Published online 1 March 2007
Published in J Environ Qual 36:498-507 (2007)
DOI: 10.2134/jeq2005.0465
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Landscape and Watershed Processes
Inorganic Nitrogen and Phosphorus in Sierran Forest O Horizon Leachate
T. M. Loupea,
W. W. Millerc,*,
D. W. Johnsonc,
E. M. Carrollb,
D. Hansederb,
D. Glassa and
R. F. Walkerc
a Hydrologic Sciences Program, College of Agriculture, Biotechnology, and Natural Resources, Univ. of Nevada, Reno, NV 89512
b Dep. of Natural Resources and Environmental Science, College of Agriculture, Biotechnology, and Natural Resources, Univ. of Nevada, Reno, NV 89512
c Dep. of Natural Resources and Environmental Science, College of Agriculture, Biotechnology, and Natural Resources, Univ. of Nevada, Reno, NV 89512
* Corresponding author (wilymalr{at}cabnr.unr.edu)
Received for publication December 19, 2005.
 |
ABSTRACT
|
|---|
High in situ concentrations of inorganic N and P have been reported in overland/litter interflow from Sierran forests, indicating that these nutrients are derived from the forest floor O horizons. To test this hypothesis, forest floor monoliths consisting of the combined Oe and Oi horizons were collected near the South Shore of Lake Tahoe, Nevada, for leaching experiments. Three monoliths were left intact, and three were hand-separated according to horizon for a total of three treatments (combined Oe+Oi, Oe only, and Oi only) by three replications. Samples were randomized and placed into lined leaching bins. Initial leaching consisted of misting to simulate typical early fall precipitation. This was followed by daily snow applications and a final misting to simulate spring precipitation. Leachate was collected, analyzed for NH4+N, NO3N, and PO43P, and a nutrient balance was computed. There was a net retention of NH4+N, but a net release of both NO3N and PO43P, and a net release of inorganic N and P overall. Total contributions (mg) of N and P were highest from the Oe and Oe+Oi combined treatments, but when expressed as per unit mass, significantly (p < 0.05) higher amounts of NO3N and PO43P were derived from the Oi materials. The nutrients in forest floor leachate are a potential source of biologically available N and P to adjacent surface waters. Transport of these nutrients from the terrestrial to the aquatic system in the Lake Tahoe basin may therefore play a part in the already deteriorating clarity of the lake.
 |
INTRODUCTION
|
|---|
DUE to the exclusion of fire and forest harvest in the eastern Sierra Nevada, forest stands have become dense and unhealthy, and are now characterized by reduced growth, increased disease, and increased insect infestation (Ansley and Battles, 1998; Stephens and Moghaddas, 2005). Although coniferous forests typically exhibit low concentrations of N and P in the soil and plant biomass compared with most other forested ecosystems (Stohlgren, 1988), Sierran forests have become dominated by young trees and have accumulated a very thick O horizon over time (Neary et al., 1999). In healthy coniferous ecosystems, rapidly decomposing litter can sustain a short-term cycling of nutrients from the canopies to the soils and back to the canopies. However, an accumulating forest floor similar to that present in the eastern Sierras tends to sequester nutrients leading to the buildup of large nutrient pools that tie up N and P, and at the same time may contribute to long-term nutrient discharge to surface waters (Piatek and Allen, 2001).
An environment where the forest litter builds up on the surface of the mineral soil faster than it degrades is thus undesirable for forest management because nutrient cycles develop progressively slower rates of turnover (Covington and Sackett, 1984), and the forest floor is no longer able to maintain a short-term nutrient cycle. In addition, the excess nutrients now present in the litter may either leach into the soil or discharge from the area by way of surface runoff and/or litter interflow. Recent research has identified the presence of high concentrations of biologically available nitrogen (NH4+N, NO3N) and phosphorus (PO43P) in coniferous forest overland flow (Miller et al., 2005). This suggests that these nutrients are derived from the overlying forest floor O horizons, and that there has been little biological uptake, leaching, or direct contact with the mineral soil where strong retention of NH4+N and PO43P would be expected. Although commonly considered to be an unimportant water flow pathway in forest ecosystems because vegetation and litter layers typically promote maximum infiltration and soil moisture and nutrient storage, it now appears that overland flow may be an important source of dissolved nutrients discharged to nearby streams and lakes. The potential magnitude, however, remains largely unknown.
The hypothesis for our study was that forest floor O horizon materials are an important source of the biologically available N and P contained in litter interflow. The test objective was to quantify, under laboratory conditions, the discharge of these nutrients by measuring their concentrations in leachate derived from intact monoliths of O horizon materials collected in situ.
 |
MATERIALS AND METHODS
|
|---|
Intact O horizon monoliths were collected from a study site located approximately 2 km southeast of South Shore Lake Tahoe, in Stateline, NV (57.5° N, 119.92° W). The slope gradient ranges from 10 to 40%. Dominant tree species are Jeffrey pine (Pinus jeffreyi [Grev. and Balf.]) and California white fir (Abies concolor var. lowiana [Gord. and Glend.] Lindl.), with substantially lesser amounts of sugar pine (Pinus lambertianna [Dougl.]). The understory vegetation consists of Sierra chinquapin (Castanopsis sempervirens [Dudl.]) and currant (Ribes spp.), with some tobacco brush (Ceanothus velutinis [Dougl. ex Hook.]), greenleaf manzanita (Arctostaphylos patula [Greene]), and antelope bitterbrush (Purshia tridentata [Pursh]). The soils developed from granitic parent material, belong to the Cagwin-Rock Outcrop complex, and are classified as mixed Typic Cryopsamments.
Six intact monoliths composed primarily of Jeffrey pine and white fir needle litter and consisting of the combined Oa (highly decomposed unidentifiable organic material), Oe (partially decomposed litter), and Oi (fresh needle fall litter) horizons were collected for greenhouse leaching experiments (Fig. 1). Adjacent field samples were taken for determination of initial nutrient (N and P) status. Air dry monoliths were collected in late summer/early fall of 2004 using 0.63 cm thick 0.61 m by 2.44 m (1.49 m2) sheets of plywood covered with a 1.22 by 4.88 m vinyl polyethylene tarp to reduce friction. A 30 cm by 3 m swath of O horizon material was first excavated to the mineral soil surface. Thickness of the O horizon ranged from about 10 to 15 cm, of which approximately two thirds consisted of the Oe material. The leading edge along the length dimension of the plastic-covered plywood was placed at the now exposed O-A horizon interface, and manually inserted along the interface until the width dimension of the plywood sampler became flush with the sidewall of the original excavation. The O horizon was then vertically severed around the perimeter of the plywood sampler, the intact monolith removed, covered in plastic, and transported to the greenhouse for study. Very little, if any, mineral soil was collected using this technique.
Once in the greenhouse, three monoliths were left completely intact, and since an Oa horizon was absent from the collection area, three were hand-separated according to the remaining horizons using the same procedure as applied in the field, for a total of three treatments (intact Oe+Oi, Oe only, and Oi only) by three replications. The criteria for Oe and Oi separation was the depth at which the largely unaltered needle litter fall was no longer present (
5 cm) and the underlying material was clearly in various stages of more fragmented decomposition (Fisher and Binkley, 2000). Nine 1.39 m2 open top box leaching bins were constructed from 1.9 cm plywood. Each of the nine bins were centered on adjacent but individual bench supports resulting in a bin separation distance of about 1.50 m. Bins were adjusted to a 2% slope and lined with 4 mil heavy-duty all purpose clear plastic sheeting to prevent leaking. A hole was drilled into the bottom of the down gradient end of each bin, and a funnel mechanism installed for drainage discharge and sample collection. The various treatments were randomized, and the individual monoliths were trimmed and fitted into the lined bins. This approach allowed for the separation of potential nutrient sources: (i) needle litter only; (ii) the partially decomposed layer; and (iii) the intact combination.
To simulate an early fall rainfall precipitation event common to the eastern Sierras, misting was applied to initially wet the O horizon material in each treatment. Municipal water was supplied via a suspended 3 m, 2.54 cm diam. PVC pipe, running the length of the 2.5 m leaching bin (Fig. 2). Seven misting nozzles (orifice size = 0.030 in.) were inserted along the supply line at distances of 1, 75, 90, 150, 210, 225, and 299 cm from the main water source. Before beginning the actual leaching experiment, several arrays of 300 mL beakers were distributed across the total misting area and the average minimum misting application rate was measured at approximately 0.05 cm min1. Leaching bins were then placed on separate bench areas, and the leaching experiments were initiated in February 2005. Before starting the experiment, six small beakers were randomly embedded in the litter materials throughout each bin. The intent was twofold: (i) to collect samples of source water from each leaching cycle for nutrient analysis; and (ii) to measure the average volume of water applied during each cycle.
Because the minimum misting rate was faster than most rainfall events common to the area, the initial wetting consisted of misting for 5 min at 30 min intervals over 3 h (six applications), for an estimated total application of approximately 1.5 cm of rainfall precipitation. The initial misting discharge period was approximately 6 h long. This was followed by seven cycles of daily snow applications (24-h discharge period), of approximately 2 to 4 cm of water equivalent each. Snow was used for most of the leaching process because it accounts for the majority (
60%) of the precipitation in this region (Johnson et al., 1997), because litter decomposition occurs mostly under a snow pack due to the insulating effect (Stohlgren, 1988), and because of the presence of available moisture which is limiting during the warmer seasons (Stark, 1973). Snow was collected using plastic shovels from a location near Mount Rose summit (approx. 45 km away), shoveled into a tarp-lined truck bed, covered, and driven to the University of Nevada, Reno campus greenhouse. The snow was then shoveled into lined wheelbarrows, and distributed evenly across each leaching bin. During the course of the overall experiment, snow fell on Mount Rose summit on average every other day, with multiple inches of accumulation between each collection event.
Following the seven snowmelt cycles, misting was used again to simulate the effects of a spring rainfall on snow event. The final misting was performed in the same manner as previously described; however, only 1 cm of water was applied to each bin over the course of 2 h. The final misting discharge period was about 5 h long, making the cumulative treatment period including the mist and snow applications approximately 10 d. The average total water application throughout the experiment for each bin was 19.6 cm, roughly 23% of the average annual precipitation for the Lake Tahoe basin.
Leachate discharged from misting and snowmelt applications was collected continuously throughout each treatment cycle, sequential and total discharge volume was measured, and subsamples were taken for nutrient analysis. During the initial and final misting applications, samples were collected on volume intervals, one for every 1000 mL that discharged from the outlet. During the snowmelt period, samples were instead collected on time intervals, waiting approximately 2 to 3 h between each sampling event. Source misting water and snow samples from each treatment cycle were also collected, and all samples were analyzed for NH4+N, NO3N, and PO43P at the Oklahoma State University Soil, Water, and Forage Analytical Laboratory (Stillwater, OK). Ammonium N was analyzed on a Lachat flow-injection analyzer using QuickChem method 12-107-06-2-A (LACHAT, 1994), NO2+NO3N was analyzed on a Lachat flow-injection analyzer using QuickChem method 12-107-04-1-B (LACHAT, 1994), and PO43P was analyzed using an inductively coupled plasma (ICP) spectrometer (APHA, 1995). Average nutrient input concentrations from the municipal source and Sierra snow are presented in Table 1.
The entire leaching experiment was performed during the month of February (mid-winter in the Sierras). Average daily and nighttime temperatures were approximately 10 and 2°C, respectively. Temperatures inside the leaching bins were measured using Onset, Co. HOBO temperature data loggers, which were placed underneath the monolith material in each of the nine leaching bins. During the misting portions of the experiment, temperatures ranged from approximately 15 to 30°C, which is also representative of the average range for ambient daytime greenhouse temperatures throughout the experiment. Nighttime greenhouse temperatures were approximately 5°C. During the snowmelt cycles the temperatures remained below 17°C even during the warmest parts of the day, and were generally below 5°C for the evening and early morning hours. Even with daily snow applications, the snow tended to melt almost entirely within a 24-h period. Nonetheless, daily applications of snow maintained continuous field capacity (24-h drainage by gravity alone) moisture conditions in the monolith materials.
Following gravity drainage from the final misting application, the total content of each bin was weighed wet and subsamples were collected for moisture content determination. For the combined horizon treatments, the organic material was separated as previously described into Oe and Oi components and weighed, with separate subsamples collected for each. This was done because the moisture content of Oi litter would be less than that of the more highly decomposed Oe portion. Although rarely present, when the bins were being emptied any mineral soil deposits were manually isolated from the organic material. This small mineral fraction could have provided an additional source of NO3N, but would generally be regarded as a potential sink for NH4+N and PO43P. Sticks, cones, and bark material were also separated from the rest of the bin contents. Needles and decomposed components have been shown to contain much higher concentrations of N and P than wood, bark, or cones (Smethurst and Nambiar, 1990). For this reason, and because of the different moisture storage capacities compared with that of the Oe and Oi materials, separate moisture contents were again determined for the wood, bark, and cones. The wet weight of the sticks, bark and cone materials was subtracted from the total wet weight of the bin contents before dry mass determination for the remaining organic materials.
An individual input/output water balance was developed for each of the nine leaching bins using the following relationship:
 | [1] |
Due to the time of year (winter), ambient greenhouse temperatures, and the comparatively short duration of each wetting cycle, evaporation was assumed to be minimal compared with the total inflow and, therefore, disregarded as a major component of the water balance. The mass water content of each organic layer immediately following complete gravity drainage from the final misting treatment was taken as the maximum litter moisture storage after drainage for the overall experiment.
Combining the inflow/outflow water balance with measured concentrations of NH4+N, NO3N, and PO43P in the misting water, each snow application, and discharged leachate, a nutrient mass balance was determined for each treatment from a similar relationship:
 | [2] |
Where Ci is input concentration, Co is output concentration, and r is nutrient discharge or retention. Any difference between the nutrient content entering the bins in the input water and that discharged during leaching/drainage was assumed to be either retained or released from the O horizon materials. The nutrient discharge or retention was then converted to nutrient mass per unit mass of organic litter using the total dry weight content of each bin. A simple analysis of variance (ANOVA) was performed for both the overall discharge mass and the mass per dry weight of bin material for each inorganic nutrient form. Finally, for overall nutrient loading throughout the experiment, the mass per unit mass of each nutrient retained or discharged was converted to kg ha1, based on the actual dry mass of bin material per unit area (kg m2) of each leaching bin.
 |
RESULTS AND DISCUSSION
|
|---|
Water Balance
The actual volume of water input proved difficult to quantify. The six drop can (beaker) placements in each bin were ineffective at accurately characterizing water input. This was due, in part, to changes in air flow and water drop distribution as a result of bin placement on each misting bench. In addition, snowmelt often circumvented collection due to preferential flow around the edges of the open mouth beakers. As a result, the initial sum of the moisture storage and the measured outflow was often larger than the estimated inflow volume based on projected misting rates and the moisture equivalent of snow applications. Because the total outflow plus the litter moisture storage cannot exceed the total inflow (Eq. [1]), the actual volume of water applied during the initial and final misting treatments had to be greater than the 0.05 cm min1 application rate would predict (i.e., 20832 cm3 bin1 and 13888 cm3 bin1, for first and last misting, respectively). Once the litter moisture storage was satisfied, the inflow for each subsequent leaching cycle could be assumed equal to the outflow. Assuming inflow to equal outflow for the final misting cycle, new input volumes during the initial and final misting treatments were estimated to be 21535 cm3 and 14357 cm3, respectively.
The water balance for the overall experiment is presented in Table 2. Based on these data and the constraints of Eq. [1], the average estimated total inflow across all treatments was 272558 cm3, or 19.6 cm of applied water, ± 1.4 cm. During the initial misting, the time was noted at which the first outflow of discharge water was observed. The Oi treatments generally demonstrated the shortest time to discharge, but the relationship was significant (t-test, p < 0.05) only when compared with the combined Oi+Oe treatments, in which the monoliths were more intact. Since the Oe layers in coniferous forests are largely water repellent, thus restricting infiltration and uniform wetting (Brock and DeBano, 1982; Naslas et al., 1994; Cleveland et al., 2004), the Oe only and the Oi+Oe combined treatments were expected to exhibit longer times to initial outflow. The relatively short time to outflow observed in two of the Oe only treatments may have been the result of disturbance caused in the natural framework of the litter mat when the layers were hand-separated, or to the presence of natural preferential flow paths that became exposed to positive pressure heads following removal of the overlying Oi materials. In any case, it was apparent that the moisture storage capacity of the Oi materials was easily satisfied during the initial misting cycle.
The moisture storage capacity of the Oe horizons was found to be satisfied following the first snowmelt cycle, and was from then on presumed to be at the maximum water content measured on termination of the overall experiment. Although it was later found that several of the Oe horizons contained nonwetted areas, it is believed that these remained intact throughout the experiment and that water flow and overall moisture storage was dominated by well established preferential flow paths (Burcar et al., 1994; Naslas et al., 1994). Applying Eq. [1] and assuming
LMS to be zero for all Oi materials after the initial misting and for all Oe materials after the first snowmelt cycle, a reasonable estimate of water inflow/outflow was derived for each leaching cycle (i.e., two misting, seven snowmelt) for each bin, thus providing an overall and sequential water balance.
O Horizon Total Nitrogen and Phosphorus Concentrations
Initial total N and P content (mg kg1) of O horizon materials from the field samples are presented in Table 3 along with the water-extractable nitrogen and phosphorus loads determined from the leaching experiments. The NO3N leached from the O horizon materials (mg kg1 dry mass) for the overall experiment was less than 1% of the total N concentration (mg kg1) measured in the litter before leaching. The PO43P leached was about 15% of the total P contained in the Oi horizon material, and less than 5% from the Oe.
View this table:
[in this window]
[in a new window]
|
Table 3. Total N and P concentrations in litter before leaching and inorganic N and P measured in discharge leachate throughout the experiment.
|
|
Leachate Nitrogen and Phosphorus Concentrations
Nitrogen
The NO3N and NH4+N concentrations in the leachate are presented in Fig. 3 and 4, respectively, for all three treatment types from each misting and snowmelt cycle. Concentrations of both NO3N and NH4+N in the outflow during the initial misting were sporadic, with reciprocating low and high concentrations measured throughout (Fig. 3 and 4), and peak concentrations of about 0.45 mg L1 and 0.80 mg L1, respectively. Of particular interest, however, was the apparent inverse relationship between discharge concentrations of NO3N and NH4+N in the leachate (Fig. 5). This trend of low NO3N discharge concentrations with correspondingly high NH4+N discharge concentrations (and vice versa), generally continued throughout the following six snowmelt cycles.

View larger version (48K):
[in this window]
[in a new window]
|
Fig. 3. Nitrate-N leachate concentrations from each misting and snowmelt cycle for the (A) Oi, (B) Oe, and (C) the Oi + Oe combined treatments. Note the scale differences.
|
|

View larger version (50K):
[in this window]
[in a new window]
|
Fig. 4. Ammonium-N leachate concentrations from each misting and snowmelt cycle for the (A) Oi, (B) Oe, and (C) the Oi + Oe combined treatments. Note the scale differences.
|
|
For example, snowmelt cycles 1, 3, and 5 typically exhibited lower NO3N concentrations (Fig. 3) in the leachate during initial discharge (<0.1 mg L1) than toward the end of the discharge period (
0.5 mg L1). Conversely, NH4+N concentrations in the leachate for the same cycles were higher (
0.4 mg L1) at the beginning of the sampling period (Fig. 4) than toward the end (<0.1 mg L1). For snowmelt cycles 2, 4, and 6, this trend typically continued in reverse; i.e., the higher initial NO3N concentrations (
0.1 to 0.55 mg L1) gradually began to decline, and the initially low NH4+N concentrations (
0.01 to 0.2 mg L1) correspondingly began to increase (if not peak) throughout the sampling period. These general patterns indicate that nitrification may be a factor during the latter stages of odd numbered snowmelt cycles (1, 3, and 5), likely due to the insulating effect of the snow pack (Stohlgren, 1988). Nitrate-N discharge concentrations then continued to decline during even numbered snowmelt cycles (2, 4, and 6), following which the overall process was found to repeat.
An exception to this pattern was noted during the last snowmelt cycle (cycle 7), which exhibited low concentrations of both NO3N and NH4+N (<0.1 mg L1) in the discharge leachate. These data suggested that the slow leaching characteristic of snowmelt had discharged most of the soluble inorganic nitrogen by the time the last cycle was initiated. However, during the final misting concentrations of NO3N and NH4+N in the leachate again increased. Since the misting application rate was greater than that during snowmelt, it may be that new areas of inherently water repellent pockets began to wet, or more likely that immobilized water containing higher nutrient concentrations during periods of slow snowmelt became mobilized at the higher application rate of the final misting process.
Phosphorus
Phosphate-P was discharged from all treatments during each misting and snowmelt period throughout the course of the experiment (Fig. 6). The highest discharge concentrations, however, were observed during the initial misting cycle for all treatment types. Phosphate-P concentrations in the leachate from each replicate of the Oi only treatment during the initial misting cycle reached the highest levels recorded throughout the experiment (5.6, 4.8, and 4.1 mg L1, respectively) (Fig. 6A). From the time of the second snowmelt cycle, concentrations of PO43P in the leachate remained fairly constant at <1.0 mg L1 throughout each individual sampling period.

View larger version (38K):
[in this window]
[in a new window]
|
Fig. 6. Phosphate-P leachate concentrations from each misting and snowmelt cycle for the (A) Oi, (B) Oe, and (C) the Oi + Oe combined treatments. Note the scale differences.
|
|
Although Susfalk and Johnson (2002) have reported substantial soil P fluxes (0.68 kg P ha1 yr1) from resin-based lysimeters in a Sierran coniferous forest in nearby Little Valley, NV, P fluxes from adjacent ceramic cup lysimeter samples from the same study were below the level of detection. Findings from our leaching study along with the resin-based lysimeters of Susfalk and Johnson (2002) could indicate that the ceramic cup lysimeters may not have collected the main pulse of percolating phosphorus because it had already passed by the time suction was re-established in the lysimeters following the dry summer months.
Comparison with Lake Tahoe Basin Water Quality Standards
Water quality data collected on various tributaries of the Tahoe basin by the Lake Tahoe Interagency Monitoring Program (LTIMP) between 1989 and 1996 have shown NH4+-N concentrations typically less than 0.01 mg L1, but ranging up to 0.075 mg L1, and discharge-weighted concentrations of PO43 (reported as phosphate) and total phosphorus (TP) ranging from <0.01 to 0.02 and 0.02 to 0.22 mg L1, respectively (Reuter and Miller, 2000). In terms of ground water quality, Loeb and Goldman (1979) reported the well chemistry from four wells in the Ward Valley watershed to contain average concentrations of less than 0.015 mg L1 NH4+, 0.16 mg L1 NO3, and 0.07 mg L1 soluble reactive P. Accordingly, the Lahontan Regional Water Quality Control Board (LRWCB) has established storm water effluent limitations for surface water runoff that enters directly into Lake Tahoe (or into lake tributaries) for total N concentrations of 0.5 mg L1 and total P of 0.1 mg L1; and surface water guidelines in the Lake Tahoe basin watershed for maximum allowable total N concentrations of 0.21 mg L1, and soluble reactive P concentrations of 0.009 mg L1 (CA-EPA, 1994).
Based on the above, the observed peak concentrations of NH4+N and NO3N found in leachate from the overall course of this study (0.8 and 0.75 mg N L1, respectively) are clearly of a sufficient magnitude for concern. Furthermore, since the waters of Lake Tahoe are now considered to be trending toward a P-limiting condition (Goldman, 1988), the observed leachate concentrations of inorganic P are of even greater concern. Initial discharge concentrations of PO43P were as high as 5.6 mg P L1, 50 times greater than LRWCB surface runoff limitations and almost three orders of magnitude greater than surface water guidelines for the Lake Tahoe basin.
Nutrient Mass Balance and Discharge Loading
Nitrate Nitrogen
Variable amounts (mg) of NO3N were discharged from all treatments during each leaching cycle, with sporadic peaks recorded throughout the entire experiment, indicating a continual discharge via O horizon leaching. This was anticipated because the soluble inorganic NO3 anion would not normally be expected to adsorb to organic colloidal materials. ANOVA showed that the combined Oi+Oe treatment contributed more N as NO3 throughout the overall experiment than did either of the individual litter layer treatments. This difference was found to be only moderately significant (p < 0.10). Although not significant, the Oe only treatment also discharged more NO3N (mg) than did the Oi only, largely due to the comparative total mass of respective litter materials. In the area of monolith collection, Oi horizon materials averaged about 1.8 kg m2, with about 5.4 kg m2 of underlying Oe material.
When converted to mass N per unit mass of dry litter, the differences between treatments were not significant according to the ANOVA. However, the Oi only treatment tended to release more N as NO3 per dry unit mass of material than did either the Oe or the combined Oi+Oe horizon treatments.
Ammonium Nitrogen
Nitrogen in the form of NH4+ was generally retained and/or transformed through nitrification in all litter treatments over the course of the experiment. Only two of the leaching bins (#4 and #9) exhibited a cumulative release of NH4+N, both of which contained the combined Oi+Oe treatments. This effect in itself was not surprising in that NH4+ in particular is readily immobilized during the first 6 mo to 1 yr of litter decomposition (Rustad and Cronan, 1988; Monleon and Cromack, 1996). Alternatively, a predominant mechanism for abiotic nitrogen immobilization is thought to involve the reaction of NH4+ with plant materials containing large concentrations of lignin (Schimel and Firestone, 1989). Since coniferous forest litter contains a substantial amount of lignin, exposure to these organic constituents at and within needle surfaces during early stages of decay may have led to abiotic chemical immobilization (commonly termed ligand exchange), and hence, the observed retention of NH4+N overall (Schimel and Firestone, 1989). Furthermore, in the absence of extremely acidic conditions (pH < 4.5), soluble NH4+ cations could be expected to adsorb to the charged surfaces of colloidal organic or mineral components associated with the forest floor matrix.
Based on ANOVA, the amount of NH4+N retention in the Oi treatments was significantly higher for both total mass (p < 0.05), and mass per unit dry mass (p < 0.005) of litter material. Although net biotic N accumulation has frequently been reported during the early stages of the decay process (Klemmedson et al., 1985; Yavitt and Fahey, 1986; Stohlgren, 1988), the Oi horizon consists largely of intact only slightly decayed needle litter. Hence, the specific mechanism of NH4+ retention in the needle litter remains unclear.
Less NH4+N retention and even some discharge was observed in leachate from the intact monoliths containing both the Oi+Oe horizons. These monolith treatments were undisturbed throughout the experiment and microbial mineralization (i.e., ammonification) may have remained more active compared with the separate, albeit slightly disturbed, litter layers. Elevated NH4+ concentrations have been reported in forest leachate after the microbial population has consumed a substantial part of the extractable organic N (Cortina et al., 1995), but in the Sierras, this process appears to be most prominent during the winter due to enhanced moisture content and the incubating effect of the overlying snow pack (Stark, 1973).
Phosphate Phosphorus
The greatest amount (mg) of PO43P was discharged from all treatment types during the initial misting cycle and declined thereafter. Measured differences throughout the course of the experiment between the treatments were found to be nonsignificant (p > 0.1) based on ANOVA. However, the combined horizon treatment tended to contribute greater amounts of PO43P in the discharge leachate than did the Oi only treatment. When expressed on a per unit mass basis (mg kg1), ANOVA showed a significantly greater (p < 0.05) amount of PO43P discharged from the fresh litter (Oi only) than from either of the other treatment types. Although not found to be significant, the Oe only treatment also contributed larger amounts of PO43P than did the combined horizon treatment.
These findings are consistent with the literature in that P has been reported to be rapidly lost by leaching during the initial stages (the first 3 to 6 mo) of litter decomposition (Edmonds, 1979; Rustad and Cronan, 1988; Cortina et al., 1995), with continued P release over time but at a much slower rate (Monleon and Cromack, 1996). Indeed, the reported P loss was so rapid during the early stages of litter decay for two separate studies on litter nutrient release, that the proportion of P remaining in the litter was about 50% after 1 yr, and 30% after 6 mo of incubation, respectively (Rustad and Cronan, 1988; Monleon and Cromack, 1996). Compared with the newly fallen litter, the PO43P discharged per unit mass from the Oe only treatments remained lower throughout the experiment, most likely because the older organic material had already lost a substantial portion of its phosphorus from leaching during the previous season.
In summary, there was a net discharge of biologically available N and P from the combined Oi+Oe materials during leaching. Specifically, when considering the combined horizon treatments because they best represent in situ field conditions, net contributions of N were approximately 4 mg N kg1 (0.45 kg N ha1) and for P approximately 17 mg P kg1 (2 kg P ha1) (Table 4). By way of comparison to the field, we can multiply the N and P yields per unit O horizon mass from this study by the mass of O horizon material measured in the field near where the monoliths were collected (93200 kg ha1). This produces a value for N yield of 0.37 kg N ha1 and for P yield of 1.58 kg P ha1, numbers very close to those estimated above from the leaching bin area. The N leaching values calculated either way are similar to the annual leaching values using resin lysimeters as reported by Murphy et al. (2006) for the field site location (an unburned control location) from which the monoliths were collected (0.37 ± 0.07 kg N ha1 yr1). However, the P values contained in O horizon leachate reported here are much greater than those reported by Murphy et al. (2006) for soil leaching flux (0.48 ± 0.16 kg P ha1 yr1). In both cases it should be noted that the values measured by Murphy et al. (2006) represented 8 mo of soil leaching fluxes over the winter season whereas the values in this study were for a much shorter period. Thus, N and P in runoff (O horizon leachate) appear to be of greater magnitude than from in situ soil leaching, suggesting that O horizon runoff may be potentially discharged to adjacent surface waters by way of overland/litter interflow.
View this table:
[in this window]
[in a new window]
|
Table 4. Summary of inorganic N and P discharge loads. Negative values indicate net retention. Area based on a 1.39 m2 leaching bin monolith.
|
|
 |
CONCLUSIONS
|
|---|
Our hypothesis that O horizon materials are an important source of the inorganic N and P contained in litter inflow from Sierran watersheds was supported. There was a net discharge of both inorganic N (4 mg N kg1, 0.45 kg N ha1) and P (17 mg P kg1, 2 kg P ha1) from O horizon materials. Although the Oi horizon contribution of NO3N and PO43P was greatest on a mass per unit mass basis (mg N or P kg1 dry matter), the Oe will likely contribute more in situ due to the larger total mass present. Nitrogen discharge concentrations were not as high as those previously reported in situ by Miller et al. (2005). However, we believe this is due to the comparatively short-term nature of the study and the application of only 23% of the typical MAP. Furthermore, N demonstrated a fluctuating discharge characteristic, which may explain why many investigators have reported early in situ peak concentrations in snowmelt discharge (Coats et al., 1976; Yavitt and Fahey, 1986; Williams and Melack, 1991), whereas others have identified late season peaks (Johnson et al., 2001). Phosphorus discharge, on the other hand, was very rapid, occurring primarily during the initial misting cycle and remaining fairly constant at much lower concentrations thereafter. This early discharge effect could explain why high PO43P concentrations in adjacent tributaries are seldom reported in the literature.
The effect of almost a century of fire suppression has been a general decline of forest and watershed health, including the excessive build up of forest litter containing much larger pools of soluble N and P. These nutrients, when discharged by way of forest floor leachate through litter interflow represent a potential source of biologically available N and P to adjacent surface waters. Transport of these nutrients from the terrestrial to the aquatic system in the Lake Tahoe basin may therefore play a role in the deterioration of lake clarity.
 |
ACKNOWLEDGMENTS
|
|---|
We gratefully acknowledge the participation and support of the Nevada Agricultural Experiment Station, McIntyre-Stennis, and the US Forest Service Lake Tahoe Basin Management Unit and Joint Fire Sciences Program. NAES publication #52055642.
 |
NOTES
|
|---|
A contribution of the Dep. of Natural Resources and Environmental Science. Research supported in part by Nevada Agricultural Experiment Station, publ. # 52055642.
 |
REFERENCES
|
|---|
- America Public Health Association. 1995. Standard methods for the examination of water and wastewater. 19th ed. American Public Health Association, Washington, DC.
- Ansley, J.S., and J.J. Battles. 1998. Forest composition, structure, and change in an old-growth mixed conifer forest in the northern Sierra Nevada. J. Torrey Bot. Soc. 125:297308.
- Brock, J., and L. DeBano. 1982. Runoff and sedimentation potentials influenced by litter and slope on chaparral community in central Arizona. p. 372377. In E.C. Conrad and W.G. Oechel (Tech. Coords.) Dynamics and Management of Mediterranean-type Ecosystems: Proceedings of the symposium: 1981 June 2226; San Diego, CA. Gen. Tech. Rep. PSW-58. U.S. Dep. of Agriculture, Forest Service, Pacific Southwest Forest and Range Experiment Station, Berkeley, CA.
- Burcar, S., W.W. Miller, S.W. Tyler, and D.W. Johnson. 1994. Seasonal preferential flow in two Sierra Nevada soils under forested and meadow cover. Soil Sci. Soc. Am. J. 58:15551561.[Web of Science]
- CA-EPA. 1994. Lahontan Regional Water Quality Control Board, Basin Plan. California Environmental Protection Agency, Sacramento, CA. Available at www.waterboards.ca.gov/lahontan/BPlan/BPlan_Index.htm (verified 27 Nov. 2006).
- Cleveland, C., J. Neff, A. Townsend, and E. Hood. 2004. Composition, dynamics, and fate of leached dissolved organic matter in terrestrial ecosystems: Results of a decomposition experiment. Ecosystems 7:275285.
- Coats, R., R. Leonard, and C. Goldman. 1976. Nitrogen uptake and release in a forested watershed, Lake Tahoe Basin, California. Ecology 57:9951004.
- Cortina, J., J. Romanya, and V.R. Vallejo. 1995. Nitrogen and phosphorus leaching from the forest floor of a mature Pinus radiata stand. Geoderma 66:321330.[CrossRef][Web of Science]
- Covington, W.W., and S.S. Sackett. 1984. The effects of a prescribed burn in southwestern Ponderosa pine on organic matter and nutrients in woody debris and forest floor. For. Sci. 30:183192.
- Edmonds, R.L. 1979. Decomposition and nutrient release in Douglas-fir needle litter in relation to stand development. Can. J. For. Res. 9:132139.
- Fisher, R.F., and D. Binkley. 2000. Ecology and management of forest soils. John Wiley & Sons, New York.
- Goldman, C.R. 1988. Primary productivity, nutrients, and transparency during the early onset of eutrophication in ultra-oligotrophic Lake Tahoe, California-Nevada. Limnol. Oceanogr. 33:13211333.
- Johnson, D.W., R.B. Susfalk, and R.A. Dahlgren. 1997. Nutrient fluxes in forests of the eastern Sierra Nevada Mountains, USA. Global Biogeochem. Cycles 11:673681.[CrossRef][Web of Science]
- Johnson, D., R. Susfalk, R. Dahlgren, T. Caldwell, and W.W. Miller. 2001. Nutrient fluxes in a snow-dominated, semi-arid forest: Spatial and temporal patterns. Biogeochemistry 55:219245.
- Klemmedson, J.O., C.E. Meier, and R.E. Campbell. 1985. Needle decomposition and nutrient release in Ponderosa pine ecosystems. Forest Sci. 31:647660.
- LACHAT. 1994. QuickChem Method 12-107-04-1-B (Nitrate) and 12-107-06-2-A (Ammonium). LACHAT Instrument, Milwaukee, WI.
- Loeb, S.L., and C.R. Goldman. 1979. Water and nutrient transport via groundwater from Ward Valley into Lake Tahoe. Limnol. Oceanogr. 24:11461154.
- Miller, W.W., D.W. Johnson, C. Denton, P.S.J. Verburg, G.L. Dana, and R.F. Walker. 2005. Inconspicuous nutrient laden surface runoff from mature forest Sierran watersheds. Water Air Soil Pollut. 163:317.
- Monleon, V., and K. Cromack, Jr. 1996. Long-term effects of prescribed underburning on litter decomposition and nutrient release in ponderosa pine stands in central Oregon. For. Ecol. Manage. 81:143152.
- Murphy, J.D., D.W. Johnson, W.W. Miller, R.F. Walker, E.M. Carroll, and R.R. Blank. 2006. Wildfire effects on soil nutrients and leaching in a Tahoe basin watershed. J. Environ. Qual. 35:479489.[Abstract/Free Full Text]
- Naslas, G.D., W.W. Miller, R.R. Blank, and G.F. Gifford. 1994. Sediment, nitrate, and ammonium in surface runoff from two Tahoe basin soil types. Water Resour. Bull. 30:409417.
- Neary, D., C. Klopatek, L. DeBano, and P. Ffolliott. 1999. Fire effects on belowground sustainability: A review and synthesis. For. Ecol. Manage. 122:5171.
- Piatek, K.B., and H.L. Allen. 2001. Are forests in mid-rotation stands of loblolly pine (Pinus taeda) a sink for nitrogen and phosphorus? Can. J. For. Res. 31:11641174.
- Reuter, J.E., and W.W. Miller. 2000. Aquatic resources, water quality, and limnology of Lake Tahoe and its upland watershed. p. 215402. In D.D. Murphy and C.M. Knopp (ed.) The Lake Tahoe watershed assessment: Volumes 1 and 2. General Tech. Rep. PSW-GTR-178/176. USDA Forest Service Pacific Southwest Research Station, Albany, CA.
- Rustad, L.E., and C.S. Cronan. 1988. Element loss and retention during litter decay in a red spruce stand in Maine. Can. J. For. Res. 18:947953.
- Schimel, J.P., and M.K. Firestone. 1989. Inorganic N incorporation by coniferous forest floor material. Soil Biol. Biochem. 21:4146.
- Smethurst, P., and E. Nambiar. 1990. Distribution of carbon and nutrients and fluxes of mineral nitrogen after clear-felling a Pinus radiata plantation. Can. J. For. Res. 20:14901497.
- Stark, N. 1973. Nutrient cycling in a Jeffrey pine forest ecosystem. Montana Forest and Conservation Experiment Station, Univ. of Montana, Missoula, MT.
- Stephens, S.L., and J.J. Moghaddas. 2005. Experimental fuel treatment impacts on forest structure, potential fire behavior, and predicted tree mortality in a California mixed conifer forest. For. Ecol. Manage. 215:2136.
- Stohlgren, T.J. 1988. Litter dynamics in two Sierran conifer forests: II. Nutrient release in decomposing leaf litter. Can. J. For. Res. 18:11361144.
- Susfalk, R.B., and D.W. Johnson. 2002. Ion exchange resin based soil solution lysimeters and snowmelt solution collectors. Commun. Soil Sci. Plant Anal. 33(78):12611275.
- Williams, M., and J. Melack. 1991. Solute chemistry of snowmelt and runoff in an alpine basin, Sierra Nevada. Wat. Res. 27:15751588.
- Yavitt, J.B., and T.J. Fahey. 1986. Litter decay and leaching from the forest floor in Pinus contorta (Lodgepole pine) ecosystems. J. Ecol. 74:525545.
This article has been cited by other articles:

|
 |

|
 |
 
T.M. Loupe, W.W. Miller, D.W. Johnson, J.S. Sedinger, E.M. Carroll, R.F. Walker, J.D. Murphy, and C.M. Stein
Effects of Mechanical Harvest plus Chipping and Prescribed Fire on Sierran Runoff Water Quality
J. Environ. Qual.,
February 6, 2009;
38(2):
537 - 547.
[Abstract]
[Full Text]
[PDF]
|
 |
|