Figures
Abstract
Forested ecosystems in the southeastern United States are currently undergoing an invasion by the hemlock woolly adelgid (HWA). Previous studies in this area have shown changes to forest structure, decreases in canopy cover, increases in organic matter, and changes to nutrient cycling on the forest floor and soil. Here, we were interested in how the effects of canopy loss and nutrient leakage from terrestrial areas would translate into functional changes in streams draining affected watersheds. We addressed these questions in HWA-infested watersheds at the Coweeta Hydrologic Laboratory in North Carolina. Specifically, we measured stream metabolism (gross primary production and ecosystem respiration) and nitrogen uptake from 2008 to 2011 in five streams across the Coweeta basin. Over the course of our study, we found no change to in-stream nutrient concentrations. While canopy cover decreased annually in these watersheds, this change in light penetration did not translate to higher rates of in-stream primary production during the summer months of our study. We found a trend towards greater heterotrophy within our watersheds, where in-stream respiration accounted for a much larger component of net ecosystem production than GPP. Additionally, increases in rhododendron cover may counteract changes in light and nutrient availability that occurred with hemlock loss. The variability in our metabolic and uptake parameters suggests an actively-infested ecosystem in transition between steady states.
Citation: Northington RM, Webster JR, Benfield EF, Cheever BM, Niederlehner BR (2013) Ecosystem Function in Appalachian Headwater Streams during an Active Invasion by the Hemlock Woolly Adelgid. PLoS ONE 8(4): e61171. https://doi.org/10.1371/journal.pone.0061171
Editor: Sarah C. Davis, University of Illinois Urbana-Champaign/Energy Biosciences Institute, United States of America
Received: September 2, 2012; Accepted: March 7, 2013; Published: April 22, 2013
Copyright: © 2013 Northington et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was supported by a National Science Foundation award DEB-02218001 and DEB-0823293 from the Long Term Ecological Research Program to the Coweeta LTER Program at the University of Georgia. Any opinions, findings, conclusions, or recommendations expressed in the material are those of the authors and do not necessarily reflect the views of the National Science Foundation or the University of Georgia. Funding was also provided by a grant from the United States Forest Service. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Biological invasions have the potential to greatly affect ecosystem structure and processes. Alterations in compositional structure of ecosystems are one of the most direct effects of invaders, while changes in biogeochemical cycles or productivity are a secondary consequence [1], [2]. Significant modifications to carbon [3] and nitrogen [4], [5] inputs along with shifts in forest community composition [6], [7] have been noted throughout the extent of the infestations by both fungal [8] and insect invaders [9], [10], [11], [12]. Previous work in Coweeta [13] noted significant losses of riparian canopy cover and subsequent increases in light intensities and stream temperatures, all attributable to the influence of the hemlock woolly adelgid (HWA, Adelges tsugae).
Currently, forests spanning the eastern United States are being invaded by the HWA [14], which are limited by water availability and temperature [15]. Even so, their range may continue to expand with increases in annual warming [16]. HWA was first noted in southeastern forests in 2003 [17]. Within 5 years, there was 33% tree mortality [13], and currently there is a near total loss of hemlock trees [18]. The most direct effect of HWA appears to be increased canopy openness [13], [19], which is allowing for previously suppressed species such as oaks, hardwoods, and rhododendron (Rhododendron maximum) to become more abundant [6], [7]. Thus, an overstory becoming more dominated by deciduous trees will allow more light to reach both the forest floor and stream ecosystems during times of leaf-off [6], [13].
Concurrent with an increase in canopy openness will be inputs of large hemlock wood and needles to both the forest floor [4] and streams [13]. In-stream concentrations of nitrate have increased significantly in some watersheds with death of hemlocks resulting from the leaching of nitrogen [5], [20], in addition to increased annual discharge brought about due to a shift to a more deciduous watershed [21]. Thus, increases in light and nutrient availability may increase in-stream primary production [22]. Over the long term, nitrogen may eventually be retained in the forest floor due to immobilization of the nitrate by heterotrophic microbes colonizing the newly available litter resources [3], [4], [23].
Forest disturbance has been shown to have a significant impact on ecosystem processes in streams. As noted earlier, increases in canopy openness allows for greater light availability to streams in watersheds affected by HWA [13], [19]. A significant relationship between canopy openness and primary production has been noted in a variety of stream ecosystems spanning the globe [24], [25]. Earlier studies in Coweeta [26] and British Columbia [27] demonstrated significant increases to periphyton biomass in clearcut watersheds, mostly due to increases in light penetration.
Measures of watershed-scale processes, such as nutrient uptake and stream metabolism, are valuable tools to assess the role of invasive species as a disturbance within an ecosystem. Biogeochemical cycling in streams has been a key component of many ecosystem studies, as it integrates changes to nutrients over large scales of time and space into smaller, more measurable units of study [28], [29]. Additionally, relating carbon dynamics (fixation and mineralization) with nutrient demand helps to better assess the energetics of an ecosystem [30], [31], [32], especially one that is going through an active disturbance. Therefore, streams are ideal for measuring ecosystem response to a wide variety of disturbances, especially given the current ease of capturing changes to nutrients and metabolism over a range of time scales [33], [34]. In this study, we assessed the indirect effect of a current HWA infestation on ecosystem function in streams within watersheds where riparian hemlock is abundant. We hypothesized that changes in metabolism and nutrient uptake should occur because of a loss of riparian hemlock canopy. As the canopy around streams becomes more open, we predict that in-stream primary production should increase. Additionally, greater leaching of nutrients from hemlock-dominated watersheds should increase nutrient uptake by autotrophs and stimulate primary production.
Materials and Methods
Study Sites
This study was performed at Coweeta Hydrologic Laboratory in southwestern North Carolina (35°03′35″N, 83°25′51″W) as part of long-term research of the effects of HWA on forest canopy structure and ecosystem function in this region (e.g. [13]).
Previous work on ecosystem effects of HWA was conducted using 9 sites within the larger Coweeta Creek watershed; here, we chose a subset of five lower order streams (1st–2nd) used in a previous study in Coweeta [13]. The streams used in the present study are physically similar (Table 1) and have low background concentrations of important nutrients, such as inorganic nitrogen (Table 2). Hemlock basal area contribution to vegetated riparian corridors surrounding the streams varied from 28.4% (Hugh White Creek, Lower) to a high of 41.6% in Mill Branch [13]. We did not use reference non-hemlock sites due to the prevalence of this species in forests of this region.
Physical Stream Measures
Stream discharge (Q) at all sites was estimated using the sodium chloride slug method [35] prior to measures of metabolism and uptake (described below) each year. Stream wetted widths, depths, and cross sectional areas were also measured during each sampling period.
Light
Full methods for collecting light intensity and canopy openness data were described previously [13]. In short, light intensity over the study period was collected using HOBO Pendant data loggers (Onset Computer Corp., Bourne, ME, USA) mounted on 1-m tall posts every 10 m along each stream in the riparian zone. Data for relative light intensity (in lux) were recorded every 5 minutes since the beginning of the experiments in 2007. Here, we only report values for July light intensities in order to pair them with our measures of metabolism and nutrient uptake taken during that month each year. Annual patterns in light intensity for these streams have been reported in a previous study [13]. Canopy openness at each site was measured annually prior to leaf-out at permanent stream locations using a digital camera [13].
Metabolism
We monitored in-stream ecosystem respiration (ER) and GPP every July during the study period using the open-channel, one-station method (e.g. [36], [37]). Multi-probe sondes (Hach-Hydrolab ) were placed in each stream to record dissolved oxygen concentrations (mg L−1), oxygen saturation (%), temperature (°C), and specific conductance (µS cm−1) every two minutes for 36 hours. Sulfur hexafluoride (SF6) injections occurred simultaneously with metabolism measurements and ammonium injections (described below). Replicate water samples were injected into evacuated glass vials to allow for the headspace to accumulate SF6.
Headspace air was analyzed on a gas chromatograph (SRI 8610/9300 equipped with an ECD detector, SRI Instruments, Torrance, CA 90503) for relative abundance of SF6, and reaeration was estimated using the slope of the log-corrected loss of gas with distance downstream. Data for net ecosystem production (NEP) were calculated by the equation NEP = GPP – ER, where more positive values indicate an autotrophic system and negatives suggest more heterotrophy [38].
Uptake
We estimated ammonium (NH4+) uptake using the plateau injection method [39]. In order to not over-fertilize the stream (e.g. [40]), we attempted to raise in-stream NH4+ concentration by <50 µg L−1 over background. A co-injection of ammonium and chloride was added to each site at a known rate with a FMI metering-pump, until specific conductance of the stream (as measured with a YSI-30 handheld conductivity probe, Yellow Springs, OH) reached plateau. Replicate samples of stream water were then filtered through 0.7-µm GFF syringe filters into plastic sample bottles at 6 to 7 stations along the length of the stream reach. Samples were frozen prior to analysis.
Ammonium concentrations were determined using the phenate method [41] with a flow-injection analyzer (Lachat Quickchem 8500, Lachat Corporation, Loveland, CO). Nitrate and chloride were estimated using ion chromatography (Dionex DX500 ion chromatograph, Thermo Fisher Scientific, Inc., Waltham, MA, USA). The log-transformed loss of nutrient over stream length was used to derive an uptake length (Sw; [42]). Sw is the average distance traveled by a nutrient atom as it goes from inorganic to organic form [39]. The uptake length was then used to calculate areal uptake (U) [39]. We chose to use ammonium as our measure of nutrient uptake as it is more readily immobilized than nitrate in Coweeta streams [43], [44], [45]. Additionally, ammonium uptake has been used as a measure of nutrient cycling in numerous studies, contributing to long-term datasets at Coweeta [32], [43], [44].
Statistical Analyses
Changes in nutrient uptake parameters, GPP, ER, NEP, and July relative light intensities over time were analyzed using Regional Kendall tests for trend, which is a non-parametric test that examines monotonic trends in long-term data [39], [45]. Differences in uptake and metabolism were analyzed separately for year and site using Kruskal-Wallis tests, with post-hoc Wilcoxin multiple comparisons [46]. Non-parametric Pearson correlations were performed between functional parameters and stream physicochemical factors. Computations for the Regional Kendall tests were completed using the software described in [45]. All other analyses were performed on JMP v.9.0 statistical software (SAS Institute, Cary, NC).
Results
Light
Light intensities to streams significantly increased throughout the Coweeta watershed since 2006 [13], with canopy openness also increasing significantly in both sections of Hugh White Creek and Reynolds. In our study, which focused only on July, we see variable light intensities in the streams over time, with only small increases in 2010 (Fig. 1). Even though some sites (e.g. Mill) show an upward trend in light intensity, these changes were not statistically significant (Kendall τ = 0.07, p = 0.90).
Distribution of July light intensities (lux) at the five study sites in Coweeta Hydrologic Laboratory over the course of the study.
Nutrient uptake
Background levels of ammonium remained low over the course of the study period in all sites (CV = 0.25), while nitrate concentrations were greater and more variable (CV = 0.76, Table 2). Hugh White Creek (both upper and lower sites) and Cunningham had the most consistent nitrate levels during the study. Mill and Reynolds both had peak nitrate during 2008, with much lower concentrations in subsequent years (Table 2).
Both Sw and U were variable across sites and years (CV = 0.60). No overall trends in Sw (Kendall τ = 0, p = 1.0) or U (Kendall τ = 0.40, p = 0.10) were seen among sites over the course of this study. Uptake was significantly lower only in 2009, but Sw was statistically similar throughout the four years of the study (Fig. 2A–B, Table 3). Stream-specific changes in uptake parameters were only seen in Cunningham and Lower Hugh White Creek, both of which had significantly longer uptake lengths than Upper Hugh White Creek, Mill, and Reynolds (Fig. 2D–E).
Changes in ammonium uptake length (Sw) and areal uptake (U) for low-order streams in Coweeta Hydrologic Laboratory over the years of the study (A–B) and in each stream (C–D). Error bars represent 1 SE of the mean, and different letters represent significant differences (p<0.05) based on Wilcoxon post-hoc multiple comparison tests.
Uptake parameters generally related to physical, as opposed to chemical or metabolic factors (Table 3). Sw was negatively related to light (both canopy openness and light intensity) and positively associated with depth, while U showed no relationship with light (Table 3). Uptake was more strongly related to discharge, while Sw responded more strongly to stream size (e.g., width and depth; Table 3). The availability of inorganic nitrogen species appeared to not have a significant relationship with any uptake parameter.
Metabolism
Metabolic parameters were quite variable over stream locations and across years (CV = 0.60 for both GPP and ER). Levels of GPP were very low (Fig. 3A,D), ranging from 0.01 to 0.18 g O2 m−2 d−1 in Upper Hugh White and Reynolds Creek, respectively. ER ranged from a low of 0.4 g O2 m−2 d−1 in Lower Hugh White Creek in 2008 to a high of 5.4 g O2 m−2 d−1 in Mill Branch in 2010. Negative values of NEP indicated that all streams were net heterotrophic at the times of our measurements (Fig. 3).
Changes in gross primary production (GPP), ecosystem respiration (ER), and net ecosystem production (NEP) for low-order streams in Coweeta Hydrologic Laboratory over the years of the study (A–C) and in each stream (D–F). Error bars represent 1 SE of the mean, and different letters represent significant differences (p<0.05) based on Wilcoxon post-hoc multiple comparison tests.
As with uptake, no overall trends in metabolism were seen across sites over the course of this study: GPP (Kendall τ = 0.27, p = 0.27), ER (Kendall τ = 0.27, p = 0.29), NEP (Kendall τ = −0.27, p = 0.29). Strong differences in stream metabolism were seen in 2008 and 2010 (Fig. 3A–C), with NEP significantly lower in 2010 compared to 2008 (Fig. 3C). This effect was mostly driven by significant differences in respiration (Fig. 3B, Table 3), as GPP was not significantly different among years. No significant differences in metabolic parameters were found among sites over the course of the experiment (Fig. 3D–F).
There were no significant relationships between stream physicochemical variables and metabolic parameters (Table 3). Uptake parameters also had no significant associations with metabolism (Table 3). There appears to be a suggestive response of ecosystem respiration and NEP to changes in benthic organic matter availability (Table 3), although these relationships were only based on data for 2008.
Discussion
Primary Production
Contrary to our predictions, there has been no appreciable change to in-stream GPP, even with basin-wide increases in light availability to streams. Historically, Coweeta streams had low nutrient concentrations [47], [48] and have been shown to be nutrient-limited [26]. Our data demonstrate that this condition has not changed over the course of our study (Table 2), in spite of documented losses of nitrogen from terrestrial watersheds affected by HWA [5], [23]. Even if light becomes more available to stream ecosystems, autotrophic production is dependent on the availability of inorganic nutrients, especially N and P [49], [50], [51]. Further, nutrient limitation of primary production may only become apparent under conditions where light is more abundant [52].
Although we were unable to demonstrate trends in uptake over time, the generally short Sw and low overall uptake are consistent with previous studies [44], [53], [54], [55] suggesting that streams of this region are nutrient limited [26]. Significantly shorter Sw, coupled with similar levels of uptake in Upper Hugh White Creek, Mill, and Reynolds potentially indicate increases in N-immobilization due to hemlock mortality [18]. Even though hemlock detritus is a relatively low-quality resource, a previous study in Coweeta [53] suggested that the woody input from dying trees may provide a greater surface area for microbial colonization and nutrient immobilization as suggested by our data.
Light limitation of autotrophic production in these headwater streams may not be alleviated by the canopy loss due to HWA infestation. Previous work in Coweeta demonstrated significant annual increases in light intensity across the basin over the 4 years of their study [13]. Further, light limitation has been demonstrated in heavily shaded streams in Tennessee during the summer months after manual removal of riparian vegetation, where there was a significant increase in in-stream primary production [55]. Even so, this short-term experiment did not demonstrate any compositional changes to riparian plant communities in White Oak Creek, Tennessee. Therefore, shading (not species composition) was a more significant contributor to rates of in-stream primary production in this low order forested stream.
The increase in light that was evident from annual means [13] was not observed in July. Instead, July light has been relatively unaffected over the 4 years of this study (Figure 1). The greatest differences in canopy openness and light infiltration into the streams occur during the winter and early spring months of the year [13], but we do not have annual estimates for uptake and metabolism to pair with these annual patterns in light. Thus, we were only able to address relationships between light, metabolism, and nutrient uptake during the times when we have data for each of those factors (i.e. July).
While canopy opening due to HWA infestation does increase light availability, growth of other understory and riparian species may be stimulated. Significant, negative relationships between rhododendron and hemlock have been demonstrated previously [13], suggesting a possible suppression of rhododendron due to previously heavy hemlock cover. With the loss of hemlock canopy in watersheds affected by HWA, rhododendron appears to be re-establishing close to streams [3], [7], [56]. The lack of July trends in light data here also suggests that riparian re-growth of rhododendron is providing shading to the streams that was previously due to hemlock canopy, thus continuing to suppress primary production in our system.
Metabolic changes and respiration
Changes in GPP were not associated with metabolic change, while more significant changes to NEP did occur due to ER (Fig. 3, Table 3). Although forested headwater systems such as these are typically heterotrophic (e.g. [57]), there appeared to be movement towards even greater heterotrophy from 2008 to 2010 (Fig. 3). Although we did not have complete BOM data throughout our study, the significant relationship between BOM, ER, and NEP does suggest an even greater role for organic matter in controlling ecosystem metabolism (Table 3).
Large amounts of wood have been measured in these streams [13], which is a resource that may remain in the streams for decades [53], [58], [59]. Reynolds Branch, which had a large contribution of hemlock to overall riparian basal area [13] may be indicative of how streams in this region may respond to further loss of hemlocks. Small shifts in respiration (Fig. 3), possibly due to greater litter fall in Reynolds, combined with relatively short Sw (Fig. 2) may indicate a greater heterotrophic nutrient demand (e.g. [31], [60]) relative to Cunningham and HWC.
System-wide effects and future work
In Coweeta, the immediate effects of infestation by HWA (i.e. mass infestation and killing of hemlocks) may be characteristic of the “acute” stage of invasion [61]. As such, systems undergoing an active disturbance will not show immediate responses due to transitional stages that will occur as ecosystems move between stable states and nutrient retentive ability [56], [62]. We were unable to fully address changes to nutrient dynamics during times of peak light inputs into Coweeta streams, which occur in April [13]. A more thorough, annual assessment of nutrient uptake would be valuable in the future. Our initial predictions of changes to primary production, respiration, and nutrient uptake due to overstory canopy loss may be better addressed over annual time scales.
Future increases in hemlock detritus to streams will also increase standing stocks of organic matter due to the recalcitrance of needles [62] and generally slow breakdown of wood [20], [53]. Further measures of hemlock inputs and subsequent processing of particulates is needed. Additionally, woody inputs may ultimately change the retentive nature of streams [60], [63], thus affecting function in ways yet to be determined (e.g. [63], [64], [65]). These “chronic” effects of HWA invasion, where inputs from dead or dying hemlocks may lead to subsequent alterations of ecosystem structure and function, will take time to manifest in our system [61].
Conclusions
Even though large, significant changes to metabolism and nutrient uptake were not seen in our study, the inherent variability in the data was even more important. This study, and most others relating to HWA effects, occurred during an active infestation, with the variability most likely indicating that stream ecosystems in Coweeta are transitioning to new steady states. In-stream primary production did not increase over the course of the study, partially due to the consistent shading of streams during July by broad-leaved species. Net ecosystem production changes were mostly driven by increased respiration, likely related to changes in benthic organic matter. On the other hand, nutrient uptake showed no consistent patterns over time. The full extent of the impacts of HWA-mediated hemlock loss on stream ecosystems may not be known for decades.
Acknowledgments
We thank the USDA Forest Service Southern Research Station Coweeta Hydrologic Laboratory and the Coweeta LTER for logistical support. Additionally, we would like to thank all of the graduate and undergraduate students who helped in the field and laboratory. Also, thanks to J. Moore and an anonymous reviewer for comments on earlier drafts of this paper.
Author Contributions
Conceived and designed the experiments: JRW EFB BRN. Performed the experiments: RMN BMC. Analyzed the data: RMN. Wrote the paper: RMN JRW.
References
- 1. Ehrenfeld JG (2010) Ecosystem consequences of biological invasions. Annu Rev Ecol Evol Syst 41: 59–80.
- 2. Simberloff D (2011) How common are invasion-induced ecosystem impacts? Biol Invasions 13: 1255–1268.
- 3. Nuckolls AE, Wurzburger N, Ford CR, Hendrick RL, Vose JM, et al. (2009) Hemlock declines rapidly with hemlock woolly adelgid infestation: impacts on the carbon cycle of southern Appalachian forests. Ecosystems 12: 179–190.
- 4. Stadler B, Müller T, Orwig D (2006) The ecology of energy and nutrient fluxes in hemlock forests invaded by hemlock woolly adelgid. Ecology 87: 1792–1804.
- 5. Jenkins JC, Aber JD, Canham CD (1999) Hemlock woolly adelgid impacts on community structure and N cycling rates in eastern hemlock forests. Can J Forest Res 29: 630–645.
- 6. Small MJ, Small CJ, Dreyer GD (2005) Changes in a hemlock-dominated forest following woolly adelgid infestation in southern New England. J Torrey Bot Soc 132: 458–470.
- 7. Krapfl KJ, Holzmueller EJ, Jenkins MA (2011) Early impacts of hemlock woolly adelgid in Tsuga canadensis forest communities of the southern Appalachian mountains. J Torrey Bot Soc 138: 93–106.
- 8. Elliott KJ, Swank WT (2008) Long-term changes in forest composition and diversity following early logging (1919–1923) and the decline of the American Chestnut (Castanea dentata). Plant Ecol 197: 155–172.
- 9. Eshleman KN, Morgan RP, Webb JR, Deviney FA, Galloway JN (1998) Temporal patterns of nitrogen leakage from mid-Appalachian forested watersheds: Role of insect defoliation. Water Resour Res 34: 2005–2016.
- 10. Townsend PA, Eshleman KN, Welcker C (2004) Remote sensing of gypsy moth defoliation to assess variations in stream nitrogen concentrations. Ecol Appl 14: 504–516.
- 11. Lewis GP, Likens GE (2007) Changes in stream chemistry associated with insect defoliation in a Pennsylvania hemlock-hardwoods forest. For Ecol Manage 238: 199–211.
- 12. Riscassi AL, Scanlon TM (2009) Nitrate variability in hydrological flow paths for three mid-Appalachian forested watershed following large-scale defoliation. J Geophys Res-Biogeosci 114 .
- 13. Webster JR, Morkeski K, Wojculewski CA, Niederlehner BR, Benfield EF, et al. (2012) Effects of hemlock mortality on streams in the southern Appalachian Mountains. Am Midl Nat 168: 112–131.
- 14. Fitzpatrick MC, Preisser EL, Porter A, Elkinton J, Ellison AM (2012) Modeling range dynamics in heterogeneous landscapes: invasion of the hemlock woolly adelgid in eastern North America. Ecol Appl 22: 472–486.
- 15.
Ward JS, Montgomery ME, Cheah CASJ, Onken BP, Cowles RS (2004) Eastern hemlock forests: guidelines to minimize the impacts of hemlock woolly adelgid. NA-TP-03-04. U.S. Department of Agriculture, Forest Service, Northeastern Area State & Private Forestry, Morgantown, WV.
- 16. Paradis A, Elkinton J, Hayhoe K, Buonaccorsi J (2008) Role of winter temperature and climate change on the survival and future range expansion of the hemlock woolly adelgid (Adelges tsugae) in eastern North America. Mitig Adapt Strat Global Change 13: 541–554.
- 17. Elliott KJ, Vose JM (2011) The contribution of Coweeta Hydrologic Laboratory to developing an understanding of long-term (1934–2008) changes in managed and unmanaged forests. For Ecol Manage 261: 900–910.
- 18. Ford CR, Elliott KJ, Clinton BD, Kloeppel BD, Vose JM (2012) Forest dynamics following eastern hemlock mortality in the southern Appalachians. Oikos 121: 523–536.
- 19. Orwig DA, Foster DR (1998) Forest response to the introduced hemlock woolly adelgid in southern New England, USA. J Torrey Bot Soc 125: 60–73.
- 20. Ellison AM, Bank MS, Clinton BD, Colburn EA, Elliott K, et al. (2005) Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Front Ecol Env 3: 479–486.
- 21. Ford CR, Vose JM (2007) Tsuga canadensis (L.) Carr. mortality will impact hydrologic processes in southern Appalachian forest ecosystems. Ecol Appl 17: 1156–1167.
- 22. Rowell TJ, Sobczak WV (2008) Will stream periphyton respond to increases in light following forecasted regional hemlock mortality? J Freshw Ecol 23: 33–40.
- 23. Knoepp JD, Vose JM, Clinton BD, Hunter MD (2011) Hemlock infestation and mortality: impacts on nutrient pools and cycling in Appalachian forests. Soil Sci Soc Am J 75: 1935–1945.
- 24. Lamberti GA, Steinman AD (1997) A comparison of primary production in stream ecosystems. J N Am Benthol Soc 16: 95–104.
- 25. Clapcott JE, Barmuta LA (2010) Forest clearances increases metabolism and organic matter processes in small headwater streams. J N Am Benthol Soc 29: 546–561.
- 26. Lowe RL, Golladay SW, Webster JR (1986) Periphyton response to nutrient manipulation in streams draining clearcut watersheds. J N Am Benthol Soc 5: 221–229.
- 27. Kiffney PM, Richardson JS, Bull JP (2003) Responses of periphyton and insects to experimental manipulation of riparian buffer width along forest streams. J App Ecol 40: 1060–1076.
- 28. Bunn SE, Davies PM, Mosisch TD (1999) Ecosystem measures of river health and their response to riparian and catchment degradation. Freshw Biol 41: 333–345.
- 29. Young RG, Matthaei CD, Townsend CR (2008) Organic matter breakdown and ecosystem metabolism: functional indicators for assessing river ecosystem health. J N Am Benthol Soc 27: 605–625.
- 30. Fellows CS, Valett HM, Dahm CN, Mulholland PJ, Thomas SA (2006) Coupling nutrient uptake and energy flow in headwater streams. Ecosystems 9: 788–804.
- 31. Hall RO, Tank JL (2003) Ecosystem metabolism controls nitrogen uptake in streams in Grand Teton National Park, Wyoming. Limnol Oceanogr 48: 1120–1128.
- 32. Tank JL, Meyer JL, Sanzone DM, Mulholland PJ, Webster JR, et al. (2000) Analysis of nitrogen cycling in a forested stream during autumn using a 15N-tracer addition. Limnol Oceanogr 45: 1013–1029.
- 33. Mulholland PJ, Webster JR (2010) Nutrient dynamics and the role of J-NABS. J N Am Benthol Soc 29: 100–117.
- 34. Stanley EH, Powers SM, Lottig NR (2010) The evolving legacy of disturbance of stream ecology: concepts, contributions, and coming challenges. J N Am Benthol Soc 29: 67–83.
- 35.
Gordon ND, McMahon TA, Finlayson BL, Gippel CJ, Nathan RJ (2004) Stream Hydrology: an Introduction for Ecologists, 2nd edition. New York, NY: John Wiley & Sons, Inc. 444 p.
- 36.
Bott TL (2006) Primary productivity and community respiration. In: Hauer FR, Lamberti GA, editors. Methods in Stream Ecology, 2nd edition. New York, NY: Elsevier. pp. 663–690.
- 37.
Grace M, Imberger S (2006) Stream metabolism: performing and interpreting measurements. Water Studies Centre, Monash University, Murray Darling Basin Commission and New South Wales Department of Environmental and Climate change. Available: http://www.sci.monash.edu.au/wsc/docs/tech-manual-v3.pdf. Accessed 2008 Nov 10.
- 38.
Allan JD, Castillo MM (2007) Stream Ecology, 2nd edition. Dordrecht, The Netherlands: Springer. 450 p.
- 39.
Webster JR, Valett HM (2006) Solute dynamics. In: Hauer FR, Lamberti GA, editors. Methods in Stream Ecology, 2nd edition. New York, NY: Elsevier. pp. 169–186.
- 40. Mulholland PJ, Tank JL, Webster JR, Bowden WB, Dodds WK, et al. (2002) Can uptake length in streams be determined by nutrient addition experiments? Results from an interbiome comparison study. J N Am Benthol Soc 21: 544–560.
- 41.
American Public Health Association, American Water Works Association, Water Environment Federation (APHA, AWWA, WEF) (2005) Method 4130: Inorganic Nonmetals by Flow Injection Analysis In: Eaton AD, Clesceri LS, Rice EW, Greenberg AE, editors. Standard Methods for the Examination of Water and Wastewater, 21st edition. Washington, DC: American Public Health Association. pp. 4–13.
- 42. Stream Solute Workshop (1990) Concepts and methods for assessing solute dynamics in stream ecosystems. J N Am Benthol Soc 9: 95–119.
- 43. Webster JR, D'Angelo DJ, Peters GT (1991) Nitrate and phosphate uptake in streams at Coweeta Hydrologic Laboratory. Verh Internat Verein Limnol 24: 1681–1686.
- 44. Webster JR, Tank JL, Wallace JB, Meyer JL, Eggert SL, et al. (2000) Effects of litter exclusion and wood removal on phosphorous and nitrogen retention in a forest stream. Verh Internat Verein Limnol 27: 1337–1340.
- 45. Helsel DR, Frans LM (2006) Regional Kendall Test for trend. Environ Sci Technol 40: 4066–4073.
- 46.
Zar JH (1999) Biostatistical Analysis, 4th edition. Upper Saddle River, NJ: Prentice Hall. 960 p.
- 47.
Swank WT, Waide JB (1988) Characterization of baseline precipitation and stream chemistry and nutrient budgets for control watersheds. In: Swank WT, Crossley DA, editors. Ecological Studies, Vol. 66: Forest Hydrology and Ecology at Coweeta, New York, NY: Springer-Verlag. pp. 52–79.
- 48. Swank WT, Vose JM (1997) Long-term nitrogen dynamics of Coweeta forested watersheds in the southeastern United States of America. Global Biogeochem Cy 11: 657–671.
- 49. Elwood JW, Newbold JD, Trimble AF, Stark RW (1981) The limiting role of phosphorous in a woodland stream ecosystem: effects of P enrichment on leaf decomposition and primary producers. Ecology 62: 146–158.
- 50. Grimm NB, Fisher SG (1986) Nitrogen limitation in a Sonoran Desert stream. J N Am Benthol Soc 5: 2–15.
- 51. Elser JJ, Bracken MES, Cleland EE, Gruner DS, Harpole WS, et al. (2007) Global analysis of nitrogen and phosphorous limitation of primary producers in freshwater, marine, and terrestrial ecosystems. Ecol Lett 10: 1135–1142.
- 52. Larned ST (2010) A prospectus for periphyton: recent and future ecological research. J N Am Benthol Soc 29: 182–206.
- 53.
Morkeski K (2007) In-stream hemlock twig breakdown and effects of reach-scale twig additions on Appalachian headwater streams. MS Thesis. Virginia Polytechnic Institute and State University.
- 54. Valett HM, Thomas SA, Mulholland PJ, Webster JR, Dahm CN, et al. (2008) Endogenous and exogenous control of ecosystem function: N cycling in headwater streams. Ecology 89: 3515–3527.
- 55. Hill WR, Ryon MG, Schilling EM (1995) Light limitation in a stream ecosystem: responses by primary producers and consumers. Ecology 76: 1297–1309.
- 56. Roberts SW, Tankersley R Jr, Orvis KH (2009) Assessing the potential impacts to riparian ecosystems resulting from hemlock mortality in Great Smoky Mountains National Park. Environ Manage 44: 335–345.
- 57. Mulholland PJ, Fellows CS, Tank JL, Grim NB, Webster JR, et al. (2001) Inter-biome comparison of factors controlling stream metabolism. Freshw Biol 46: 1503–1517.
- 58. Wallace JB, Webster JR, Egger SL, Meyer JL, Siler ER (2001) Large woody debris in a headwater stream: Long-term legacies of forest disturbance. Int Rev Hydrobiol 86: 501–513.
- 59. Warren DR, Kraft CE (2008) Dynamics of large wood in an eastern US mountain stream. For Ecol Manage 256: 808–814.
- 60. Roberts BJ, Mulholland PJ (2007) In-stream biotic control on nutrient biogeochemistry in a forested stream, West Fork of Walker Branch. J Geophys Res-Biogeosci 112 .
- 61. Strayer DL, Eviner VT, Jeschke JM, Pace ML (2006) Understanding long-term effects of species invasions. Trends Ecol Evol 21: 645–651.
- 62. Valett HM, Crenshaw CL, Wagner PF (2002) Stream nutrient uptake, forest succession, and biogeochemical theory. Ecology 83: 2888–2901.
- 63. Bilby RE, Likens GE (1980) Importance of organic debris dams in the structure and function of stream ecosystems. Ecology 61: 1107–1113.
- 64. England LE, Rosemond AD (2004) Small reductions in forest cover weaken terrestrial-aquatic linkages in headwater streams. Freshw Biol 49: 721–734.
- 65. Warren DR, Bernhardt ES, Hall RO, Likens GE (2007) Forest age, wood and nutrient dynamics in headwater streams of the Hubbard Brook Experimental Forest, NH. Earth Surf Proce Land 32: 1154–1163.