon. Nutrient con- centrations in throughfall and stemflow under canopies artificially defoliated with Paraquat (600{ of needles lost) showed little difference with those in an untreated plot (Tiedemann et al., 1980). Crossley and Seastedt (1981) found that throughfall concentrations of NH,, PO, and SO, under individual black locust and red maple trees was significantly related to the percentage of canopy consumed by insects. Defoliation of red pine by European saw flies added to the canopy by Kimmins (1972), increased the rate of leaching loss of I3,Cs. Much remains to be learned about the effects of insect consumption and defoliation on nutrient recycling, especially during large-scale infestations. H. Other Factors Stand altitude influences nutrient deposition in throughfall in a complex manner since altitude is correlated with higher amounts of precipitation, higher wind speeds and lower temperatures and soil and vegetation types, among other factors. Johnson (1975), Turner and Singer (1976), Ugolini e t a / . (1977); Cronan (1978) and Falconer and Falconer (1979), have investigated subalpine systems, where high winds speeds and the compact nature of krummholz vegetation are associated with high rates of water and aerosol impaction (Schlesinger and Reiners, 1974; Cronan, 1978). In alpine systems much incident precipitation comes in indirect forms such as fog, cloud droplets and rime ice, which can be considerable (e.g. Vogelmann e ta / . , 1968), though difficult to quantify (Kittredge, 1948; Cronan, 1978; Cronan and Schofield, 1979; Falconer and Falconer, 1979). Such precipitation can be more concentrated than direct precipitation, especially in acids (Falconer and Falconer, 1979). Leaching losses can be much greater in such systems since fog, mist and cloud water can be intercepted by lower as well as upper foliar surfaces. Rainwater is the most common precipitation form studied, but snow (Verry and Timmons. 1977; Fahey, 1979; Tiedemann e f al., 1980), rime ice and hoarfrost (Schlesinger and Reiners, 1974; Cronan, 1978; Cronan and THROUGHFALL, STEMFLOW IN FOREST NUTRITION 85 Schofield, 1979), fog and cloud water (Cronan, 1978; Falconer and Falconer, 1979) and dew (Tukey, 1970a) may also remove material from canopies. Fahey (1979) found concentrations of K, Mg and Ca higher in the fresh snowpack under three subalpine canopies than for snow in the open. VI. PROCESSES OF THROUGHFALL ENHANCEMENT The quality of throughfall deposition is determined by several distinct processes capable of changing concentrations and amounts of precipitation. There are at least five such processes: (1) Incident wet deposition to the canopy during the event. (2) Evaporation of intercepted water. (3) Washing, by precipitation, of deposits accumulated upon the canopy (4) Leaching of material from internal plant tissues. (5) Uptake, sorption or permanent attachment of solutes, gases or particles by foliage and epiphyllic biota. All of these processes (diagrammed below in Fig. 11) contribute to the alteration of precipitation quality; the importance of a given process varies according to the factors previously described in this review. Each of the component processes and their potential contributions to throughfall quality is described in the following sections. between events. Wet precipitation Dry precipitotion / Sedimentation Absorption lmpoction Consumer + activities Canopy minerol Folior uptoke 1T Leoching Woshoff Y Throughfall ond stemflow Fig. 11. Processes affecting the composition of throughfall. 86 G. G. PARKER A. Incident Precipitation The quality of incident precipitation may affect the quality of throughfall through its effects on washing, leaching and sorption, but the importance of particular constituents of incident precipitation (e.g. acids, surfactants) in affecting throughfall quality remains uninvestigated in field situations. Solute concentrations in precipitation may be important: for example, leaching may occur at low concentrations but foliar absorption may take place at high concentrations (Yawney and Leaf, 1971). B. Evaporative Concentration The evaporation of intercepted water and the consequent increase in concentration of the remaining precipitation only explains a fraction of the total enhancement in rainwater concentrations. If the solute mass in incident precipitation is completely conserved in throughfall via evaporative con- centration, then the solute concentration increase is directly proportional to the degree of evaporation, as is expressed in the following equation, 5 D, C i = D where the terms have been previously defined. In other words, the con- centration ratio expected by evaporation alone equals the reciprocal of the deposition ratio of precipitation amounts. Such a relation is given in Fig. 12 for mean concentration ratios of K in single events under White pines in Virginia (authors data). All the points in this figure are substantially above the line of evaporative concentration indicating that other processes are important in enhancing throughfall concentrations. Such relations are evident for other ions as well. Evaporative concentration is but a minor process for annual nutrient depositions in throughfall (Banaszak, 1975). Evaporation of precipitation in smaller-sized storms, which exhibit highest solute concentrations (Madgwick and Ovington, 1959; Attiwill, 1966; Wol- aver and Lieth, 1972; Parker, 1982), may produce extremely high solute concentrations in water in the canopy, possibly damaging to the leaf (Tukey, 1970a). In even smaller events, where the rainfall is insufficient to penetrate the canopy, the dissolved materials are either precipitated onto the foliage, or taken up into leaf tissues through foliar absorption. While evaporation is the dominant process of throughfall enhancement in these cases, the low deposition contributes little to the annual nutrient budget. Unfortunately, since the majority of studies seek to estimate annual fluxes, small storm processes are not commonly documented. Throughfall is now more com- monly sampled by event (e.g. Comerford and White, 1977; Krause, 1977; Manokaran, 1978; Lindberg et a/., 1979; Tiedemann et al., 1980) rather than THROUGHFALL, STEMFLOW IN FOREST NUTRITION 87 weekly or monthly, so that small-storm effects will be easier to discern. The maximum evaporative contribution is probably less than 5% of the total annual enhancement, depending on the compound. P F .- C 0 e c - C 0 0 0 5 u) ln .- a 0 0 I 2 3 4 5 6 7 8 9 1 0 Deposition ratio for precipitotion amounts Fig. 12. Throughfall enrichment for potassium (0 ) in single precipitation events under white pine (Pinus strohu.~). The solid line represents concentration increases porpor- tional to the evaporation of intercepted water. The concentration ratio is the concentration of throughfall -concentration of incident precipitation and the de- position ratio for precipitation amounts is throughfall as a fraction of incident precipitation. Numbers refer to points off scale. C. Inter-event Deposition Dry deposition, which combines all the input processes between storms (Galloway and Parker, 1980), is much greater to canopies than to either flat ground or even the collectors commonly used to sample bulk (White and Turner, 1970; Eaton et a/., 1973; Art et ul., 1974; Miller et al., 1976; Bache, 1977) or dry precipitation (Swank and Henderson, 1976; Tiedemann, 1980). 88 G . G . PARKER The extent of the difference is a matter of substantial and continuing debate (e.g. Belot and Gauthier, 1975; Horntvedt, 1975; Droppo, 1976; Chamberlain, 1980; Graustein, 1980; Sehmel, 1980). The difference is due to the geometry of the canopy, its surface wetness (Chamberlain, 1975a) and meteorological factors (Miller et a/., 1978). Dry deposition is made up of at least three distinct physical processes which depend on the phase and size of the airborne matter. Gases are sorbed or actively removed from the atmosphere. Particles fall onto canopy surfaces (when large) or impact and remain on them (when small). Dry deposition is mostly due to particle impaction (White and Turner, 1970; Wedding ct al., 1976) and gaseous absorption (Hill, 1971; Chamberlain, 1975b), though sedimentation is observed (Tamm and Troedsson, 1955; Russel, 1965). For a given forest the dominant process ultimately depends on the compound in question, its phase, chemical form and size distribution. I . Sedimentation The disappearance of visible dust deposits on roadside foliage following rains is a common observation. Tamm and Troedsson (1955) showed that these sedimented materials can contain considerable amounts of plant nutrients. Hamman (1956) indicates that surface dusts and spray residues can be considered contaminants in the analysis of foliar tissues. Sedimentation (dryfall, fallout) is the most important dry deposition process for large-sized airborne particles (> 10pm in diameter) whose settling rates depend on ambient concentration and particle size distribution, not on the complexity of the receiving surface (Witherspoon, 1972 and Chamberlain, 1975a). Since the sediment sees only the projected surface area dusts are generally found on the uppermost canopy strata; the smaller particles on lower layers are deposited by other means. Dry deposition by sedimentation is most conspicuous in the vicinity of large particle sources such as alkali flats, recently plowed or fertilized fields, gravel roads, or industrial stacks. Also, sea salt elements in coastal forest throughfall probably originate via sedimentation (Ingham, 1950b; Rapp, 1969; Art et al., 1974; Clements and Colon, 1975; Potts, 1978; Westmann, 1978). Large, continuous forests probably do not derive much nutrient input from sedimentation. 2. Gaseous Absorption Absorption by and uptake into plant tissues is the major route ofentry into the forest system for numerous gaseous compounds. The process is influenced by wind speed and turbulence, by the concentration and solubility of the gas and by the wetness and stomata1 behaviour of the foliage (Hill, 1971; Bennet and Hill, 1975; Mudd, 1975; Bache, 1977; Chamberlain, 1980). Much of the absorbed material may be fixed, metabolized or retranslocated, depending on the compound. How much of these materials are routinely retained on plant surfaces, or how much may be remobilized by precipitation, is unknown. For THROUGHFALL, STEMFLOW IN FOREST NUTRITION example, Horntvedt et al., (1980) found that much more sulphate sulphur could be washed from leaves immediately after fumigation with SO2 than after longer intervals. Therefore, while gaseous absorption is a major input pathway for carbon, nitrogen and sulphur and other elements, its importance in throughfall is unclear. 3. Gas and Aerosol Impaction Gases and small sized aerosols ( < 2 pm in diameter) impact upon and are retained by plant surfaces by turbulence above and within the canopy (Droppo, 1976). The transfer is described mathematically as a process of turbulent diffusion (Slinn, 1977). A common approach for the estimation of the impaction flux is the deposition velocity, vg (or vd, cm sec -I), which may be interpreted as the speed at which airborne material at some reference height (commonly 1 m above the surface) is transferred to the receiving surface. Nutrients fluxes via impaction are estimated as the product of ambient concentrations of the gas or aerosol and deposition velocities, which vary from 10.0 to .05 cm sec-' over forests (White and Turner, 1970; Droppo, 1976; Droppo, 1980; Sehmel, 1980). Particulate deposition velocities rise with particle density, solubility and diameter (when greater than 0-1 pm), with surface roughness and wetness (and smaller obstacle sizes), wind speed and possibly, with atmospheric stability (e.g. Chamberlain, 1975a; Slinn, 1977; Sehmel, 1980). Gas deposition velocities are additionally affected by canopy resistances (Chamberlain, 1980). Many methods exist for estimating vg (White and Turner, 1970; Chamberlain, 1976; Droppo, 1976; Bache, 1977; Chamberlain, 1980; Lind- berg and Harriss, 198 1) but unfortunately most are complicated and produce variable results (see the review of vg estimates by Droppo, 1980). An alternative, though less frequently employed, measure of canopy filtering potential is the impaction efficiency, the amount of material caught on the obstacle divided by the amount that would have passed by were the obstacle not present. Its value is low, usually less than 0.01, closer to 0.001 (Gregory, 1961; White and Turner, 1970; Belot and Gauthier, 1975). However, since a hectare of forest annually interacts with many cubic kilometers of air even low impaction efficiencies can imply large fluxes. As with the deposition velocity, the impaction efficiency is a difficult and variable quantity to measure. The contribution of gaseous absorption to dry deposition is, as previously discussed, unknown. The impaction component will probably dominate sedimentation for the majority of nutrient elements since ( 1 ) airborne gas and small particle concentrations are high; ( 2 ) the geometry of foliar elements (twigs and leaves and their surface micro-structure) is well suited to impaction (Belot and Gauthier, 1975; Holloway, 1971), and total canopy areas and leaf area densities are large, (Whittaker and Woodwell, 1967), and (3) the 89 90 G. G. PARKER roughness of the whole canopy produces a local turbulence, enhancing the transfer of material-laden air to the canopy. D. Leaching The removal of internal plant parts by external solutions is a biologically passive process which can transfer a variety of substances from plants and temporarily reduce the tissue concentrations of those substances (Tukey, 1970a). The materials removed (leachates) evidently derive from ion replace- ment reactions involving dilute water (the leachant) and a pool of exchange- able ions in the intercellular free space or outer space of leaves and other tissues (Mecklenburg et al., 1966; Epstein, 1972). This pool is connected to the transpiration and translocational streams since (1) amounts of solutes removed can total more than the differences in foliar levels before and after leaching (Tukey, 1970a) and (2) rates of root uptake and translocation may increase during leaching (Tukey et a/ . , 1958). The species, vigor, tissue (its position and age), as well as the amount and composition of the leachant also affect leaching rates. Materials are leached most easily during the initial stages of wetting and additional applications of leachant yield proportionately less solute (Clements et a/., 1972). Lausberg (1935) suggests that losses from the leachable pool are replenished within 3-4 days after a large storm. Organic substances, especially carbohydrates, are leached in largest quantities, but HCO,, K, Ca, Mg and Mn are easily removed as well (Tukey 1970a). Radioisotope studies by Tukey et al. (1958) indicate that more than 25% of total foliar Na and Mn can be leached; between I and 10% of foliar Ca, Mg, K and Sr and of Fe, Zn, P, and C1, less than 1%. Ammonium, nitrate and nitrite are not easily leached. The leaching potential of silicate, sulphate, and the organic forms of N, P and S has not been investigated in the laboratory. Interested readers are referred to Tukeys ( 1970a) excellent review of leaching physiology. Leaching is probably the major source of net throughfall for many elements such as potassium and carbon compounds. In spite of its known rapidity and magnitude in laboratory trials however, the importance of leaching is unresolved for other elements in field situations. The quality of incident precipitation may affect leaching. Several mech- anisms might be operating, for example: (1) low leachant concentrations could increase the concentration gradient, and (2) certain substances might influence the ionic exchanges, for example, K and Na salts promote foliar losses whereas Ca salts inhibit them (Tukey, 1970a). High concentrations of hydrogen ions in artificial rainwater or mist are known to increase cation leaching from leaves (Fairfax and Lepp, 1975; Wood and Borman, 1975; Abrahamson et al., 1976; Horntvedt et al., 1980; authors data), probably THROUGHFALL, STEMFLOW IN FOREST NUTRITION 91 through a process of ion exchange. I t has been suggested that recent, anthropogenically-induced increases in precipitation acidity (Likens and Bormann, 1974) might accelerate release of minerals from the canopy, among other effects (Tamm and Cowling, 1977; Jacobsen, 1980; Tukey, 1980). However, no field study has yet demonstrated clearly that more acid storms produce more concentrated throughfall. In fact, Eaton e ta / . (1973) calculate that only 27% of cations released from a northern hardwood canopy are accounted for by protons retained. Similar results were observed for throughfall under white pine and mixed hardwood canopies in rural Virginia (Parker, 1982). E. Foliar Uptake Plant leaves can remove materials from solution by a process analogous to the reversal of leaching (Boynton, 1954; Wittwer and Teubner, 1959; Wittwer and Bukovac, 1969). For some compounds this is an efficient pathway of entry into plants; fertilizers are often applied to the leaves of crop species. In the nutrient cycling context foliar uptake includes not only active and passive absorption through cuticles and stomates, but also accumulation of solutes by phyllosphere organisms. Epiphytic uptake may not immediately affect plant nutrition (though the materials become available after litterfall), but it does contribute to changing precipitation concentrations. For most nutrient elements there is a net leaching effect. However, canopies in oligotrophic systems can scavenge materials from precipitation (Jordan et al., 1980). In eutrophic systems, limiting nutrients such as nitrogen and phosphorus are often removed from incident rainwater, (see Sections VIII . G , and H). Calcium and sulphate-sulphur may also be removed occasionally. Canopy scavenging of protons is also observed but involves ion replacement as well as ion exchange processes (Hoffman et al., 1980a). As far as I know, foliar uptake rates have never been directly measured for tree leaves in the forest. Washoff and leaching are probably the dominant processes in throughfall enhancement. Foliar uptake can be most important for nutrients in low supply (such as nitrogen) and for oligotrophic systems (such as epiphyte communities or dune swamp, and bog forests, and rain forests). Con- centration by evaporation may be physiologically important to foliage but is of minor importance as a long-term nutrient flux. The major processes in net throughfall deposition can be summarized for a given element by the following conservation relation (afier Horntvedt, 1974 and Bjor e t al., in Bache. 1977): NTF=W + L -- A 92 G. G. PARKER Where W, L and A are the portions washed off, leached from or sorbed to the canopy (kg h a - period-). In the absence of field evaluations for A, the variables L and A are combined in the term ( L -A), the net leachate. VII. PARTITIONING NET THROUGHFALL Net throughfall and its various sources pose a dilemma for balancing ecosystem nutrient budgets: the recycling of materials already in the system and input of external ones are combined in this one flux. This confusion may affect not only estimates of tree uptake rates but also input/output balances for whole forests, particularly (1) in forests where incident precipitation or dry deposition are major nutrient inputs or (2) where net throughfall fluxes are on the order of those in organic pathways (litterfall and root decomposition). Furthermore, uncertainties in fluxes can affect mineral cycling indices such as the fraction of inputs recycled in the system (Jordan et al., 1972). Separation of these sources would not only help balance forest nutrient budgets but also quantify losses of some atmospheric elements, which has proven very difficult by other means (Droppo, 1980). I t has long been recognized that net throughfall is a mixture of at least two classes of sources, those driving from within and from outside the plant-soil system (e.g. Ingham, 1950a; Hesse, 1957; Stenlid, 1958; Carlisle et al., 1966b; Henderson et al., 1977; Likens et al., 1977). Only lately has the problem of separating its origins been experimentally addressed. For some elements there exists some agreement as to which source dominates (K is clearly leached, for example); for most others the major processes are unknown. Some authors imply that net throughfall materials are atmospheric in origin (Ingham, 1950a,b; Tamm 1958; Nihlgard, 1970; Mayer and Ulrich, 1972; Westman, 1978). In much of the literature, net throughfall is often termed leachates (e.g. Stenlid, 1950; Tamm, 1951; Voigt, 1960b) or losses (e.g. Foster and Gessel, 1972; Johnson and Risser, 1974). Horntvedts (1975) review of the throughfall chemistry literature stress the lack of consensus regarding sources of net throughfall. Numerous approaches have been employed to separate the washoff and leachate portions of net throughfall. Few direct evaluations have been obtained for either component in field situations; most methods are indirect and applicable only under certain unique situations (Galloway and Parker, 1979). In the following sections I discuss the various approaches, their assumptions and limitations as general methods. A. Direct Approaches Only Banaszak (1 975) has reported comparisons of throughfall concen- trations of trees exposed to dry deposition with those protected from such THROUGHFALL, STEMFLOW IN FOREST NUTRITION 93 inputs. A plot of young loblolly pines (Pinus taeda) was enclosed in a plastic tent which was removed during storms. Concentrations of throughfall in this plot showed little difference with those under an adjacent, uncovered plot, for all elements analyzed, including sulphate and silica. The author concluded that net throughfall consisted largely of leachates. Usually, however, direct methods for separating leaching and washoff have involved the estimation of dry deposition to inert surfaces intended to simulate canopy filtering potential. Artificial fir trees (Schlesinger and Reiners, 1974), stacks of plastic sheeting (Nihlgard, 1970), vertical plexiglass baffles (Etherington, 1967) or screening (Lakhani and Miller, 1980), and polyethylene (Hart and Parent, 1974; authors data) or stainless steel screening (Iwatsubo and Tsutsumi, 1967) or both (Juang and Johnson, 1967) have been used. Actual precipitation is collected both under and adjacent to such surfaces. While depositions are always higher under such elaborate surfaces than to open, bulk funnels (e.g. Nihlgard, 1970; Parent, 1974; Miller and Miller, 1980) it is not clear that they simulate canopy filtering potential. Occasionally, inert surfaces such as Teflon or polyethylene plates (Lindberg et al., 1979), petri dishes and filter paper (White and Turner, 1970) or actual foliage rendered inert with acrylic spray (Art et al., 1974) are exposed to dry deposition in the field, retrieved and washed in the laboratory. In either case, washoff from actual canopies is estimated from the amount of material mobilized from these surfaces (that is, net throughfall from the artificial canopy). It is unknown how well such surfaces simulate the filtering action of actual plants. The precise leaf geometry and chemistry, both on large (leaf shape, arrangement and wetness) and small scales (microstructure and chemical binding sites) are critical in determining transfer rates. Furthermore, such simulations necessarily neglect foliar uptake, surface chemical conversions and interactions with incident precipitation. The controlled washing of plant parts is another direct approach. Foliage or whole shoots are periodically sampled and rinsed in solutions of known composition. Increases in wash water concentration are presumed to be due to dry deposition, if the leaching contribution over short intervals is constant (Lindberg et al., 1979). Such leaf washing has been employed in laboratory radiotracer experiments (Little, 1973, 1977; Raybould et al., 1977). White and Turner (1970) washed leaves and branches of several tree species exposed to known aerosol loads to calibrate the trapping efficiency of wind-vane mounted filter papers. Aerosol input to actual canopies was then estimated as the product of ambient concentrations sampled by the filter paper and the estimated trapping efficiency. The efficiencies reported are probably an overestimate, however, since the washings surely include some leached materials. Art et al. (1974) estimated sea salt impaction by multiplying the net throughfall deposition of sodium by the concentration ratios of various 94 G . G. PARKER cations to sodium in washings of acrylic-coated leaves placed in the upper canopy of a dune forest. They assumed that all of the net throughfall sodium derives from dry deposition and that no fractionation has occurred in the aerosol. However changes in sea salt ratios are observed over time (e.g. Clayton, 1972; Miller and Miller, 1980). A similar calculation was made by Graustein ( 1 980) for two canopies in high elevation New Mexico. However, instead of measuring sodium-to-cation ratios from captured aerosols, he employed those from incident precipitation. Such ratios are not likely to be identical for both wet and dry deposition. Direct approaches have focused on the estimation of dry deposition. Canopy simulation has the drawback that the fidelity of simulation is unknown, and dry deposition may be easily over- or under estimated. Leaf washings contain leachates in addition to dry deposits. Finally, ratio approaches must be used with caution since ratios can change. B. Indirect Approaches Net throughfall sources may also be partitioned with methods that do not estimate dry deposition directly. Mayer and Ulrich (1972) suggest that the washoff rate in a deciduous forest is equal to the rate of net throughfall flux during the leafless period. In computing a yearly income by canopy filtering, they assume the impaction efficiency of the leafless and fully expanded summer canopy are equal, neglecting that ( I ) non-leafy plant parts can be leached (Tukey, 1970a), (2) trapping efficiencies increase with canopy elaboration, and total input with canopy area and, (3) both aerosol concentrations and nutrient depositions in incident precipitation may exhibit seasonal variations (Galloway and Parker, 1979) as can the amount of precipitation. Eaton et af. (1978) have estimated sulphur dry deposition inputs to a northern hardwood watershed assumed to be in steady-state for sulphur, as the differences between bulk precipitation inputs and streamwater outputs. They argue that this flux (6.1 kg S ha yr -') appears consistent with estimates employing reported gas and particle measurements and reasonable deposition velocities. If all of this input were remobilized by precipitation it would represent 29% of the net throughfall. However, since leafless forests can produce appreciable net throughfall depositions of sulphur (Mayer and Ulrich, 1972) and since throughfall was collected only during the 5 month growing season (Eaton et al., 1973), this value may be an overestimate. This method relies on the steady state assumption which is certainly not valid for a given year but may be justified for long-term trends. Miller (l963), investigating a hard beech forest in maritime New Zealand, found the deposition ratios for major sea salt elements to be very similar ( 1.25 f .09), suggesting a close relation between wet and dry deposition. The THROUGHFALL, STEMFLOW IN FOREST NUTRITION 95 dry deposition rate was thus assumed to be 25% of the wet deposition rate for all elements. The net throughfall nutrient deposition in excess of this dry deposition was ascribed to leaching, which could be represented as: Leaching rate=THF- 1.25 INC where THF and INC are deposition rates in throughfall and incident precipitation, respectively. Attiwill ( 1966) points out that net throughfall of sea salt elements in Miller (1963) cannot be dry deposition alone since the sea salt ratios of gross throughfall are not observed in net throughfall. He argues that similarity of deposition ratios does not imply a constant dry deposition fraction, and suggests, instead, that leaching supplies most of the increases for all elements. Miller et a/. (1976) estimate the leaching fraction of a given element as the positive y-intercept of linear regressions of throughfall deposition on incident wet deposition. This implies that net throughfall for very small storms will consist largely of leachates. However, leaf washing studies using tracers have shown that soluble deposits are easily and quickly released on wetting (Little, 1973, 1977). Also, for most elements, the relation is probably not linear but curvilinear and convex upwards (Madgwick and Ovington, 1959; Attiwill, 1966; Bernhard-Renversat, 1975; Parker, 1982). Lakhani and Miller (1980) have proposed a refinement of this regression approach, requiring additional incident precipitation sampling under screened collectors and the assumption that additional depositions in screened collectors compared to bulk precipitation collectors are proportional to aerosol inputs to forest canopies. Other investigators have suggested that the importance of leaching in net throughfall is predicted by the mobility of the solute ion (Eaton et d., 1973; Henderson et al., 1977). Leachability is indicated in leaching studies (e.g. Gosz et a/., 1975) or by the ratio of net throughfall flux to total canopy content for each element (the inverse of the canopy turnover time via net throughfall). Such an index indicates the minimum level of the leaching contribution but will be of little value for elements well represented in both leachates and aerosols, such as Ca, Na, Si and S. Graustein and Armstrong ( 1 978) found that tree boles and airborne dusts in a New Mexico watershed had distincly different "SI-/'~S~ isotope ratios. Assuming the net throughfall isotope ratio to be a linear combination of the bole (leaching) and dust (washoff) values, they calculated the percentage of strontium in net throughfall due to dust input at 65% and 26% for a Spruce fir and an aspen canopy, respectively. The strontium results were extended to other elements using ionic ratios (Graustein, 1980). The extrapolation suffers from the drawbacks in the ratio approach previously mentioned, but the isotope approach has much merit. Temporal observations on throughfall quality may provide some clues as to 96 G . G . PARKER the relative importances of washoff and leaching. Within storm measurements of throughfall quality are few and show a variety of patterns (Sollins and Drewry, 1970; Yawney et af., 1971; Cole and Johnson, 1977; Richter and Granat, 1978; Lindberg, Parker, unpublished data). Sequential throughfall concentrations show either: (1) wide oscillations without clear trend, es- pecially in small storms (e.g. Richter and Granat, 1978), (2) patterns which roughly mimic those of incident precipitation (Sollins and Drewry, 1970; Parker, unpublished data) or, (3) a regular, nearly exponential decline with successive amounts of precipitation (Yawney and Leaf, 1971; Lindberg, Parker, unpublished data). Such a pattern of decline is shown for successive measurements of electrical conductivity in small, frontal storms in rural Virginia in Fig. 13 (below). Dry deposited material, which is quickly released on wetting of the canopy (Little, 1973, 1977), appears in the early fractions of throughfall, whereas 250 200 - E 0 S \ E a 150 ? =. c u 3 D 0 c E 100 u - i? W 50 0 \ 0 1 2 3 4 0 1 2 3 4 5 6 7 8 Cumulotlve precipitation during o storm mm Fig. 13. Electrical conductivity in sequentially sampled incident precipitation and throughfall under white pine (Pinus srrobus). Numbers are storm designations. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 97 leachates certainly dominate the later, more dilute fractions. Richter and Granat (1978) concluded from the lack of trend in sequential sulphate concentrations that dry deposition was not important for sulphur in net throughfall. On the other hand, their data show a strong decrease in throughfall SO,/K ratios during the storms, suggesting that most sulphate is lost initially. Clearly, detailed investigations of sequential throughfall may provide additional insights into the processes of material transfer. A promising method which has been inadequately explored involves the marking of the pool of leachable materials. Radioisotope or dye tracers could be introduced into the tree's translocation stream. A separation of leaching and washoff could be obtained by monitoring throughfall and incident rainwater for changes in activity and concentration of labelled and unlabelled compounds. Similar techniques were used in the original work on the physiology of leaching and field studies directed at the movement of material through a whole plant system (Witherspoon, 1964; Waller and Olsen, 1967; Thomas, 1967, 1969; Dayton, 1970; Kimmins, 1972). While those studies have shed light on the pathways and rates of recycling for calcium, cesium and strontium (and their analogs), they were not designed to address the problem of recycling and input in net throughfall. Below is compiled a table (Table 4) describing the percentage contributions of leachates to throughfall enhancement for major cations as found by various methods discussed in this section. Table 4 Estimates computed from the literature for the percentage of net throughfall due to leaching for several cations. Element Forest system Na Ca Mg K Source 200 m from ocean New York Corsican pine. Scotland Solling beech forest, Central Germany Hard beech, mari- time New Zealand Spruce firh High elevation New Mexico 0" 53.6 8.2 81.4 Art er al.. 45.51 30.0 58.8 56.7 Miller P I a/.. 39.3 65.0 53.9 74.3 Mayer and 1974 I976 Ulrich, 1972 0" 50.0 0" 84.4 Miller, 1963 0" 48.4 65.7 82.8 Graustein, 1980 ( -50) (20.9) (48.6) (73) Graustein and Armstrong ( 1978) "Defined as zero in the method employed, strontium isotope-ratio method). Using sodium-ratio method (parenthetical values 98 G. G. PARKER There is much agreement between these estimates considering the diversity of systems studied and methods employed. Leaching appears to supply 60-90% of potassium, 50-60% of magnesium, 20-60% ofcalcium and 40-50x of the sodium in net throughfall. In maritime forests deposition sources for Na and Mg dominate. VIII. ELEMENTS IN THROUGHFALL A. Carbon Where comparisons exist, carbon is found to be the most concentrated element in throughfall (Carlisle et af . , 1966b; Eaton eta/., 1973; Hoffman e ta / . , 1980a,b). Organic carbon appears to dominate (Carlisle et a/., 1966b) and sugars comprise much of the soluble portion (Carlisle, 1965; Carlisle et a/., 1966b; de Boois and Jansen, 1975) though numerous other compounds are present (Hoffman er al., 1980a,b). Bicarbonate is found in regions away from acidified precipitation (Nye, 1961; McColl, 1970; Feller, 1974; Johnson, 1975; Ugolini et a/., 1977; Cole and Johnson, 1977) where it would be purged as CO, (Cronan, 1978). Bicarbonate may be the dominant throughfall anion in undisturbed areas, since leachate cations commonly transfer as bicarbonate salts (Tukey, 1970a). Stemflow is also quite concentrated in carbon (Carlisle et a/., 1967; Mina, 1967; Gersper and Hollowaychuck, 1971; Mahendrappa. 1974), containing coloured compounds (Nye, 1961; Stark, 1973; Yadav and Mishra, 1980) which are probably stem and bark decomposition products. Though incident precipitation may contain carbon (Carlisle, 1965; Carlisle et a/., 1966b, 1967; Eaton et a/., 1973; Galloway et al., 1976), throughfall carbon is nearly all derived from the plant. However, some derives from decom- position and consumer activities on plant surfaces, behaving as a washed-off deposit though originating within the system. B. Hydrogen In regions removed from acid rain, the pH of throughfall is higher than of incident precipitation for hardwood stands. Coniferous forest throughfall usually has a lower pH than incident precipitation. This represents an increase in free acidity under conifers and a decrease under hardwoods. In areas with acidic precipitation even hardwood throughfall pHs can be lower than incident ones. This effect is controlled by the presence of the foliage canopy because in deciduous forest throughfall acidity varies with season. Hoffman et a/. (1980a) found throughfall pHs to be lower than for incident precipitation under a leafless Quercus prinus canopy, but during the growing season the reverse pattern was observed. Apparently the leafy canopy buffers hydrogen THROUGHFALL, STEMFLOW IN FOREST NUTRITION 99 ion input through ion exchange, but the bark and stem tissues are net proton sources. pH measures only a solution's free acidity. Other contributions to total acidity must be obtained by other means. An acidity titration by Gran's method (Gran, 1952; Askne and Brosset, 1972; Brosset, 1976) can separate the total titratable acidity into components from strong acids (such as anthro- pogenically produced H2SO4 and HNO,) and weak ones (such as organic acids and metal complexes). Hoffman et al. (1980a) found that total titratable acidity of precipitation remained relatively constant before and after canopy interaction. However, the percentage contribution by weak acids was altered, being small (32%) in incident rain but larger (55%) in throughfall. The summertime canopy removed strong acids from precipitation and replaced them with weaker organic ones. In more pristine areas, throughfall has appreciable concentrations of bicarbonate (Nye, 1961; Johnson et al., 1975; Cole and Johnson, 1977; Feller, 1977) and its pH is usually higher than for incident precipitation. C. Potassium Potassium commonly shows the highest throughfall concentrations of the inorganic nutrients K depositions in incident precipitation are low, on the order of a few kg ha-' yr-' (higher in tropical or coastal areas). Therefore its deposition ratio is consistently high, averaging 11.2 (cf. Table 1). Leaching studies indicate this monovalent cation is extremely mobile (Tukey et al., 1958; Gosz et al., 1975). Foliar concentrations of K are quite high, on the order of 0.6-1.5% (Mitchell, 1936). K has a low canopy residence time (the ratio of canopy load to net throughfall flux); its uptake and turnover rates are both considerable. Though potassium is extremely mobile, Stone and Kszystyniak (1977) found that 407; of foliar potassium in a red pine stand could be traced to a fertilization 23 years previous. Much conservation of this element is certainly affected in throughfall recycling. Since ambient air concentrations of the element are low, while its mobility and foliar levels are extremely high, net throughfall potassium derives almost entirely by leaching, probably more than 90%, even in maritime regions. D. Calcium Calcium also exhibits significant rates of deposition in throughfall. Incident concentrations are higher than for potassium yet the deposition ratios can be appreciable, averaging 2.86 (cf. Table 1). Coastal forests (e.g. Art et a/., 1974) and those near alkali flats (Parent, 1972; Hart and Parent, 1974) exhibit much calcium in both incident and throughfall precipitation. 100 G . G . PARKER Much net throughfall calcium undoubtedly derives from leaching (Meck- lenberg et af., 1966; Thomas, 1969; Tukey, 1970a). Calcium is not re- translocated and appears to accumulate in apical tissues and leaves (Epstein, 1972), where it appears to be in a mobile pool. The importance of canopy filtering falls off rapidly with distance from a source since airborne calcium is contained largely in large particle soil dusts and sea salts, which have a high settling velocity. Ambient air and incident precipitation concentrations of calcium can be high (particularly in dusty episodes), but forest foliage contains much calcium (0.8% of dry weight) and leaching studies (Gosz et af., 1975; Clements et af., 1972; Tukey, 1970a) have shown that the foliar load is relatively mobile (between 1 and 10% leachable). Probably 60-800/, of net throughfall calcium is leached, especially away from dust and sea salt sources. E. Magnesium Forest canopies commonly enhance magnesium deposition by a factor of 7 (see Table 1). Probably most of this is in dry deposition, both soil and sea salt derived. Low foliar levels of magnesium and a moderate leaching percentage (l-lOo/, according to Tukey, 1970a) suggest that dry deposition probably supplies SO-SO% of net throughfall magnesium and probably > 80% in marine areas. F. Sodium Foliar sodium is very mobile during leaching (> 25% of foliar amounts removable, Tukey, 1970) but has low concentrations (on the order of .02-.0So/,, Guha and Mitchell, 1966). Incident depositions are not small and yet sodium deposition ratios average 2.41. Net throughfall sodium surely originates from dry deposition in maritime situation (> 90%) and since the leaching process is inadequate to supply observed depositions, I suspect that much net throughfall sodium derives from dry deposition even in forests removed from strong aerosol sources. G. Phosphorus The canopy pool of phosphorus is large (0.2% by leaf dry weight) (Mitchell, 1936) but is mostly sequestered in immobile, organic forms (Epstein, 1972). Its large deposition ratios (3.94) are due more to small incident depositions than to appreciable throughfall. Some phosphorus could derive from soil dusts but phosphate minerals have low solubilities in the pH range of precipitation (3.5-5.5). Nearly negligible amounts of P in ambient air suggest that leachates THROUGHFALL, STEMFLOW IN FOREST NUTRITION 101 contribute 70-90% of throughfall phosphate, probably > 90% of total P and nearly all of the organic P. H. Nitrogen Net throughfall nitrogen derives from leaching, foliar uptake and decom- position processes in the leaf surface community, with probably less contri- bution from gaseous or aerosol input. Canopy nitrogen is tightly conserved, often through foliar uptake of nitrogen from incident precipitation. Negative net throughfall is most commonly reported for NH,-N (Horntvedt and Joranger, 1974; Paivanen, 1974; Wells et a/., 1975; Feller, 1977; Verry and Timmons, 1977; Richter and Granat, 1978; author's data) though NO3-N (Paivanen, 1974; Verry and Timmons, 1977; Schlesinger, 1978; Miller et al., 1976; Graustein, 1980; Weaver and Brown, 1980) and total nitrogen are also removed from incident precipitation (Miller, 1963; Carlisle et al., I966b; Foster and Gessel, 1972; Foster, 1974; Wells and Jorgenson, 1974; Rolfe eta/., 1978; Jordan et al., 1980). Apparently the leaf and leaf-dwelling organisms can assimilate nitrogen very rapidly for this is observed even when biocidal agents have been added to throughfall samplers to prevent microbial uptake of N (Paivanen, 1974; Schlesinger, 1978; Westman, 1978). Foliar levels are high (1.0-3.5%D.W., second only to carbon) yet leaching is very low (Tukey, 1970a). Airborne nitrogen is present as gaseous NO,, H N 0 3 and in small ammonium sulphate and organic nitrogen aerosols (Butcher and Charlson, 1972) which ought to be efficiently filtered by forest vegetation (Chamberlain, 1975a). Little is known about NO, uptake by forest canopies. In spite of potentially large aerosol sources, the presence of a large canopy pool of nitrogen and rapid foliar uptake rates of ammonium suggest leaching (or foliar uptake) will dominate the net throughfall processes. I. Chlorine Chlorine wet and dry depositions can be substantial, even in continental areas. Errikson (1 955) concluded that steady state river water concentrations of chloride required large (several kg ha -' yr -') dry atmospheric inputs and suggested that aerosol capture by trees and subsequent washoff by rain could provide balancing amounts. Juang and Johnson (1967) argue that one third of total annual chlorine input to the Northern Hardwoods ecosystem is dry deposition by canopy filtration. Chlorine deposition ratios are high (3.03f 1.49) but foliar contents (0.14% D.W. in corn, Epstein, 1972) and leaching mobility (< 1%, Tukey, 1970a) are low, suggesting that most of chlorine in net throughfall derives from atmospheric dry deposition, probably 90% in maritime regions. Whether this input I02 G. G. PARKER derives from gaseous or particulate impaction or remobilization of absorbed gases is unclear. J. Sulphur Sulphur is commonly the major anion in both incident precipitation and often also in throughfall, in industrialized (e.g. Haughbotn, 1973; Heinrichs and Meyer, 1977) and pristine areas (Graustein, 1980; Jordan et al., 1980). Deposition ratios are consequently not exceptionally high (averaging 2.2, see Table 1). Dry deposition is potentially large since atmospheric concentrations of sulphur-containing particles and gases are high. However, the leaching physiology of sulphur is poorly known (Turner et al., 1977; Turner and Lambert, 1980) and consequently, the partition of net throughfall sulphur is very much in question. This confusion is evident in Table 8 which presents literature estimates of leaching contributions to net throughfall depositions. These range from 13-100% and show no clear trend. In certain situations the degree of throughfall sulphate enhancement may correspond to the proximity and strength of sulphur emission sources (see Section V. C . ) , the airborne sulphur loads between storms (Bache, 1977; McColl and Bush, 1978) or the length of the dry interval between storms (McColl and Bush, 1978). As previously noted, coniferous canopies appear to be better aerosol receptors than hardwood canopies under these conditions. Table 5 Incident (INC), Throughfall (THF) and Net Throughfall (NTF) sulphate-sulphur depositions in several systems where attempts were made to partition net throughfall sources. Adapted from Parker et al. (1980). References Raybould if ul., 1977 Eaton et a/., 1978 _ _ _ _ ~ . _ _ _ _ ~ Lindberg rt ul., 1979 Bache, 1977 Ulrich rf a/., 1978 Miller, 1963 Sampling period Z N T E Depositions (kg ha ) due to dry INC T H F NTF deposition 66 daysd INC many years T H F 5 months 2 years 153 days 127 days 8 years 1 year 2.06 4.42 2.36 13 12.7 33.7 21.0 21.9 13.0 32.0 19.0 26.3 5 .8 19.6 13.8 89.1 ~ ~ 4.5 14.8-44.4 23.8 53.1 29.3h 88.9 - f2 .7 * 11.2 k 9 . 4 - + 12.0 8.4 10.4 2.0 100 Using data of Bjor (in Bache, 1977), canopy . Includes stemflow, Standard deviation, A wheat THROUGHFALL, STEMFLOW IN FOREST NUTRITION 103 However, such clear source/receptor relationships are not usually obvious for net throughfall sulphur (e.g. Lindbergetaf., 1979; Parker, 1982). There are indications that atmospheric inputs are not dominant. Bjor (in Bache, 1977), Bache (1977), Lindberg (personal communication) and the author have found that net throughfall sulphate is strongly correlated with precipitation depths (even during very rainy periods when atmospheric loads are low), suggesting a leaching source. Leachates probably comprise more than 50% of net throughfall sulphate in pristine regions (Parker et a/., 1980). This may be less under hardwoods and less still under conifers in the vicinity of strong sulphur sources. This indication is preliminary; work is necessary on the chemical forms and mobility of foliar sulphur, its leaching physiology and particularly, on the fate of canopy-trapped sulphur gases. K. Silicon, iron and aluminium Major soil dust elements silicon, iron and aluminium occur in low con- centrations in both incident and throughfall precipitation (McColl, 1970; Stark, 1973; Golley et af . , 1975; Heinrichs and Mayer, 1977; Feller, 1977; McColl and Bush, 1978). Their deposition ratios barely exceed unity. The low precipitation concentrations of these elements may be due to solubility limitations, since their airborne particulate concentrations can be high (Lindberg and Harriss, 198 1). Tree leaves contain much silicon (0.149/, D.W., Guha and Mitchell, 1966) but less iron and aluminium (0.01-0.09% D.W., Guha and Mitchell, 1966). Iron (and presumably aluminium) is leached with great difficulty (< I%, Tukey, 1970a) and silicate in net throughfall is extremely low relative to its foliar content. The soil mineral elements are probably supplied to net throughfall mainly via dry deposition, but rates of solute deposition are low. L. Heavy metals Comparisons of depositions between throughfall and incident precipitation have been reported for heavy metal elements by Heinrichs and Mayer (1977), Mayer and Ulrich (1 972), Stark (1973), Golley et af . (1 9759, McColl and Bush (1978) and Lindberg et af . (1979). Heinrichs and Mayer (1977) observed net throughfall enhancements for Ni, Pb, Sb, Bi, Cd, Mn and T1 and foliar uptake for Zn, Cr, Cu and Hg for a beech forest in central Germany. Golley et al. (1975) calculated only negative net depositions for Sr, Mn, Cu, Co, Zn and Pb (these annual budgets, however, are based on only a few samples). Stark (1973) has observed positive net throughfall for Cu and Zn in Nevada. For many of these metals both airborne and foliar levels are so low that it is unknown which sources control net throughfall depositions. For most elements that leaching process is responsible for the major part of 104 G . G. PARKER nutrient depositions in net throughfall except in particular situations. On a global basis the importance of recycling through leaching relative to inputs via dry deposition probably decreases in the order C = K > P > N > Ca > S > CI > Na > Mg> Si = Al= Fe. This scheme does not apply where net throughfall is negative or where it involves the exchange of ions input and leached (such as for hydrogen, Hoffman et af., 1980a). Figure 14 (below) summarizes the range of contri- bution to net throughfall deposition by leaching or dry deposition estimated for various nutrients reported in the literature (for example, net throughfall carbon ranges from 90-1000/, from leaching, or from zero to 10% via dry deposition). 0 % Leaching 50 I00 I00 50 0 % Dry Deposition Fig. 14. Probable ranges for the percentage of net throughfall due to dry deposition and leaching for major throughfall elements (leaching plus dry deposition equals 100%). IX. RECOMMENDATIONS FOR FUTURE RESEARCH Large amounts ofdata o n throughfall and stemflow now exist in the literature (cf. Appendix table). Little purpose would be served in continued studies of a solely descriptive nature. Instead, there is a need for more research directed at THROUGHFALL, STEMFLOW IN FOREST NUTRITION the mechanisms controlling throughfall, its requirements by and effects upon the plant-soil system. Attention would be profitably focused on the following questions: (1) To what extent is vegecation dependent on nutrient recycling in throughfall? What fraction of throughfall nutrients are actually taken up? Are some elements more or less critical than their mobility in throughfall would suggest? (2) Throughfall is an important route for the transfer of information-rich allelochemic compounds between foliar strata (e.g. Whittaker and Feeney, 1971). Does the spatial pattern in soil properties and in understory vegetation reflect the influence of the overstory, or vice versa? Manipulative experiments where throughfall from various species is distributed to the understories of other species may shed some light on these processes. (3) The importance of throughfall to grasses, shrubs and crop plants has not been adequately investigated. Only a few studies exist for non-forest vegetation (Egunjobi, 1971; Best, 1976; Raybould et al., 1977; Calvo de Anta et af . , 1979; Stinner, 1981), though several studies have incorporated ground flora throughfall into overall throughfall estimates (Carlisle et al., 1967; Astrup and Biilow-Olsen, 1974; Verry and Timmons, 1977). (4) The problem of the sources of net throughfall solutes is far from settled and needs more work. Since dry deposition is so difficult to measure, quantification of leaching rates may be fruitful experimental approach. Several approaches are possible such as radiotracers studies on individual trees or wind tunnel experiments with known aerosol loads. Alternatively, throughfall collection under particular canopy types in transects away from strong aerosol sources might shed light on the process of air filtration. Also, more attention must be paid to (1) the effects of acidified precipitation on throughfall quality in field situations and, (2) the role of canopy scavenging of gases such as SO2 and NO,. (5) Finally, methods of throughfall sampling, handling and analysis (chemical and statistical) require standardization, as has been suggested for incident precipitation (Galloway and Likens, 1978). A great many estimates in the literature are based on too few samples or too little attention to factors which alter precipitation chemistry. 105 X. CONCLUSIONS ( 1 ) Throughfall and stemflow contribute significant amounts of mineral nutrients to the floors of the worlds forests. For all but protons and some species of nitrogen, this exceeds those in incoming precipitation and, for K, S, and Na, is more than in litterfall. Moreover, unlike litterfall, such dissolved materials are immediately available for reabsorption by plants. The canopy, 106 G. G. PARKER therefore, has considerable short-term control over the circulation of many elements. ( 2 ) Net throughfall and stemflow combine materials originating outside of the system (dry deposition between storms) with those from within it (leachates, exudates and decomposition products). Incident precipitation is therefore an underestimate of input and net forest water an overestimate of canopy-mediated recycling. Air filtration by forest canopies may exert some influence on ambient air concentrations. As anthropogenic atmospheric inputs increase, so will the importance of quantifying the ability of plants to remove these materials. Without estimates of dry deposition and subsequent washoff, nutrient cycles are unbalanced and atmospheric losses poorly quantified. (3) Throughfall and stemflow show much variability in concentration and deposition, far more than for incident precipitation. Descriptions of such variability ought to be a standard part of throughfall studies. Sampling and analyses of forest waters present special chemical and statistical problems. For nutrient budgets, a major difficulty lies in obtaining an accurate and independent estimate of the forest hydrological budget. (4) Stemflow is a concentrated and localized nutrient flux which contri- butes only 5-20/, of the nutrient deposition in gross forest water. However it is the major mineral flux to a narrow area about the bole. Nutrient deposition in stemflow is quite variable and difficult to quantify, particularly in mixed and uneven-aged forests. (5) For the majority of elements, leaching is the major contribution in net throughfall. The importance of leaching relative to dry deposition in enhancing throughfall is argued to follow the sequence C = K > P > N > Ca > S > CI > Na > Mg > Si =A1 = Fe. Ultimate controls over the partitioning of net throughfall sources are due to the forest system itself, especially factors associated with (1) the degree of biological intervention in nutrient circulation (growing season length, season- ality and possibly, temperature), ( 2 ) the vigor and nutritional state of the system, (3) the ambient load and character of depositable material (source strength and proximity), (4) the filtering capacity of the plant itself (receptor geometry) and also, the nature of the annual precipitation income (amount, intensity, frequency, type and composition). XI. ACKNOWLEDGEMENTS This work was sponsored by the US Department of Energy. I would like to thank Drs D. A. Crossley, Jack Ewel, James Galloway, Bruce Haines, Bruce Hayden, Eugene Odum, Michael Pettelle, Roger Pielke, Thomas Wolaver and THROUGHFALL, STEMFLOW IN FOREST NUTRITION 107 Messrs. William Keene and David Johnson for their comments on this work, Kula Campbel l and Brenda Pat terson for preparat ion of the manuscript, Thomas A d a m s and Cather ine Parker for suppor t and enthusiasm. REFERENCES Abee, A. and Lavender, D. (1972). Nutrient cycling in throughfall and litterfall in 450 year-old Douglas Fir stands. In "Research on Coniferous Forest Ecosystems First Year Progress in the Coniferous Forest Biome" (Eds J. F. Franklin, L. J . Dempster and R. H. Waring), pp. 133-143. US/IBP, Proc. Symp. Northwest Scientific Association, Pacific Northwest Forest Experiment Station, Forest Service, USDA. Portland, Oregon. Abrahamsen, G., Bjor. K., Horntvedt. R. and Tveite, B. (1976). Effects of acid precipitation on coniferous forest. In "Impact of Acid Precipitation on Forest and Freshwater Ecosystems in Norway (Ed. F. H. Braekke), pp. 36-63. SNSF Project, FA 1/75, AS-NLH, Norway. Akhtyrtsev, B. P. and Sviridova, I. K . (1975). Rainfall leaching of some elements from tree crowns, forest litter and soils in the Central Russian forest-steppe. Biol. Nauhi (Mosc.) 18, 123-128. Anderson, R. L., Loucks, 0. L. and Swain, A. M. (1969). Herbaceous response to canopy cover, light intensity and throughfall precipitation in coniferous forests. Art, H. W., Bormann, F. H., Voigt, G. K. and Woodwell. G. M. (1974). Barrier island forest ecosystem: role of meteorologic nutrient inputs. Scirrice 184. 60 62. Askne, C. and Brossett, C. (1972). Determination of strong acid in precipitation, lake- water and air-borne matter. A m o s . Eriiiron. 6 , 695-696. Astrup, M. and Bulow-Olsen, M. (1979). Nutrient cycling in two Danish beech (Fugu.5 sylvatica) forests growing on different soil types. Holarct. Ecol. 2. 125-1 29. Attiwill, P. M. (1966). The chemical composition of rainwater in relation to cycling of nutrients in mature eucalyptus forest. Plant and Soil 24, 390408. Aussenac, G., Bonneau, M. and Le Tacon, F. (1972). Restitution des elements mineraux au sol par I'intermediare de la litiere et des precipitations dans quatre peuplements forestiers de I'est de la France. Orcol. Plunt. 7, I 21. Bache, D. H. (1977). Sulphur dioxide uptake and the leaching of sulfates from a Pine forest. J . Appl. Ecol. 14, 881-895. Baker, J., Hocking, D. and Nyborg. M. (1977). Acidity of open and intercepted precipitation in forests and effects on forest soils in Alberta, Canada. Water. Air ut id Soil Pollutiori 7, 449460. Banaszak, K. T. ( 1975). Relative throughfall enrichment by biological- and aerosol- derived materials in loblolly pines. Miss. Water Res. Inst., Mississippi State Univ., PBP245-258/9ST. Nat. Tech. Info. Ctr., Springfield, Virginia. Basilevich, N. I. and Rodin, L. E. (1966). The biological cycle of nitrogen and ash elements in plant communities of the tropical and subtropical zones. Forestrv Ahstrurrs 27, 357-368. (Translated from Botaniceskig Zurnal 49. 185-209.) Belot, Y . and Gauthier, D. (1975). Transport of micronic particles from atmosphere to foliar surfaces. In "Heat and Mass Transfer in the Biosphere" Part 1. "Transfer Processes in the Plant Environment" (Eds D. A. deVries and N. H. Afgan). pp. 583-591. John Wiley, New York. Bennet. T. H. and Hill, A. C. (1975). Interactions of air pollutants with canopies of Ecol0g.V 50, 255-263. 108 G. G. PARKER vegetation. In Responses of Plants to Air Pollution (Eds J . B. Mudd and T. T. Kozlowski), pp. 273-306. Academic Press, New York. Bernhard-Renversat. F. (1 975). Nutrients in throughfall and their quantitative importance in rain forest mineral cycles. In Tropical Ecological Systems-Trends in Terrestrial and Aquatic Research (Eds E. Medina and F. B. Golley), pp. 153-1 59. Springer-Verlag, New York. Best, G. R. (1976). Treatment and biota of an ecosystem affect nutrient cycling. PhD Dissertation. University of Georgia, Athens, Georgia. Best. G. R. and Monk, C. D. (1975). Cation flux in hardwood and White Pine watersheds. I n Mineral Cycling in Southeastern Ecosystems (Eds F. G. Howell, J . B. Gentry and M. H. Smith), pp. 847-861. Natl. Tech. Inf. Ctr., Springfield, Virginia. Bollen, W. B., Chen. C. S., Lu, K. C. and Tarrant, R. F. (1968). Effect of stemflow precipitation on chemical and microbiological soil properties beneath a single alder tree. I n Biology of Alder (Eds J. M. Trapp, J. F. Franklin, R. F. Tarrant and G. M. Hansen), pp. 149-156. Pacific Northwest Forest and Range Experiment Station USDA Forest Service, Portland, Oregon. Boynton, D. (1954). Nutrition by foliar application. Ann. Rev. PI. Physiol. 5. 31-54. Bray. J . R. and Gorham. E. ( 1 964). Litter production in forests of the world. Adv. Ecol. Bringmark. L. (1980). Ion leaching through a podsol in a Scots Pine stand. In Structure and Function of Northern Coniferous Forests ~ an Ecosystem Study (Ed. T. Persson). Ecol. Bull. (Storkkoln7) 32, 341-361. Brinson. M . M.. Bradshaw. H. D., Holmes, R. N. and Elkins, J . B., Jr. (1980). Litterfall, stemflow and throughfall nutrient fluxes in an alluvial swamp forest. Emlogy 61. 827-835. Brosset. C. (1976). A method of measuring airborne acidity: its application for the determination of acid content on long-distance transported particles in drainage water from spruces. Water, Air and Soil Pollution 6. 259-275. Brown. A. H. F. (1974). Nutrient cycles in oakwood ecosystems in northwest England. I n The British Oak. Its history and natural history (Eds M. G. Morris and F. H. Perring). Pendragon Press, Cambridge. Browne. F. G. (1932). Bleeding of Meranti Sarang Punai. Malay Forester 1. 177. Butcher, S. B. and Charlson. R. J. (1972). An Introduction to Air Chemistry 241 pp. New York. New York. Calvo de Anta. R. M.. Gonzalez, A. P. and Viqueira, F. D. (1979). Nuevos datos sobre la influencia de la vegetacion en la formacion del suelo en Galicia. 11. Aportes de elementos por lavado de cubierta y tronco. An. Edufol. Agrobiol. 38, 1675-1692. Carlisle. A. (1965). Carbohydrates in the precipitation beneath a Sessile Oak, Quercus pitrnca (Mattushka) Liebl. canopy. Plunt and Soil 24, 399400. Carlisle, A,. Brown, A. H. F. and White, E. J. (1966a). Litterfall, leaf production and the effects of defoliation by Tor1ri.v viridana in a Sessile Oak (Quercu.7 petraea) woodland. J . Ecol. 54. 65-85. Carlisle, A,. Brown, A. H. F. and White, E. J . (1 966b). The organic matter and nutrient elements in the precipitation beneath a Sessile Oak (Quercus petraea) canopy. J . G,ol. 54. 87-98. Carlisle. A,, Brown, A. H. F. and White. E. J . (1967). The nutrient content of tree stemflow and ground flora litter and leachates in a Sessile Oak (Quercus perraea) woodland. J . Ecol. 55. 615-627. Carrier. C. L. (1958). Rain from a Sliorea das~ph~~ l la tree. Malaj- Forester 21, 274. Chamberlain. A. C. (197%). The movement of particles in plant communities. In RCJS. 2. 101.-157. THROUGHFALL, STEMFLOW I N FOREST NUTRITION 109 "Vegetation and the Atmosphere" Vol. I . "Principles" (Ed. J. L. Monteith). pp. 155-203. Chamberlain, A. C. (1975b). Pollution in plant canopies. / ) I "Heat and Mass Transfer in the Biosphere" Part I , "Transfer Processes in the Plant Environment" (Eds D. A. deVries and H. H. Afgan). pp. 561-582. John Wiley, New York. Chamberlain, A. C. (1980). Dry deposition of sulfur dioxide. I I I "Atmospheric Sulfur Deposition" (Eds D. S. Shriner. C. R. Richmond and S. E. Lindberg). pp. 185 197. Ann Arbor Science, Ann Arbor. Michigan. Clayton, J. L. (1972). Salt spray and mineral cycling in two California coastal ecosystems. Ecologj, 53, 74-8 1. Clements, C . R.. Jones, L. P. H. and Hopper. M. J. (1972). The leaching of some elements from herbage plants by simulated rain. J . Appl. E d . 9, 249 260. Clements, R. G. and Colon, J . A. (1975). The rainfall interception process and mineral cycling in a Montane Forest in eastern Puerto Rico. h i "Mineral Cycling in Southeastern Ecosystems" (Eds F. G. Howell, J. B. Gentry and M. H. Smith). pp. 813-823. Natl. Tech. Inf. Ctr.. Springfield, Virginia. Clesceri, N. L. and Vasudevan. C. (1980). Acid precipitation, throughfall chemistry and canopy processes. 111 "Ecological Impact of Acid Precipitation" (Eds D. Drablras and A. Tollan). pp. 258-259. SNSF Project, Oslo-As. Cole, D. W., Gessel. S. P. and Rice. S. F. (1967). Distributions and cycling of nitrogen. phosphorus. potassium and calcium in a second-growth Douglas Fir ecosystem. h i "Symposium on Primary Productivity and Nutrient Cycling in Natural Ecosys- tems" pp. 197-232. University of Maine Press, Orono. Maine. Cole, D. W. and Johnson, D. M. (1977). Atmospheric sulfate additions and cation leaching in a Douglas Fir ecosystem. M/utu Rrsour. Rrs. 13. 313-317. Cole, D. W. and Rapp. M. (1981). Elemental cycling in forest ecosystems. 111 "Dynamic Properties of Forest Ecosystems" (Ed. D. E. Reichle), pp. 341409. Cambridge University Press, New York. Comerford, N. B. and White, E. H. (1977). Nutrient content of throughfall in Paper Birch and Red Pine stands in northern Minnesota. Ccrri. J . For. Rex 7. 556-561. Corlin, J. W. (1971). Nutrient cycling as a factor in site productivity and forest fertilization. In "Tree Growth and Forest Soils" (Eds C. T. Youngberg and C. B. Davey), pp. 313-325. Oregon State University Press, Corvalis, Oregon. Cronan, C. S. (1978). Solution chemistry of a New Hampshire subalpine ecosystem: biogeochemical patterns and processes. PhD Thesis. Dartmouth College. Cronan, C. S. and Schofield, C. L. (1979). Aluminium leaching response to acid precipitation: effects on high elevation watersheds in the northeast. Scieticr 204, Cronan, C. S., Reiners, W. H. and Reynolds, R. C.. Jr. (1980). The impact of acid precipitation on forest canopies and soils in the northeastern US. "Ecological Impact of Acid Precipitation" (Eds D. Drablm and A. Tollan), pp. 158-159. SNSF Project, Oslo-As. Crossley, D. A,, Jr. and Seastedt, T. R. (1981). Effects of canopy arthropod consumption and leaf biomass on throughfall chemistry of a successional forest in the southern Appalachians. Bull. Ecol. SOC. A m . 62, 106. Dansereau, P. (1957). "Biogeography: An Ecological Perspective" 394 pp. Ronald Press, New York. Dassman, R. E. (1976). "Environmental Conservation" 427 pp. Wiley, New York. Dayton, B. R. (1970). Slow accumulation and transfer of radiostrontium by young De Boois, H. M. and Jansen, E. (1975). Effects of nutrients in throughfall water and of 304-306. loblolly trees (Pinus ruedo L.). Ecologj. 51, 204-216. 110 G. G. PARKER litterfall upon fungal growth in a forest soil layer. Pedohiologiu 16. 161- 166. Denaeyer-DeSmet, S. (1966). Bilan annuel des apports delements minereaux par les eaux de precipitation sous couvert forestier dans la forkt melangee caducifoliee de Blaimont. Buff. Soc. R. BOI. Befg. 99, 345-375. Droppo, J . G. (1976). Dry removal of air pollutants by vegetation canopies. In Proc. 4th Natl. Conf. on Fire and For. Met. (Tech. Coords. B. H. Baker and M. A. Basberg), pp. 200-208. USDA Genl. Tech. Rpt. Droppo, J . G. (1980). Experimental techniques for dry-deposition measurements. In Atmospheric Sulfur Deposition (Eds D. S. Shriner. C. R . Richmond and S. E. Lindberg), pp. 209-221. Ann Arbor Science, Ann Arbor, Michigan. Durant, C. L. (1932). Bleeding of trees. Malay Forester 1. 220. Duvigneaud, P. and Denaeyer-DeSmet, S. ( 1 970). Biological cycling of minerals in temperate deciduous forest. In Ecological Studies, * * 1. Analysis of Temperate Forest Ecosystems (Ed. D. E. Reichle), pp. 199-225. Springer-Verlag. New York. Duvigneaud, P. and Denaeyer-DeSmet, S. (1975). Mineral cycling in terrestrial ecosystems. In Productivity of World Ecosystems (Eds D. E. Reichle, J . F. Franklin and D. W. Goodall). pp. 133-154. National Academy of Sciences, Washington, D.C. Eaton, J. S., Likens, G. E. and Bormann, F. H. (1973). Throughfall and stemflow chemistry in a northern hardwood forest. J . Ecology 61, 495-508. Eaton, J . S., Likens, G. E. and Bormann, F. H. (1978). The input of gaseous and particulate sulfur to a forest ecosystem. Tellus 30, 546-551. Egunjobi, J . T. (1971). Ecosystem processes in a stand of U1e.u europaeus L. 11. The cycling of chemical elements in the ecosystem. J . Ecof. 59, 669-678. Epstein, E. (1972). Mineral Nutrition of Plants: Principles and Perspectives 412 pp. John Wiley and Sons, New York. Eriksson, Erik. (1955). Airborne salts and the chemical composition of river water. Tellus I. 243-250. Etherington, J . H. (1967). Studies of nutrient cycling and productivity in oligotrophic systems. 1. Soil potassium and windblown sea spray in South Wales dune grassland. J . Ecology 55, 743-752. Ewel, K. C., Gamble, J. F. and Lugo, A. E. (1975). Aspects ofmineral-nutrient cycling in a southern mixed-hardwood forest in north central Florida. In Mineral Cycling in Southeastern Ecosystems (Eds F. G. Howell, J . B. Gentry and M. H. Smith), pp. 700-714. USDA Natl. Tech. Info. Ctr., Springfield, Virginia. Fahey, T. J . (1979). Changes in nutrient content of snow water during outflow from Rocky Mountain coniferous forest. Oikos 32, 422428. Fairfax, J . A. W. and Lepp, N. W. (1975). Effect of simulated acid rain on cation loss from leaves. Nature (London) 255, 324-325. Falconer, R. E. and Falconer, P. D. (1979). Determination of cloud water acidity of a mountain observatory in the Adirondack mountains of New York State. Atmos- pheric Sciences Research Center, Publ. No. 741. State University of New York, Albany, New York. Falconer, P. D. and Kadlecek, J . A. (1980). Cloud chemistry and meteorological research at Whiteface Mountain: Summer 1979. Atmospheric Sciences Research Center, Publ. No. 748. State University of New York. Albany, New York. Feller, M. C. (1977). Nutrient movement through Western Hemlock-Western Red Cedar ecosystems in southwestern British Columbia. Ecology 58, 1269-1 283. Finn, J. T. (1976). Measures of ecosystem structure and function derived from analysis of flows. J . Theor. Biol. 56, 363-380. Finn, J. T. (1980). Flow analysis of models of the Hubbard Brook Ecosystem. Ecology 61, 562-571. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 111 Ford, E. D. and Deans, J. D. (1978). The effects of canopy structure on stemflow, throughfall and interception loss in a young Sitka spruce-stand. J . Appl. E d . 15, 905-917. Foster, N. W. and Gessel, S. P. ( 1 972). The natural addition ofnitrogen, potassium and calcium to a Pinus hanksianu Lamb. forest floor. Can. J . For. Ri~s . 2, 448455. Foster, N. W. (1974). Annual macroelement transfer from Pinus hunksiunn Lamb. forest to soil. Can. J . For. RFS. 4, 470-476. Galloway, J. N., Likens, G. E. and Edgerton, E. S. (1976). Hydrogen ion speciation in the acid precipitation of the northeastern United States. W n w , Air und Soil Pollutirin 6. 423433. Galloway, J. N. and Likens, G. E. (1978). The collection of precipitation for chemical analysis. Tellus 30, 7 1-82, Galloway, J . N. and Parker, G . G. (1979). Sulfur deposition in the eastern United States. In MAP3S Update: Progress Report for FY-I977 and FY-1978 (Proj. Dir. M. C. McCracken), pp. 124-134. Nat. Tech. Info. Ctr.. Springfield, Virginia. Galloway, J . N. and Parker, G. G . (1980). Difficulties in measuring wet and dry deposition on forest canopies and soil surfaces. 117 Effects of Acid Precipitation on Terrestrial Ecosystems (Eds T. C. Hutchinson and M. Havas). pp. 57-68. Plenum Press, New York. Gersper, D. L. and Hollowaychuck. N. (1970). Effects of stemflow water on a Miami soil under a beech tree I. morphological and physical properties. Soil Sci. Soc. Am. Proc. 34. 779-786. Gersper, D. L. and Hollowaychuck, N. (1971). Some effects of stemflow from forest canopy trees on chemical properties of soils. Ecology 52. 691-702. Golley, F. B., McGinnis, J. T., Clements, R. G., Child, G. 1. and Duever. M. J . (1975). Mineral Cycling in a Tropical Moist Ecosystem 248 pp. University of Georgia Press, Athens, Georgia. Gorham, E. (1953). Chemical studies on the soils and vegetation of water-logged habitats in the English Lake District. J . Ecology 41, 345-360. Gosz, J . R., Likens, G. E., Eaton. J . S. and Bormann, F. H. (1975). Leaching of nutrients from leaves of selected tree species of New Hampshire. I n Mineral Cycling in Southeastern Ecosystems (Eds F. G. Howell, J. B. Gentry and M. H. Smith), pp. 630-641. Natl. Tech. Info. Ctr.. Springfield, Virginia. Gosz, J . R., Likens, G. E. and Bormann, F. H . (1976). Organic matter and nutrient dynamics of the forest and forest floor in the Hubbard Brook Forest. Orculogiu 22. 305-320. Gran, G. ( 1952). Determination of the equivalence point in potentiometric titrations. Anal-yst 77, 661. Graustein, W. C. (1980). The effects of forest vegetation on chemical weathering and solute acquisition: a study of the Tesuque Watersheds near Santa Fe, New Mexico. PhD Dissertation. Department of Geology and Geophysics, Yale University. Graustein, W. C. and Armstrong, R. L. (1978). Measurement of dust input to a forested watershed using 87Sr/Sr ratios. Grol. Sric. Am. Abstracts with Progrums 10, 41 I . Gregory, P. H. (1961). The Microbiology of the Atmosphere 377 pp. Leonard Hill, London. Guha, M. M. and Mitchell, R. L. (1966). The trace and major element composition of the leaves of some deciduous trees. 11. Seasonal changes. Plunt Soil 24, 90 112. Hamman, H. F. (1956). Effects of spray residues and other contaminants on leaf analysis. Plunt Physiol. 31, 256-257. Hart, G. S. and Parent. D. R. (1974). Chemistry of throughfall under Douglas Fir and Rocky Mountain Juniper. A m . M i d . Nut. 92, 191 201. I12 G. G. PARKER Haughbotn, 0. ( 1973). Nedbrarunderserkelsor i Sarpsborgdistriktet og undersrakelser over virkninger av forsurende nedfall pA jordas kjemiske egenskaper, As-NLH 1973. 151 pp. (referenced in Horntvedt, 1975). Heinrichs, H. and Mayer, R. (1977). Distribution and cycling of major and trace elements in two Central European forest ecosystems. J . Env. Qual. 6, 402407. Helvey, J. D. and Patric, J. H. (1965). Canopy and litter interception of rainfall by hardwoods in the eastern United States. Water Resour. Res. 1, 193-206. Henderson, G. S., Harris, W. F., Todd, D. E., Jr. and Grizzard, T. (1977). Quality and chemistry of throughfall as influenced by forest type and season. J . Ecology 65, Herrera, R., Jordan, C. F., Klinge, H. and Medina, E. (1978). Amazon ecosystems. Their structure and functioning with particular emphasis on nutrients. Intercienciu Hesse, P. R. (1957). Sulfur and nitrogen changes in forest soils of East Africa. Plant Soil 9, 86-96. Hill. A. C. (1971). Vegetation: a sink for atmospheric pollutants. J . Air Pollut. Contr. Assoc. 21, 341-346. Hoffman, W. A,, Jr., Lindberg, S. E. and Turner, R. R. (1980a). Precipitation acidity: the role of the forest in acid exchange. J . Environ. Qual. 9, 95-100. Hoffman, W. A., Jr., Lindberg, S. E. and Turner, R. R. (1980b). Some observations of organic constituents in rain above and below a forest canopy. Environ. Sci. Technol. 14, 999 1002. Holmes. R. N. and Brinson, M. M. (1977). Phosphorus cycling in an alluvial Swamp Forest in the North Carolina coastal plain. ASB Bull. 24, 58. Holloway, P. J. (1971). Chemical and physical characteristics of leaf surfaces. In The Ecology of Leaf Surface Microorganisms (Eds T. F. Preece and T. H. Dickerson), pp. 39-55. Academic Press, London. Horntvedt, R. (1974). Nedbrarens fordeling og kjemiske innhold i skog. Intern rapport I R 2/74. pp. 25 37. SNSF-prosjektet, Oslo-As, Norway. Horntvedt. R. (1975). Kjemisk innhold i nedbrar under traer. et litteratursam- mendrag. Teknisk notat TN 18/75. SNSF-prosjektet, 1432 As-NLH, Norway. Horntvedt, R. and Joranger, E. (1974). Nedbrarens fordeling og kjemiske innhold under traer: Juli-november 1973. Teknisk notat T N 3/74. 29 pp. SNSF-prosjektet, Oslo-As, Norway. (referenced in Horntvedt, 1975). Horntvedt, R., Dollard, G . J . and Joranger. E. (1980). Effects of acid precipitation on soil and forest. 2. Atmosphere-vegetation interactions. In Ecological Impact of Acid Precipitation (Eds D. Drablras and A. Tollan) pp. 192-193. SNSF Project, Oslo-As. Hughes, M. K., Lepp, N. W. and Phipps, D. A. (1980). Aerial heavy metal pollution and terrestrial ecosystems. Adv. E d . Res. 11, 21 7--327. Hutchinson, I . and Roberts, M. C. (1981). Vertical variation in stemflow generation. J . Appl. E d . 18, 521-527. Ingham, G. (1950a). Effect of materials absorbed from the atmosphere in maintaining soil fertility. Soil Sci. 70, 205-212. Ingham, G . (1950b). Mineral content of air and rain and its importance to agriculture. J . Agric. Sci. 40, 50-6 1. Iwatsubo, G . and Tsutsumi, T. (1967). On the amount of plant nutrients supplied to the ground by rainwater in adjacent open plot and forest. Kyoto University For. Bull. 39, 110-120. Jacobsen, J. S. (1980). Experimental studies on the phytotoxicity of acidic pre- cipitation: the United States experience. In Effects of Acid Precipitation on 365-374. 3, 223-232. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 113 Terrestrial Ecosystems (Eds T. C. Hutchinson and M. Havas), pp. 141-150. Plenum Press, New York. Johnson, D. W. (1975). Processes of elemental transfer in some tropical, temperate, alpine and northern forest soils: factors influencing the availability and mobility of major leaching agents. PhD Thesis. University of Washington. Johnson, F. L. and Risser, P. G. (1974). Biomass, annual net primary productivity, and dynamics of six mineral elements in a post oak-blackjack oak forest. Eco1og.v 55, Johnson, D. W., Cole, D. W. and Gessel, S. P. (1 975). Processes of nutrient transfer in a tropical rain forest. Biotropica 7 , 208-2 15. Jordan, C. F. (1978). Stemflow and nutrient transfer in a tropical rain forest. Oikos31, 257-263. Jordan, C. F. and Kline, J. R. (1972). Mineral cycling: some basic concepts and their applications in a tropical rain forest. Ann. Rev. Ecol. Syst. 3, 33-50. Jordan, C. F., Kline. J. R. and Sasscer, D. S. (1972). Relative stability ofmineral cycles in forest ecosystems. Am. Nut. 106, 237-253. Jordan, C., Golley, F.. Hall, J. and Hall, J. (1980). Nutrient scavenging of rainfall by the canopy of an Amazonian rain forest. Biotropica 12, 61-66. Juang, F. H. T. and Johnson, N. M. (1967). Cycling of chlorine through a forested watershed in New England. J . Geosphys. Rex 72 , 5641-5647. Junge, C. E. (1963). Air Chemistry and Radioactivity 382 pp. Academic Press, New York. Junge, C. E. and Werby, R. T. (1958). The concentration of chloride, sodium, potassium, calcium and sulfate in rainwater over the United States. J . Meteorol. 15, 41 7425. Kaul, 0. N. and Billings, W. D. (1965). Cation content of stemflow in some forest trees in North Carolina. Indian For. 91, 367-370. Kazimirov, N. I. and Morozova, R. N. (1973). Biological Cycling of Matter in Spruce Forests of Karelia. Nauka Publ. House, Leningrad (reference in Cole and Rapp, 1981). Kellman, M. (1979). Soil enrichment by neotropical savanna trees J . Eco. 67, 565- 577. Kelly, J. M. ( I 979). Camp Branch and Cross Creek Experimental Watershed Projects: es and Ecological Characteristics. USEPA - 600/7-79453. Washington, DC. Kenworthy, J. B. (1970). Water and nutrient cyclingin a tropical forest. In The Water Relations of Malesian Forests, First Aberdeen-Hull Symposium on Malesian Ecology (Ed. J. R. Flenly), pp. 49-59. University of Hull, Dept. of Geography, Inst. for South-East Asian Biology, Univ. of Aberdeen. Hull, England. Khanna, P. K. and Ulrich, B. (1981). Changes in the chemistry of throughfall under stands of beech (Fagus s.y/vatica) and spruce (Picea abies) following the addition of fertilizers. Acta Oecol. Oecol. Plant 2, 155-164. Killingbeck, K. T. and Wali, M. K. (1978). Analysis of a North Dakota gallery forest: nutrients, trace elements and productivity relations. Oikos 30, 29-60. Kimmins, J. P. (1972). Relative contributions of leaching, litterfall and defoliation by Neodiprion sertifer (Hymenoptera) to the removal of cesium-1 34 from red pine. Oikos 23, 226-234. Kimmins. J. P. (1973). Some statistical aspects of sampling throughfall precipitation in nutrient cycling studies in British Columbia Coastal Forest. Ecology 54, 1008-1019. Kittredge, J. (1948). Forest Influences 394 pp. McGraw-Hill, New York. Knabe, W. and Gunther. K. H. (1976). Investigation on effects of the forest canopy on 1246-1 258. 114 G. G. PARKER acid sulfur precipitation in the Ruhr district, Germany. In Proc. 1st Intl. Symp. Acid Precip. and the Forest Ecosystem (Eds L. S. Dochinger and T. A. Seliga), pp. 895. USDA For. Ser. Gen. Tech. Rept. NE-23. Krause. P. D. (1977). Mineral cycling in aspen: the interactions of throughfall. litterfall, road salt and insect defoliation. Ph.D. dissertation. Department of Biology, University of New Mexico, Albuquerque, New Mexico. Lakhani, K. H. and Miller, H. G. (1980). Assessing the contribution ofcrown leaching to the element content of rainwater beneath trees. In Effects of Acid Precipitation on Terrestrial Ecosystems (Eds T. C. Hutchinson and M. Havas), pp. 151-172. Plenum Press, New York. Lang, G. E., Reiners, W. H. and Heier, R. K. (1976). Potential alteration of precipitation chemistry by epiphytic lichens. Oecologia 25, 229-241. Larson, T. E. and Hettick, 1 . (1956). Mineral composition of rain-water. T e l h 8. 19 1-201. Lausberg, T. (1935). Quantitative Untersuchungen uber die kutikulare Exkretion des Laubblattes. Jurh. wiss. Botunik 81, 769-806. Lemee, G. (1974). Recherches sur les ecosysttmes des reserves biologiques de la forCt de Fontainbleau 1V. -entrees delements mineraux par les precipitations et transfert au sol par le pluviolesivage. Oecol. Plant. 9, 187-200. Leonard, R. E. (1966). Mathematical theory of interception. In Forest Hydrology (Eds W. E. Sopper and H. W. Lull), pp. 131-136. Pergamon Press, New York. Leyton, L. (1958). The mineral requirements of forest plants. In Encyclopedia of Plant Physiology IV: Mineral Nutrition of Plants (Ed. W. Ruhland), pp. 1026-1039. Springer-Verlag, Berlin. Likens, G. E. and Bormann, F. H. (1974). Acid rain: a serious regional environmental problem. Science 184, 1176-1 179. Likens, G. E., Bormann, F. H., Pierce, R. W., Eaton, J . S. and Johnson, N. M. (1977). Biogeochemistry of a Forested Ecosystem, 146 pp. Springer-Verlag, New York. Lindberg, S. E. and Harr is , R. C. (1981). The role of atmospheric deposition in an eastern deciduous forest. Water, Air Soil Pollut. 16, 13-3 1. Lindberg, S. E., Harr is , R. C., Turner, R. R., Shriner, D. S., Huff, D. D. (1979). Mechanisms and Rates of Atmospheric Deposition of Selected Trace Elements and SulFdte to a Deciduous Forest Watershed. Oak Ridge Natl Lab. Environmental Sciences Division, Publ. No. 1299. ORNL/TM-6674. Little, P. (1973). A study of heavy metal contamination of leaf surfaces. Environ. Pollut. 5, 159-172. Little, P. (1977). Deposition of 2.75,5.0 and 8.5 pm particles to plant and soil surfaces. Environ. Pollut. 12, 293-305. Little, P. and Wiffin, R. D. (1977). Emission and deposition of petrol exhaust Pb. I . deposition of exhaust Pb to plant and soil surfaces. Atmos. Env. 11, 347447. Lunt, H. A. (1934). Distribution of soil moisture under isolated forest trees. J . Agric. Res. 49, 695-703. Madgwick, H. A. I. and Ovington, J. D. (1959). The chemical composition of precipitation in adjacent forest and open plots. F0restr.p 32, 14-22. Mahendrappa, M. K. (1974). Chemical composition of stemflow from some Eastern Canadian tree species. Cun. J . For. Rtv . 4, 1-7. Mahendrappa, M. K. and Ogden, E. 0. (1973). Effects of fertilization of a Black Spruce stand on nitrogen contents of stemflow, throughfall and litterfall. Can. J. For. Res. 3, 54-60. Malcolm, R. L. and McCracken, R. J. (1968). Canopy drip: a source of mobile soil organic matter for mobilization of iron and aluminium. Soil. Sci. Soc. Am. Proc. 32, 8 34-838. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 1 I5 Manokaran, N. (1978). Nutrient concentration in Precipitation, Throughfall and Stemflow in a Lowland tropical Rainforest in Peninsular Malaysia. Malay. Nut. J . 30, 423432. Mayer, R. and Ulrich, B. (1972). Conclusions on the filtering action of forests from ecosystems analysis. 0cw)l. Plant. 9, 157-1 68. McColl, J. G. (1970). Properties of some natural waters in a tropical wet forest in Costa Rica. Biosciencc 10, 1096 ~ 1 100. McColl, J. G. and Bush, P. S. (1978). Precipitation and throughfall chemistry in the San Francisco Bay area. J . Enriron. Qua/. 7 , 352-357. Mecklenburg, H. A,, Tukey, H. B.. Jr. and Morgan. J. V. (1966). A mechanism for leaching of calcium from foliage. Plant Physiol. 41, 610-613. Miller, H. G. (1979). The nutrient budgets of even-aged forests. I n The Ecology of Even-Aged Forest Plantations (Eds E. D. Ford, D. C. Malcolm and J. Atterson), pp. 221-256. Institute of Terrestrial Ecology. Cambridge. United Kingdom. Miller, H. G. and Miller, J. D. (1980). Collection and retention of atmospheric pollutants by vegetation. I n Ecological Impact of Acid Precipitation (Eds D. Drables and A. Tollan) pp. 33~-40. SNSF Project, Oslo-As. Miller, H . G., Cooper, J. M. and Miller, J. D. (1976). Effect of nitrogen supply on nutrients in litterfall and crown leaching in a stand of Corsican Pine. J . Appl. E d . 13, 233-248. Miller, J. M.. Galloway, J. N. and Likens, G. E. (1978). Origin ofair masses producing acid precipitation in Ithaca, New York. Geophys. Rrs. Lett. 5, 757-760. Miller, R. B. (1963). Plant nutrients in Hard Beech 111: the cycle of nutrients. N . Z . J . Sci. 6, 388413 . Mina, V. N. (1965). Leaching of certain substances from woody plants and its importance in the biological cycle. Soviet Soil. Sci. 1965, 605-617. Mina, V. N. (1967). Influence of stemflow on Soil. Sol& Soil Sci. 1967, 1321-1329. Mitchell, H. L. (1936). Trends in the nitrogen phosphorus, potassium, and calcium content of the leaves of some forest trees during the growing season. Black Rock Forest Puprr 1, 30-44. Mudd, J. B. (1975). Sulfur dioxide. I n Responses of Plants to Air Pollution(Eds J. B. Mudd and T. T. Kozlowski), pp. 9-23, Academic Press, New York. New York Times. (1978). Atlas ofthe World 268 pp. Times Books Limited, London. Nicholson, 1. A,, Cape, N., Fowler, D., Kinnaird, J. W. and Paterson. I. S. (1980). Erects of a Scots pine (Pinus liv.srr.is L. ) canopy on the chemical composition and deposition pattern of precip tion. Iri Ecological Impact of Acid Precipitation (Eds D. Drablras and A. Tollan). pp. 192 193. SNSF Project. Oslo-As. Nieboer, E., Richardson, D. H. S. and Tomassini, F. D. (1978). Mineral uptake and release by lichens: an overview The Bryologist 81, 226--246. Nihlgard, B. (1970). Precipitation, its chemical composition and effects on soil water in a Beech and Spruce forest in South Sweden. Oikos 21, 208-217. Nortcliff, S. and Thorncs, J. B. (1978). Water and cation movement in a tropical rainforest environment. 1 . Objectives, experimental design and preliminary results. Acta Amazonica 8, 245-258. Nye, P. H . (1961). Organic matter and nutrient cycles under moist tropical forest. Plant Odum, E. P. (1969). The strategy of ecosystem development. Science 144, 262-270. Odum, E. P. (1971). The Fundamentals of Ecology 574 pp. W. B. Saunders Co.. Philadelphia, Pennsylvania. OHare, P. J. (1967). The leaching of nutrients by rainwater from forest trees a preliminary study. i n Forest Fertilization pp. 122 ~ 130. Proc. of the Fifth Colloq. of the Intl. Potash Institute, Jyvaskylii. Helsinki, Finland. Soil 13, 333-346. 116 G. G. PARKER Ovington, J. D. (1959). The circulation of minerals in plantations of Pinuss.vlvestris L. Ann. Bof. 23, 229-239. Ovington, J. D. (1962). Quantitative ecology and the woodland ecosystem concept. Adv. E d . RES. 1 , 103-192. Paivanen, J . (1974). Nutrient removal from Scots Pine on drained peatland by rain. Arta For. Fenn. 61, 5- 19. Parent, D. R. (1972). The influence of atmospheric dust and foliar leachates on the chemical quality of throughfall in northern Utah. M.S. Thesis. Logan State University, Logan, Utah. Parker, G . G., Lindberg, S. E. and Kelly, J. M. (1980). Canopy-atmosphere interactions of sulfur in the southeastern United States. In Atmospheric Sulfur Deposition (Eds D. S. Shriner, C. R. Richmond and S. E. Lindberg), pp. 477493. Ann Arbor Science, Ann Arbor. Michigan. Parker, G. G. (1982). Quality. variability and sources of summertime throughfall in two Virginia Piedmont forests. Thesis. University of Virginia, Charlottesville, Virginia. Patterson, D. T. (1975). Nutrient return in the stemflow and throughfall of individual trees in the Piedmont. I n Mineral Cycling in Southeastern Ecosystems (Eds F. G. Howell, J. B. Gentry and W. H. Smith), pp. 800-812. Natl. Tech. Info. Ctr., Springfield, Virginia. Pike, L. H. (1978). The importance of epiphytic lichens in mineral cycling. The Bryologist 81, 247-257. Potts, M. J. (1978). The pattern ofdeposition of air-borne salt of marine origin under a forest canopy. Plant Soil 50, 233-236. Pozdnyakov, L. K. ( 1956). The role of the rain penetrating beneath the forest canopy in the process of the exchange of materials between forest and soil. Dokl. Akad. Nauk. 107, 753-756. (English Abstract, No. 1820, in Soil Fert. Newd . 19, 379.) Prebble, R. E. and Stirk, G. B. ( 1 980). Throughfall and stemflow on silverleaf ironbark (Eucal-vptus melanophloia) trees. Aust. J . Ecol. 5 , 41 9 4 2 7 . Rapp, M. (1969). Apport delements mineraux au sol par les eaux de pluviolessivage sous des peuplements de Quercus i1e.u L., Quercus lanuginosa Lamk. et Pinus Halapcwsis Mill. Oc~cologia 4, 7 1-92. Rapp, M. (1973). Le cycle biogeochimique du soufre dans une fortt de Quercus ilex L. du sud de la France. Oecol. Plant. 8, 325-334. Rapp. M. (1974). Le cycle biogeochimique dans un bois de pins dAlep. In Ecologie Forestier (Ed. P. Persson), pp. 75-96. Gauthier-Villars, Paris. Rasmussen, L. and Johnsen, I . (1976). Uptake of minerals particularly metals, by epiphytic Hypnum cupressiforrne. Oikos 21. 483487. Raybould, C. C., Unsworth, M. H . and Gregory, P. J . (1977). Sources of sulphur in rain collected below a wheat canopy. Nature 267, 14-16. Reiners, W. A. (1972). Nutrient content of canopy throughfall in three Minnesota forests. Oikos 23, 14-20. Remezov, N. P. (1961). Decomposition of forest litter and the cycte of elements in an oak forest. Soviet Soil Sci. 1961, 703-71 I . Richter, A. and Granat, L. (1978). Pine forest canopy throughfall measurements. Report AL-43, Department of Meteorology, University of Stockholm, Sweden. Rodin, L. E. and Bazilevich, N. I . (1967). Production and Mineral Cycling in Terrestrial Vegetation (Transl. Ed. G. E. Goff), 288 pp. Oliver and Boyd, Edinburgh. Rolfe, G. L.. Akhtar. M. A. and Arnold. L. E. (1978). Nutrient distribution and flux in a mature oak-hickory forest. For. Sci. 24, 122-130. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 1 I7 Russell, R. S. (1965). Interception and retention of airborne material on plants-an introductory review. Health Phys. 11, 1305-1315. Schlesinger, W. H. (1978). Community structure, dynamics and nutrient cycling in the Okefenokee Cypress Swamp Forest. Ecol. Monogr. 48, 43-65. Schlesinger, W. H. and Reiners, W. A. (1974). Deposition of water and cations on artificial collectors in Fir krummholz of New England mountains. Ecology 55, Schlesinger, W. H. and Marks, P. L. (1977). Mineral cycling and the niche of Spanish Moss. Am. J . Bot. 64, 1254-1262. Sehmel, G. A. (1980). Model predictions and summary of dry deposition velocity date. In Atmospheric Sulfur Deposition (Eds D. S. Shriner, C. R Richmond and S. E. Lindberg), pp. 223-235. Ann Arbor Science, Ann Arbor, Michigan. Slinn, W. G. N. (1977). Some approximations for the wet and dry removal of particles and gases from the atmosphere. Water, Air and Soil Pollut. 7 , 513-543. Sokolov, A. A. (1972). Chemical composition of rainfall passed through the birch and spiuce canopies. Lesovedenie 3, 103-105. (referenced in Horntvedt, 1975). Sollins, P. and Drewry, G. (1970). Electrical conductivity and flow rate of water through the forest canopy. In A Tropical Rain Forest (Eds H. T. Odum and R. F. Pigeon) pp. H 137-H 153. USAEC, Washington, DC. Sollins, P., Crier, C. C., McCorison, F. M., Cromack, K., Jr., Fogel, R. and Fredriksen, R. L. (1980). The internal nutrient cycles of an old growth Douglas-Fir stand in western Oregon. Ecol. Monogr. 50, 261-285. Spurr, S. H. and Barnes, B. V. (1973). For. Ecol. 571 pp. The Ronald Press Co., New York. Stark, N. (1973). Nutrient Cycling in a Jeffrey Pine Forest Ecosystem. Microbiology, University of Montana Press, Missoula, Mont. 389 pp. Stenlid, G. (1958). Salt losses and redistribution of salts in higher plants. In Encyclopedia of Plant Physiology Vol. IV, Mineral Nutrition of Plants (Ed. W. Ruhland), pp. 61 5 4 3 7 . Springer-Verlag, Berlin. Stinner, B. R. (1981). Nutrient budgets for conventional, no-tillage and old-field systems on the Georgia piedmont. PhD dissertation, University of Georgia, Athens, Georgia. Stone, E. L. and Kszystyniak, R. (1977). Conservation of potassium in the Pinus resinosa ecosystem. Science 198, 192-194. Strahler, A. N. (1973). Introduction to Physical Geography 468 pp. John Wiley and Sons, New York. Swank, W. T. and Henderson, G. S. (1976). Atmospheric input of some cations and anions to forest systems in North Carolina and Tennessee. Water Resour. Res. 12, Switzer, G. L. and Nelson, L. E. (1972). Nutrient accumulation and cycling in Loblolly pine (Pinus taeda L.) plantation ecosystems: the first twenty years. SoilSci. SOC. Am. Szabo, M. and Csortos, C. S . (1975). A study of the nutrient content of canopy throughfall in an oak forest (Quecetum petraeacerris) measured for one year. Acta Bot. Acad. Sci. Hung. 21, 419432. Szabo, M. (1977). Nutrient content of throughfall and stemflow water in an oak-forest (Quercetum petraeacerris) Ecosystem. Acta Agron. Acad. Sci. Hung. 26, 24 1-258. Tamm, C. 0. (1950). Growth and plant nutrient concentration in Hylocomium proliferum (L.) Lindb. in relation to tree canopy. Oikos 2, 6 0 4 4 . Tamm, C. 0. (1951). Removal of plant nutrients from tree crowns by rain. Physiol. Plant. 4, 184-188. 378-386. 541-556. J . 36, 143-147. I18 G. G. PARKER Tamm. C. 0. (1958). The atmosphere. In The Encyclopedia of Plant Physiology, Vol. IV , Mineral Nutrition of Plants (Ed. W. Ruhland), pp. 233-242. Springer- Verlag. Berlin. Tamm, C. 0. and Troedsson. T. (1955). An example of the amounts of plant nutrients supplied to the ground in road dust. Oikos 6. 61-70. Tamm. C. 0. and Cowling, E. B. (1977). Acidic precipitation and forest vegetation. Wutcv, A i r Soil Pollut. 7. 503-51 1. Tarrant. R. F., Lu, K. C., Bollen, W. B. and Chen, C. S. (1968). Nutrient cycling by throughfall and stemflow precipitation in three coastal Oregon forest types. USDA Forest Service Research Paper, PNW-54, 7 pp. Thomas, M. D., Hendricks, R. M. and Hill, G. R. (1950). Sulfur metabolism of plants. I d . Eng. Cliiwi. 42, 223 1-2235. Thomas, W. A. ( 1967). Dye and calcium ascent in Dogwood trees. Plant Phj3siol. 42, Thomas, W. A. (1969). Accumulation and cycling of calcium by dogwood trees. Ecol. Monogr. 39, 101-120. Tiedemann, A. R., Helvey, J. D. and Anderson, T. D. (1980). Effects of chemical defoliation of an Ahirsgranciis habitat on amounts and chemistry of throughfall and stem flow. J . Env. Qua/. 9, 320-328. Torrenueva, A. L. (1975). Variation in mineral flux to the forest floors of a Pine and a Hardwood stand in the Georgia Piedmont. Ph.D. Dissertation. University of Georgia. Athens, Georgia. Tsutsumi, T. (1971). Accumulation and circulation of nutrient elements in forest ecosystems. In Productivity of Forest Ecosystems (Ecology and Conservation 4) (Ed. P. Duvigneaud), pp. 543-552. UNESCO, Paris. Tsutsumi, T. (1977). Storage and circulation of mineral nutrients. In Primary Productivity of Japanese Forests, IBP Synthesis Vol. 16. (Eds T. Shidei and T. Kira), pp. 140- 168. University of Tokyo Press, Tokyo, Japan. Tukey, H. B., Jr. (1970a). The leaching of substances from plants. Annu. Rev. Plant Ph?,siol. 21, 305-329. Tukey, H. B.. Jr. (1970b). Leaching of metabolites from foliage and its implication in the tropical rain forest. In A Tropical Rain Forest (Eds H . T. Odum and R. F. Pigeon), pp. H 155-H 160. USAEC, Washington, D.C. Tukey. H. B., Jr. (1980). Some effects of rain and mist on plants, with implications for acid precipitation. h i Effects of Acid Precipitation on Terrestrial Ecosystems (Eds T. C. Hutchinson and M. Havas), pp. 141-150. Plenum Press, New York. Tukey. H. B., Jr. and Morgan. J. V. (1963). Injury to foliage and its effect upon the leaching of nutrients from above-ground plant parts. Physiol. Plant. 16, 557-564. Tukey, H. B., Jr.. Tukey, H. B. and Wittwer, S. H. (1958). Loss of nutrients by foliar leaching as determined by radiosotopes. Proc. Am. Soc. Hort . Sci. 71, 496-506. Turner, J . and Singer, M. J . (1976). Nutrient distribution and cycling in a subalpine coniferous forest ecosystem. J . Appl. Ecol. 13, 295-301. Turner. J., Cole. D. W. and Gessel, S. P. (1976). Mineral nutrient accumulation and cycling in a stand of Red Alder (Alnus ruhra). J . Ecology 64, 965-974. Turner. J . M., Lambert. M. J. and Gessel, S. P. (1977). Use of foliage sulphate concentrations to predict responses to urea application by Douglas Fir. Can. J . For. Rrs. 7, 476 480. Turner, J. and Lambert, M. J . (1980). Sulfur nutrition of forests. In Atmospheric Sulfur Deposition (Eds D. S. Shriner, C. R. Richmond and S. E. Lindberg) pp. 321 333. Ann Arbor Science, Ann Arbor, Michigan. Tuvey. N. D. (1974). Water in the nutrient cycle of a Papuan rain forest. Nature 251, 414 415. 1800- 1802. THROUGHFALL, STEMFLOW IN FOREST NUTRITION 119 Ugolini, F. C., Minden, R., Dawson, H. and Zachara, J. (1977). An example of soil processes in the Ahies amahilis zone of Central Cascades, Washington. Soil Sci. 124, 291-302. Ulrich, B. (1971). Investigations on cycling of bioelements in forests of Central Europe (preliminary results of the Solling project). In Productivity of Forest Ecosystems (Ecology and Conservation 4) (Ed. P. Duvigneaud) pp. 501-507. Proc. Brussels Symp. UNESCO, 1969. Ulrich, von B., Mayer, R., Khanna, P. K. and Prenzel. J. (1978). Ausfilterung von Schwefelverbindungen aus der Luft durch einem Buchenbestand. Z. fflanzener- naehr. Bodenkde. 141, 329-335. Verry, F. S. and Timmons, D. R. (1977). Precipitation nutrients in the open and under two forests in Minnesota. Can. J . For. Res. 7, 112-1 19. Villecourt, P. and Roose, E. (1978). Content of N and major elements in rainfall, throughfall and drainage in the Lamto savanna, Ivory Coast. Revue JEcologie et de Biologie du Sol 15, 1-20. (abstract no. 871 in Forestry Abstracts 41). Vitousek, P. M. and Reiners, W. A. (1975). Ecosystem succession and nutrient retention: a hypotheses. Bioscience 25, 376-38 1. Vogelman, H . W., Siccama, T., Leedy, D. and Ovitt, D. C. (1968). Precipitation from fog moisture in the Green Mountains of Vermont. Ecology 49, 1205-1207. Voigt, G. K . (1960a). Distribution of rainfall under forest stands. For. Sci. 6, 210. Voigt, G. K. (1960b). Alteration of the composition of rainwater by trees. Am. Mid. Nut. 63, 321-326. Voight, G. K. and Zwolinski, M. J. (1964). Absorption of stemflow by bark of young red and white pines. For. Sci. 10, 277-282. Waller, H. D. and Olson, J. S. (1967). Prompt transfers of Cesium-137 to the soils of a tagged Liriodendron forest. Ecology 48, 15-25. Weaver, G. T. and Brown, R. (1979). Nutrient inputs and transfers in southern Illinois forested watersheds. In Impact of Intensive Harvesting on Forest Nutrient Cycling p. 419. State University of New York, College of Environmental Science and Forestry. Syracuse, New York. Wedding, J. B., Carlson, R. W., Stukel, J . J. and Bazzaz, F. A . (1976). Aerosol deposition onto plant leaves. Environ. Sci. Technol. 9, 151-153. Wells, C. G. and Jorgenson, J . R. (1974). Nutrient cycling in Loblolly Pine plantations. In Proc. 4th North Am. Forest Soils Conf. pp. 137-158. Wells, C. G., Nicholas, A. K. and Buol, S. W. (1975). Some effects of fertilization on mineral cycling in Loblolly Pine. In Mineral Cycling in Southeastern Ecosystems (Eds F. G. Howell, J. B. Gentry and M. H. Smith), pp. 754-764. Natl. Tech. Info. Ctr., Springfield, Virginia. Wells, C. A., Whigham, D. and Lieth, H. (1972). Investigations ofmineral cycling in an upland piedmont forest. J . Elisha Mitchell Sci. Soc. 88, 66-78. Westman, W. E. ( 1978). Inputs and cycling of mineral nutrients in a coastal subtropical eucalypt forest. J . Ecology 66, 5 13-53 1. White, E. J. and Turner, F. (1970). A method of estimating income of nutrients in catch of airborne particles by a woodland canopy. J . Appl . Ecol. 7 . 4 4 4 6 1 . Whitehead, H. L. and Feth, J. H. (1964). Chemical composition ofrain, dry fallout and bulk precipitation at Menlo Park, California, 1957-1959. J . Gec)pph~s. Res. 69, 3319-3333. Whittaker, R. H. and Woodwell, G. M. (1967). Surface area relations of woody plants and forest communities. Am. J . Bot. 54, 931-939. Whittaker, R. H. and Feeny, P. P. (1971). Allelochemics: chemical interactions between species. Science 171, 757-770. I20 G. G. PARKER Will, G. M. (1955). Removal of mineral nutrients from tree canopies by rain. Nature Will. G. M. ( 1959). Nutrient returns in litter and rainfall under some exotic conifers. N . Wilm. H. G. (1946). The design and analyses of methods for sampling micro-climate (actors. J . A w . Stcrt. Assoc.. 41, 221-223. Witherspoon, J. P.. Jr. (1964). Cycling of Cesium-I34 in White Oak trees. Ecol. Moriogr. 34. 403420. Witherspoon. J . P., Jr. (1972). Field studies of fallout retention by plants. //? "Survival of Food Crops and Livestock in the Event of Nuclear War" pp. 396404. USAEC Symposium Series CONF. 700909. USAEC. Washington, D.C. Witkamp. M. (1970). Mineral retention by epiphyllic organisms. /ti "A Tropical Rain Forest" (Eds H. T. Odum. R. F. Pigeon), pp. H177-Hl80. USAEC, Washington, D.C. Wittwcr. S. H . and Teubner. F. G. (1959). Foliar absorption of mineral nutrients. A / i / i i ~ . RPI.. Plrrrit PhI..siol. 10. 13-32. Wittwcr. S. H. and Bukovac. M. J. (1969). The uptake of nutrients through leaf surfaces. 111 "Handbuck der Pflanzenernahrung und Dungung. Band I : Pflanzener- nahrung" (Eds K. Scharrer and H. Linser), pp. 235-261. Springer-Verlag, Vienna. Wolaver. T. M. and Lieth. H. (1972). "US Precipitation Chemistry-Theory and Quantitative Models" 75 pp. EPA. Research Triangle Park, North Carolina. Wood, T. and Bormann. F. H. (1975). Increases in foliar leaching caused by acidification of an artificial mist. A/iihio 1975. 160- 162. Wyatt-Smith. J . ( 1958). "Rain" from a Shor~a rlc1.sj~plij~llo tree. Mnloj. Forester 21, '02 203. Yadav. A. K. and Mishra. G. P. (1980). Observations on the colors of stemflow and throughfall waters in forest trees. Grwhim (Jocllip~cr) 7, 148-150. Yawney. H. W.. Mustanoja, K . J.. Leonard, R. E. and Leaf, A. L. (1970). The contribution of throughfall and stemflow to nutrient element cycling in Red Pine Pla ti ta t ions. Agrori. A hstr. 1970. 164. Yawney. H . W. and Leaf. A. L. (1971). Nutrient release from Red Pine crowns under artificial rain. A,qrci/i. Ah.str. 1971. 121. Yawney. H. W.. Leaf. A. L. and Leonard, R. E. (1978). Nutrient content ofthroughfall and stemflow in fertilized and irrigated Pi/iif,s r(>siriosrr Ait. stands. Plant Soil 50, 433-445. Zinke, P. J. ( 1962). The pattern of influence of individual forest trees on soil properties. Zinke. P. J . (1966). Forest interception studies in the United States. hi "Forest 176, 1180. Z . J . Agric. Rcs. 2. 7 19 734. E1,ologl. 43. I30 133. Hydrology" (Eds W. E. Sopper and H. W. Lull). pp. 137-161. New York. APPENDIX Nutrient depositions (Kg ha ' ) in throughfall and incident precipitation in world forests (an extension of Horntvedt, 1975). Studies are listed alphabetically, by author. Horizontal lines separate throughkill (above) from incident precipitation (below). Forest type H,O Period, Reference, and location NH,N NO,N N, P, K Ca Mg Na CI S0,S (mm) (months) notes Washington ~- ~ Douglas fir 4.65 2.92 21.83 6.50 3.40 - 12 Abee and Incident 1.30 0.23 0.1 1 2.09 1.27 239 12 Lavender. 1972 Spruce 0.2 0.2 0.07 10.6 5.5 2.2 12.1 25.6 9.9 270 6 Abrahamson ef d. , Pine 0.4 0.6 0.02 7.7 4.9 4.7 16.2 29.0 9.5 370 6 1976 Birch 0.6 0.8 0.15 7.0 3.4 1.8 8.8 17.0 6.5 400 6 Incident 1.4 1.6 0.02 0.6 0.9 0.7 7.0 9.9 4.9 470 6 Spruce 1 . I 0.2 0.25 4.5 0.7 0.7 4.0 7.4 1 . 1 70 4 Abrahamson er u / . . Pine 0.6 0.2 0.21 4.1 1.0 0.8 5.4 8.9 1 . 1 100 4 1976 Birch 0.3 0.1 0.14 2.5 0.6 0.3 1.2 2.1 0.5 90 4 Incident 0.2 0.2 0.09 0.6 0.3 0.2 1.0 1.7 0.9 150 4 Holly 44.4 27.3 33.0 240.0 - Southern Norway Arctic Norway Coastal New York - I2 Art cr u/.. 1972 - - Incident 6.2 8.6 11.5 97.0 - 12 Denmark Ftrgu.~ ,\y/iw i c w Mull soil 1.58 31.8 23.6 11.7 45.5 - 12 Astrup and - 1.30 11.9 14.2 4.5 17.3 - I2 Bulow-Olsen. 1972 Mor Soil 2.18 33.7 17.6 7.5 34.9 - 12' Incident 2.20 15.1 18.0 5.3 18.9 - 12 - - New Zealand Eucalyptus 13.38 8.01 7.29 25.43 282 12 Attiwill. 1966 Incident 2.01 2.74 5.36 16.81 387 12 - Nutrient depositions c o i i r i i i i w d Forest type and location NH,N NO,N N, P, K Ca Mg Na Eastern France Oak Incident En g I a n d Scots pine Incident Northern Alberta Spruce-upwind of power plant Spruce-downwind of power plant Incident Ivory Coast Plateau forest Valley forest Incident South Carolina Hardwoods White pine Incident Incident Incident lncident Incident Oregon Red alder Central Sweden Scots pine Scots pine Incident 0.59 0.88 0.90 0.92 1.01 1.26 1.71 .62 26 0.3 21.4 12.3 20-30 0.04 5-10 2-3 79.5 2.2 65 39.5 41 81.0 9.8 174.5 46.5 48.5 25 1.6 5.8 24 4.3 30.5 8.1 3.1 8.6 4.7 4.2 0.8 6.3 30.5 6.3 2.0 7.2 4.9 4.4 0.8 6.8 7.56 I .49 - CI s0,s -~ 19.6 5.8 - 0.83 0.33 8.58 2.41 - - 2.15 2.36 0.63 1.31 2.13 0.51 1.28 0.29 0.78 1.24 1.92 1.60 0.41 1.01 2.59 0.51 0.76 0.14 0.57 1.38 ~ I590 I640 I800 I380 1328 I760 - 1704 - - Period Reference. (months) notes ~~ I' 1 - 7 5 5 3 3 3 3 1'" I 2" I Z h I2 12 I' 12 12' I 2 I' I 2 12 12 Aussenac CI ( I / . . 1972 Bnche. 1977 Baker cr "/.. 1977 Bern hard- Renversat. 197.5 Best and Monk. 1975 Bollen cr ( I / . . 1968 Bringmark. 1980 North Carolina Swamp forest Coastal England Oak woodland Incident 9.2 91 I 5.6 I I05 - - 12' I 2 I" 12 Brinson C I ul.. 1980 10.3 1.6 12.0 15.3 7.6 5.8 0.5 3.0 4.8 1.4 Brown. 1974 3.3 0.40 28.5 19.3 0.6 5.9 0.23 3.2 6.9 5.4 - 1240 44 5 403 395 269 627 780 857 - 1359 1405 1617 - Incident Northwestern Spain Qiicrciis rohirr Pinus pitluster EuculJpru \ glob11 \ Ulex europueus Incident Querus rohur Pinus rudiutu Calve de Anta t't u/.. 1979 2.9 10.6 11.5 3.3 22.1 14.3 3.8 12.2 13.8 3.2 29.5 10.3 5.2 10.5 11.2 3.2 25.1 16.1 2.6 8.6 8.4 1.4 25.5 7.7 8.1 3.8 7.2 1.3 25.6 9.1 5.1 60.8 20.5 11.3 42.7 35.8 33.9 15.2 76.1 20.5 13.9 57.6 69.8 41.6 - - - - - - - - - 6 12 I 2 12 I 2 12 12'4 12 Incident Oakwood Incident Oakwood 1- Puerto Rico Incident En g I a n d Tropical evergreens Washington Carlisle CI [ I / . . 1966b Carlisle P I ( I / . , I967 8.82 1.31 28.14 17.18 9.36 55.55 9.54 0.43 2.96 7.30 4.63 35.34 9.31 0.92 29.74 20.98 13.23 89.36 8.71 0.28 2.84 6.72 6.10 50.76 Clements and Colon, 1975 Cole r t ul., 1967 - - 82.8 27.0 49.1 258.3 8.9 6.4 19.8 152.4 - 2.8 0.4 13.1 7.4 DOU& fir Incident 4.5' - 1.1 t 0.8 2.8 Washington Douglas fir Incident Paper birch Red pine Incident Northern Minnesota 12" Cole and Johnson. 1977 Comerford and White. 1977 15.4 1100 12.5 1400 299.4 268.2 3.59 0.93 11.80 9.06 2.00 4.37 0.13 5.45 7.30 1.45 2.85 0.31 1.92 6.03 0.82 4 4 4 Nutrient depositions conririuetl Forest type H,O Period and location NH,N NO,N N , P, K Ca Mg Na CI S0,S (mm) (months) ~ Belgium Quercus Fugus Incident QuercuS- Curpinus incident Belgium Oak-hazel forest Incident New Hampshire Northern hardwoods lExz Maritime New Zealand U1e.x ruropucw -t Northern Florida Mixed hardwoods Incident Jack pine Incident Jack pine Ontario lncident -1 Panama Tropical moist forest New Mexico Aspen riEEi3 16.8 7.1 5 19 24 1 1 4 15 24 1 1 7 - 13 4 15 5.8 11.71 0.73 30.45 7.62 2.17 0.66 6.97 26.05 478 1.79 0.05 0.37 0.89 0.17 0.35 2.53 5.07 556 8 0.34 56 14 13 70 145 7.4 0.22 10 10.7 5.3 60 100 3.2 24.7 29.7 8.1 0.9 8.2 19.3 2.9 5.25 0.13 11.6 7.14 0.8 7.9 0.10 4.0 5.6 1.25 2.7 5-7 5.8 4.1 2.6 3.5 -- 565 1519 - I095 I I75 - 758 953 62 I - - - - - 0.95 77.51 58.13 15.11 37.20 1.46 14.36 44.25 7.36 58.97 - 3.53 1.37 0.41 0.10 0.47 0.98 129.0 0.69 0.69 0.13 0.10 0.59 1.21 156.6 12' 12 12' I2 12' 12 5' 5 12 I2 12"'" 12 12"" 12 5.5" 5.5 I21 I21 3 3 Reference. notes Duvigneaud and Denaeyer-DeSmet, 1970 Duvigneaud and Denaeyer-DeSmet. 1975 Eaton et ul.. 1973 Egunjobi, 1971 Ewe1 et ul.. 1975 Foster. 1974 Foster and Gessel. 1972 Golley et ul.. 1975 Graustein. 1980 Aspen Incident Spruce fir Spruce fir Incident Conifers near factory Incident Conifer Incident Conifer Incident Beech Spruce Incident Tennessee Pirius Liriodendrori Quercus prinu.\ Quercus Corytr Incident Southern Norway Central Germany Incident Kikuyu Kenya Wattle Eucalyptus Incident Birch Spruce Pine Incident Southern Norway 0.3 1 0.79 0.08 0.72 - 0.4 0.7 0.1 0.3 0.3 0.5 1.5 1.3 3.23 2.23 0.36 0.24 1.36 1.64 282.0 0.46 1.2 0.10 0.24 0.9 1.77 308.0 5.2 4.43 0.79 0.74 2.7 3.37 126.0 0.69 0.69 0.13 0.10 0.59 1.21 156.6 ~. ~ 6.0 5.2 0.82 0.62 3.7 4.21 215.0 0.40 0.93 0.10 0.18 0.9 1.59 270.0 42.0 48.1 10.1 39.1 111.2 540 8.4 16.7 2.6 14.7 32.3 770 83.8 32.6 11.2 41.2 19.6 5.4 2.0 21.8 69.1 540 17.7 770 30.8 17.1 6.2 10.9 21.1 540 7.8 5.4 2.0 7.4 10.0 770 23.8 0.30 15.2 26.7 4.9 20.9 47.6 864 28.3 0.70 20.6 41.3 6.4 23.7 22.6 0.80 4.1 12.8 4.2 10.7 10.6 0.37 19.65 26.25 3.8 11.05 0.37 18.35 21.65 3.4 10.6 0.37 17.65 21.25 2.9 11.45 0.57 23.85 23.85 3.7 7.75 0.07 0.95 8.85 1.0 80.0 764 24.1 1060 1302 131 1 I294 1310 1528 - 1.14 .2 762 0.99-2.9 762 0.99-1.5 762 0.2 6.2 3.0 1.5 7.8 13.8 5.7 343 0.1 7.4 3.9 1.6 9.8 19.5 8.5 249 0.0 6.8 3.7 1.9 12.3 21.9 7.8 327 0.0 0.5 0.5 0.5 4.3 6.6 4.4 388 12 I2 12 12 I2 12 12' I2 12 12 12 12 12 12" 121 I21 1 2J 4.5 4.5 4.5 4.5 Haughbotn. 1973 Haughbotn. 1973 Haughbotn, 1973 Heinrichs and Mayer, 1977 Henderson p r d.. 1977 Hesse. 1957 Horntvedt and Joranger, 1974 Nutrient depositions coritiriired Forest type and location NH,N NO,N H,O N, P, K Ca Mg Na CI S0,S ( m m ) South Africa Unknown canopy 32.1 1.2 Incident 10.7 1.2 type Chaniurc?ptrri.\ypuri,s 4.9 1.2 Japan Broadleaf-evergreen 5.4 2.0 Incident 2.1 1.4 Oklahoma Postoak-Blackjack Oak Incident Puerto Rico Rain forest Incident 24.95 Laterite site Incident 2 1.20 Venezuelan rain forests - Podsol site Incident Karelia. Russia Spruce forest Incident Rain forest Malasia 33.24 21.68 - Incident 10.8 267.2 270.4 1.8 39.9 76.6 0.14 14.7 18.3 4.6 0.48 48.4 17.7 4.8 0.2 4.0 10.6 1 .1 14.3 2.07 15.0 12.1 2.77 3.9 1.07 4.0 3.9 0.86 136.8 13.0 4.3 26.0 18.2 21.8 4.9 57.2 10.69 17.07 3.60 1.32 24.79 23.41 27.0 3.37 5.09 28.91 6.65 3.97 24.93 24.61 28.40 3.52 2.7 0.3 2.9 4.8 0.9 1 . 1 - 1.0 2.3 0.9 52.5 28.9 5.25 12.5 14.0 3.25 1335 1433 1793 - - - 862 16.65 3444 44.53 3914 19.63 3505 46.63 4123 - - 650 2100 2500 - ___ 12 I2 12'' 12' 12 12 I 2 12" 12 1 2 4 129 I ? 1 2 4 7'J 7 - 6f 6J Period Reference. (months) notes ~ Ingham. 1950b lwatsubo and Tsutsumi. 1967 Johnson Risser. 1974 Jordan e r u/.. 1972 Jordan P / ( I / . . 1980 Kazimirov and Morozova. 1973 Kenworthy. 1970 North Dakota Gallery forest East slope West slope Incident Paris. France F[/'q".S s l h r k l r 4 6 years 30 years 50 90 years Over 200 years Incident Tennessee Mixed hardwoods Incident 4 Hardwood stands 9 Conifer stands Incident England New Brunswick 12" 12" 12 Killingbeck and Wali. 1978 0.20 0.88 11.5 12.1 4.0 3.1 8.6 7.4 292 0.20 0.39 16.2 23.4 5.3 4.1 9.5 14.6 381 0.30 0.43 4.1 7.7 2.5 4.0 8.9 10.6 450 - 48 3 372 363 412 554 - 12" 12" 12 1 2" 12" Lemee. 1974 8.04 6.84 14.48 0.23 9.40 14.73 5.75 7.18 7.00 14.18 0.29 11.53 15.29 6.16 8.46 6.08 14.54 0.62 17.30 18.16 6.15 10.37 8.84 19.21 0.88 36.53 28.39 11.58 5.09 4.71 9.8 0.29 2.42 8.99 2.89 32.0 13.0 - - 12 12 Lindberg C t a/.. 1979 12" 12" 12" Madgwick and Ovington. 1959 27.8 24.5 11.0 31.1 22.6 24.1 8.8 33.8 2.8 10.7 Nutrient depositions coritiriirerl Forest type and location Moscow region Ash-hazlenut-pine Incident Oxalis-fern spruce Whortleberry-spruce Mixed grass-birch Whortleberry-birch Whortleberry-pine Incident Beech Spruce Incident Mature secondary Southern Taiga Southern Sweden Ghana forest Incident Central Scotland Scots pine lncident Ireland Pinus contortu Incident Central Finland Scots Dine Incident H 2 0 Period Reference. NH,N NO,N N, P, K Cd Mg Na CI S0,S ( m m ) (months) notes ____ ~~ ~~ ~ ~ _ ~ _ 2.9 1 .o 2.2 1.9 2.0 2.4 4.3 1 .o - - 0.88 0.77 1.27 1.94 8.3 12.0 3.1 5.1 6.6 1.6 13.4 9.6 2.7 14.6 13.1 4.0 6.4 9.7 2.1 7.4 11.0 2.6 12.1 15.0 3.2 5 .1 6.6 1.6 154 4 Mina. 1965 230 4' 188 4 Mina. 1965 I53 4 217 4 175 4 177 4 273 4 - - 8.9 0.12 13.1 10.1 3.39 15.7 35.5 18.5 772 12'J4 Nihlgard. 1970 8.2 0.07 1.9 3.5 0.91 5.6 11.1 7.9 950 1 2 1 4 24.1 0.50 27.1 17.4 6.15 26 1 54.9 54.2 580 12iJy 29.4 4.1 237.5 41.6 29.1 16.4 0.41 17.5 12.7 11.3 10.0 5.4 5.6 4.0 1562 12' Nye. 1961 I847 12' - 29.0 453 12 Nicholson er u/ . . 13.5 794 12 1980 7.5 O'Hare. 1967 533 - 39.4 31.6 8.1 229.0 10.9 10.9 5.3 117.9 846 7.5 0.13 1.74 0.11 1.9 qch Paivanen, 1974 4ch Virginia Piedmont White pine 1.41 2.22 - 1.27 1.34 Oak-Tulip poplar 0.28 0.79 Incident 1.16 1.40 Piceu sitchen.\i.s Incident Western Australia Eurulyptus melanophlaicr Incident Maritime Wales Mediterranean France Ouerrus ilev Incident Quercu.) ;/ex Incident Quercus luiiugosii~u Incident Pinu.\ huleperi.sis Incident Quercus i1e.c Incident Wheat canopy Incident Central Minnesota Pine oak Marginal fen White cedar Incident Southern Sweden Piiius sy1w.srri.s Incident England 1.45 6.67 2.41 2.94 5.17 4.15 0.73 3.58 2.52 6.27 275.1 0.65 1.52 Oi22 4.00 1.39 5.24 375.9 9.22 4.13 1.08 4.24 3.02 6.36 342.0 0.76 1.50 0.26 4.52 1.52 5.36 392.4 19.4 21.7 22.8 132.2 - - - - 0.52 22.8 9.8 8.9 6.6 0.10 2.6 1.9 0.7 3.7 23.9 3.8 40.7 37.9 6.9 42.9 14.3 0.7 3.1 14.7 2.1 22.9 14.2 3.8 21.3 26.3 2.7 34.1 14.6 1.0 2.0 10.5 1.5 22.6 16.5 1.3 16.5 13.3 3.4 22.7 12.4 1 . 1 3.8 11.7 1.0 13.4 444 - - 645 718 - - 709 - - 15.7 1.4 12.8 21.7 4.4 39.6 14.6 0.8 3.8 10.2 1.7 19.1 602 22.6 16.4 - - 5.54 0.66 7.59 3.07 5.49 0.59 10.52 3.78 6.05 0.51 10.69 3.68 - - - - 4.42 -- 2.06 64 - - - - - 624 10.34 14.89 3.62 11.4 22.90 21.91 585 1.65 1.69 .47 2.3 4.05' 7.03 731 4 4 4 4 12 12' 12 12" 12" 12" 12" 12" 12" 12" 12"' 12" 2 2 12 I2 12 124 - 3J Parker. 1982 Potts. 1978 Prebble and Stirk. 1980 Rapp. 1969 Rapp, 1969 Rapp. 1969 Rapp, 1969 Rapp, 1973 Raybould et ul.. 1977 Reiners. 1972 Richter and 3/ Granat. 1978 Nutrient depositions cwri f ir i iwt l Forest type and location Pinus \\/w\tri\ Incident Southern Illinois Oak-Hickory Southern Georgia Cypress swamp Incident Russian taiga SDruce - Birch Incident Nevada Jeffrey pine i n Z G Central Hungary Qucrcus prrruru Incident Spruce Pine Incident Southern Sweden Alnus glutinow Brfulu wrrucosu Fruxinus r.ccrl.sior Qurrcus rohur Sorhus uucupuriu Southern Sweden Corylus rrvrllrflrl ~ H 2 0 Period Reference. NH,N NO,N N , P, K Ca Mg Na CI SO,S ( m m ) (months) notes _ _ _ _ _ ~ 1.48 6.43 1.98 2.19 5.89 9.33 1.79 10.5 21.69 13.46 302 1.13 5.96 0.33 1.7 3.75 6.15 378 4.5 0.9 36.5 25.1 4.6 17.7 1.3 12.1 11.1 1.7 2.68 0.96 6.55 6.24 2.59 3.32 0.22 1.75 2.23 0.69 - 3.2 0.2 10.9 2.3 2.4 0.2 1.2 0.1 8.2 1.6 5.7 1.0 9.61 2.03 6.83 5.17 2.72 0.33 5.1 0.19 0.58 0.36 0.14 - - - 1090 1 I45 1301 - 1.9 2.0 341 1.5 1.0 382 1.2 0.5 455 - - 194 21.30 2.34 28.51 26.76 5.01 2.41 15.41 49.64 - 20.09 1.32 7.43 17.40 2.31 1.45 12.62 17.83 595.3 >2.4 >2.2 >3.7 > 68 2.7 2.5 3.2 0.2 0.4 0.6 25.0 1.8 2.4 8.0 1.8 2.2 16.0 1.5 2.4 13.0 1.8 1.2 10.0 2.3 1.6 12.0 2.3 2.0 98 103 - - 194 - ~ ~ _ _ _ 3' 3' lYd Rolfe rt a / . . 1978 1 ZSd 1204 Schlesinger. 1978 I 2"q 6 Sokolov. 1972 6 6 12"" Stark. 1973 12" 9' Szaboand 94 Csortos. 1975 1.5 Tamm. 1951 1.5 1.5 Tamm. 1953. in Stenlid. 1958 Pitea abies Pinus sylvesrri.7 Incident Alder Conifer Mixed Western Oregon Incident Eastern Washington Abies grundis before defoliation after defoliation Incident Loblolly pine Hardwoods Incident Cuniprorhecu Incident Chumuecjpuris Mixed broadleaves Pinus- Chamurc~pa r is Incident Washington Red alder Incident Silver fir Georgia Japan Incident Incident Rain forest Papua New Guinea Incident 3.9 1.0 7.4 2.0 5.2 2.35 5.7 2.17 3.44 2.00 2.24 2.20 4.04 1.35 7.0 2.8 1.5 6.0 3.5 2.5 0.97 1.41 0.91 0.51 0.51 6.65 0.20 0.22 3.29 0.23 0.45 5.93 0 0.29 1.44 0.89 0.32 3.77 2.51 0.26 3.00 1.71 18.22 3.17 0.90 4.12 0.61 1 . 1 1 1.30 0.54 4.13 2.02 0.39 3.74 6.67 3.30 .67 3.72 0.75 1.40 0.20 3.20 1.5 27.5 25.3 10.0 0.7 7.0 19.6 10.0 0.12 11.51 16.03 4.11 0.42 37.10 13.82 5.12 0.45 3.37 10.43 1.8 1.08 20.99 25.56 5.77 0.63 2.57 10.71 2.63 10.6 1.3 15.0 12.4 5.58 1-7 0.3 2.2 2 . 2 0.5 2.6 0.5 12.3 6.0 2.2 1.3 0.4 0.8 0.6 0.1 34.91 23.27 6.47 83.18 0.81 0 0.27 8.35 147 206 227 I 2553 12" Tarrant er al.. 202 1 12" 1968 2251 12" 288 1 12' - - 9* Tiedemann et a/.. - 13b 1980 - 3h - - 8 Torreneuva. 1975 8 8 - - - - 12 Tsutsurni. 1977 - I2 - 12" - 12" - Iz" - - - 12 - 12 - - 12' Turner er a/ . . 1976 1370 12 - 12 Turner and 2730 12 Singer. 1976 - 2155 12 Turvey. 1974 2694 12 - - Nutrient deposi t ions c w r i r i r i i r t , t / Forest type H,O Period Relerence. and location NH,N NO,N N T PT K Ca Mg Na CI S0,S ( m m ) (nicinthb) notes ~ Solling. Germany F~7gU.5 .$J/t'CJfiCCJ Incident Washington Silver fir-Hemlock Incident Central Alaska Paper birch Black spruce Incident Aspen Black spruce Incident Southern Illinois Mixed hardwoods Northern Minnesota -- Incident North Carolina Uplands hardwoods Upland pine Loblolly pine Loblolly pine Loblolly pine Incident Loblolly pine Incident Loblolly pine Incident - 0.40 0.05 0.60 0.01 4.3 0.30 1573 1.573 0.62 2.9 0.18 0.99 0.99 0.20 2.1 0.09 0.16 0.98 0.06 1.97 1.64 8.62 1.54 18.28 18.66 4.02 1.34 1.67 1.67 7.16 0.62 4.04 5.27 1.31 1.18 2.74 2.25 7.32 0.48 1.06 3.48 0.70 1.09 0.1 I 0.15 I 0.28 4.02 7.92 0.37 0.10 0.58 6.70 0.02 1.16 1.46 5.09 0.61 18.13 14.49 4.00 0.76 2.08 4.54 0.74 9.60 9.79 2.37 0.74 1.46 3.53 0.28 0.88 3.42 0.72 4.61 0.37 4.06 4.14 1.15 4.44 0.36 3.54 4.28 1.42 3.71 0.34 6.56 5.61 1.83 5.45 0.21 1.64 2.81 0.67 1.49 2.47 8.60 0.48 I I 24 4.90 1.85 1.37 2.02 5.12 0.20 1.89 2.96 0.74 0.74 2.45 8.48 0.52 10.06 4.10 1.76 2.02 2.62 6.29 0.29 2.03 2.06 0.66 - - 287 597 596 775 - 1157 1196 1301 - - - - 9.89 - 7.89 - 12.70 - 9.59 - 12"' Ulrich or < I / . . 1978 12 I ? Ugolini cr I [ / . 12 1977 Van Cleve. in 12"" Cole and Rapp. 12 1981 5.5"" Verry and 5~5""Timmons. 1977 $5"1 I 2'.qn Weaver and I2'"" Brown. 1979 12' Wells er o/.. 1972 I" I 2 12" Wells and 12" Jorgensen, 1974 12" 12 I2 Wells C I d.. 1975 12 I Z h Wells cr d . , 1975 I Z h Coastal Australia Eucalpytus Incident New Zealand Corsican pine Incident Corsican pine Incident Corsican pine Incident New Zealand Douglas fir Incident New York Piiius resiiiosa Incident Pinus resiiiosa Incident Southern Sweden Polyethylene screening Incident Stainless steel screening Incident Steel screening Japan Incident 0.04 95