Data Catalog Summary Pages - Ecology at Coweeta LTER

Study Number:	3062
Project Title:	Nutrient effects on a detritus-based stream ecosystem.
Investigator(s):	A D Rosemond E-Mail \| Tel. 706.542.3903 J B Wallace E-Mail \| Tel. 706.542.7886 \| Biographical Sketch K Suberkropp E-Mail \| Tel. 205.348.1795 P J Mulholland E-Mail \| Tel. 423.574.7304
Affiliated Institution(s):	University of Georgia, University of Alabama, Oak Ridge National Laboratory
Address:	Amy D Rosemond Institute of Ecology University of Georgia Athens, Georigia 30602
Study Category:	Aquatic
Project Type:	Type One
Study Period:	07/1998 - 07/2003
Notes:
Funding Source(s):	NSF Ecosystems Studies Program Award #9806610 (Text Version) “Nutrient Effects on a detritus-based stream ecosystem”
Abstract:	To determine the effects of nutrient enrichment in a detritus-based stream, we have continuously enriched a headwater stream with N and P for 4.5 years and quantified the response at different scales of stream structure and function: microbial, metazoan, and ecosystem. The effects of enrichment in the treatment stream (Watershed 54, WS54) were compared to conditions in the reference stream (Watershed 53, WS53) and one year of comprehensive baseline sampling was conducted in both streams prior to enrichment. Our first objective was to assess the effects of nutrient enrichment on growth and production of organisms. Our second objective was to determine the effects of nutrients on the fate of carbon resources, which were altered via increased microbial metabolism and invertebrate consumption. Most data were analyzed using Randomized Intervention Analysis (RIA)(Carpenter et al. 1989), which was based on pre-treatment differences between the two streams to differences following nutrient addition. This study was the first to determine the long-term effects of nutrient enrichment in a system in which organisms derive most of their carbon from detritus or detritus-based prey. Nutrient addition profoundly affected the treatment stream, with effects propagating from microbes to ecosystem-level effects. Below are specific results. EFFECTS ON ORGANISMS Nutrient addition resulted in an increase in microbial biomass and production. We observed dramatic increases in leaf-associated fungal production and biomass and significant effects of nutrient addition on leaf-associated bacterial production, but not on bacterial biomass. Whereas production of fungi exhibited an overall average 211% increase in response to nutrient addition in year 2 of enrichment, bacterial production only increased by 60% compared to the control stream. The mean number of fungal species detected per sample date before enrichment and in the reference stream was 12-13 but this increased to 18.5 species during enrichment. Primary producers showed some response to nutrient addition, despite light limitation. Contrary to our predictions, we observed a significant increase in algal biomass as measured by chlorophyll a, but the response was largely limited to high light months. There was a significant effect of enrichment on biofilm AFDM, but without the seasonal effect seen with chlorophyll a. There was no significant effect on cell densities of algae. There has been no apparent change in algal community composition as a result of enrichment. Production and biomass of invertebrates increased in response to nutrient addition. There has been a significant effect of enrichment on invertebrate biomass measured monthly and invertebrate production in the treatment stream was 153% higher compared to trends in the control stream after 1 year. Leaf litter standing crop in the stream is declining. Long-term data indicate a significant regression between benthic leaf litter standing crop and invertebrate production. Data from yr. 1 indicate a much higher production in the nutrient addition stream than would be predicted from leaf litter standing crop. Our longer term predictions for invertebrate biomass in the treatment stream include a decline in production compared to the response we observed after 1 yr of enrichment, based on the decline in benthic leaf litter. Growth rates of invertebrates were enhanced by nutrient addition. We measured increased growth rates of invertebrates with short aquatic life cycles, but did not observe increased growth rates on invertebrates with longer life cycles. Chironomidae, the dominant collector/gatherers, grew ca. 30% faster in the enriched stream than in the reference stream among all size classes. Tallaperla maria, a long-lived leaf-eating stonefly, did not respond to enrichment in terms of daily growth rates. We have data on the C, N, and P content of consumers showing that some taxa have much lower body P content than others (i.e., Tallaperla maria 0.5%, Chironomidae . 1.1%). Lower body P content generally suggests a lower P requirement for growth. Thus, Tallaperla maria, and other taxa with lower body P, may not be nutrient limited. Rates of organic matter breakdown were accelerated by nutrient addition. We observed increased breakdown of rhododendron and red maple leaves as well as wood. Rates of leaf decay exhibited a ca. 70% increase in the treatment stream relative to the control after 1 year of enrichment, and a ca. 200% increase in the 2nd year of enrichment. Breakdown rates with nutrient enrichment for both leaf types were outside of the 99% confidence interval of long-term means. Rates of decay of wood veneers exhibited 477 and 795% increases in the treatment stream relative to the control after 1 and 2 years of enrichment, respectively. Organic matter standing crop and export were affected by nutrient addition. We have observed dramatic reductions in leaf litter standing crop in the treatment stream, as well as increased export of fine particulate organic matter (significant effect on seston yield via RIA). There was no effect of nutrient addition on dissolved organic carbon. Whole-stream metabolism was affected by nutrient addition. Measurements of whole-stream gross primary productivity (GPP) and respiration (R) were highly variable seasonally. Contrary to our prediction that there would be a greater response in R than in GPP, effects on GPP were greater than on R. This result, however, was based on a very strong positive response of GPP to nutrient addition in April when GPP rates were highest, particularly in yr 2, and less response of GPP at other times of the year. The positive effect of nutrient addition on R was more consistent throughout the year. Food quality of detritus, as measured by C:N and C:P increased with nutrient enrichment. We have found that P content of leaf litter is much higher in the enriched stream than in the reference stream (i.e., lower C:P and higher %P). We have not seen any remarkable changes in %N or C:N of leaf litter in the enriched stream, suggesting that microbes demand P more than N. Respiration on detrital resources increased in response to nutrient addition. Yes, we observed increased respiration on both leaves (50-70% stimulation) and wood (300-350% stimulation) in response to nutrient addition.
Location(s), Described:	Watersheds 53 and 54
Location(s), Download GPS:	ArcView Shape Files (shp.): UTM, NAD83, Zone 17
Location(s), Online Map(s):	Online Map
Methods/Experimental Design:	The treatment stream has been continuously enriched since July 2000. During this time period, we have maintained a sampling regime to assess the effects of nutrient enrichment on ecosystem metabolism, including microbial and invertebrate production and community structure, detrital quality and quantity, etc. We have successfully maintained elevated concentrations of N and P along a 190m reach of WS54 for the past 4.5 years with very few days (ca. 15-20) when no nutrients were being added to the stream. Overall mean and standard error of (NO3+NO2 )-N, NH4 -N, and SRP in the treatment stream in yr 1 of enrichment was 180 (20), 78 (14) and 37 (5), respectively and 380 (107), 123 (34) and 45 (14) µg/L, respectively in yr 2 of enrichment. These concentrations compare to longterm means of inorganic nitrogen <10 µg/L and SRP <5 µg/L. Our nutrient delivery system includes a reservoir at the top of the reach to which we add a concentrated nutrient salt solution (NH4NO3, K2HPO4, KH2PO4, at a 11:1 N:P atomic ratio). This solution is pumped into an irrigation line that has stream water flowing through it, lies in the stream, and the nutrient solution drips out of sprinkler heads along the length of the reach. We are using a solar-charged, battery-operated continuous-flow pump (LMI High Efficiency DC Powered Metering Pump). The pump receives a signal from an ISCO flow meter for every 50 L that goes through the weir, to give us flow-proportional dosing. No dosing is made during hi-flow storm events. The reference and treatment streams are tributaries of Shope Fork creek, and because of the ~100-fold dilution of the treatment stream into Shope Fork, there is no significant downstream enrichment. Mean annual fungal biomass and production was determined by methods described in Suberkropp (1997) using rates of 14C-acetate incorporation into ergosterol to estimate fungal growth rates (Newell and Fallon, 1991, Gessner and Newell 1997). Ergosterol extraction followed Newell et al. (1988), modified by Suberkropp and Weyers (1996). Leaf material was homogenized to release bacteria (Baldy et al, 1995) and bacterial biomass was estimated from direct microscopic counts (DAPI stain, epifluorescence) and measurements of representative cells to calculate biovolume (Weyers and Suberkropp, 1996). Bacterial production was estimated from rates of incorporation of 3H-leucine into protein (Kirchman 1993, Thomaz and Wenzel 1995, Weyers and Suberkropp 1996), modified by extracting protein with an initial treatment of KOH (Moran and Hodson 1992, Newell et al 1995). On a monthly basis, 5 transects from each stream were selected at random. From each transect, 5 representative leaves were selected, and 7 leaf discs were cut from each leaf. Fom each leaf, 1 disk was fixed in buffered formalin for bacterial counts, 2 were used to determine leucine incorporation into protein, 1 was dried and ashed to determine AFDM, and 2 were used to determine rates of microbial respiration. Respiration was measured by monitoring changes in oxygen concentration using a YSI Model 5100 DO meter equipped with a Clark-type oxygen electrode in 30ml sample chambers. Additionally, 300ml of streamwater was filtered (5 µm membranes, 47mm dia) at streamside, and filters were examined through microscopy to determine the numbers and species composition of aquatic hyphomycete spores. Biomass and production of periphyton was measured using AFDM and chlorophyll a on ceramic tiles, using the methods described in Rosemond (1993). Ceramic tiles were first placed in the stream in July 1998, and they have been sampled monthly. Breakdown of leaf litter was also determined by methods previously employed in these streams (Cuffney et al, 1990). Thirty litter bags each of two different species of leaf, Acer rubrum and Rhododendron, were placed in the stream every December, and retrieved over a 250-d period. The retrieved leaves were then rinsed and dried, and invertebrates rinsed from the leaves were preserved in Kahle’s solution for identification and measurement. The dried leaves were then measured for AFDM remaining. Breakdown of wood veneers (Quercus sp.) were made according to Tank and Webster (1998). Benthic organic matter and leaf detritus standing crop was measured monthly in both streams. CBOM and FPOM standing stock was collected using a stratified random sampling design. Mixed substrates (silt to cobble sized, including debris dams) were sampled (4 replicates per stream per month) using a 400cm² benthic corer to a depth of 10cm, where possible. 225 cm² rock outcrop samples (3 per stream per month) were made as described by Lughtart and Wallace (1992). Organic material was elutriated from these samples and preserved in Kahle’s solution for invertebrate identification and measurement. Leaf litter standing crop was measured at random 20cm wide transects within each 10m interval along the entire reach of each stream. The leaves were wet weighed, and subsamples were dried and ashed for AFDM. Total export from the stream was measured in the form of coarse transported organic material (CTOM), fine particulate organic material (FPOM), and dissolved organic carbon (DOC). CTOM was continuously collected for the entire stream in a 4-mm mesh trap, 5m above the flume. This trap was bank-to-bank, in order to capture all CTOM. The CTOM was collected once a month, and dried, sorted, weighed, and ashed for AFDM. Continuous measurement of FPOM export was made using Coshocton proportional samplers attached to the flumes on each stream. The samplers divert 0.6% of discharge into a series of three 100-L settling barrels where all but the finest particles settle to the bottom. These sediments are then subsampled for AFDM and %ash measurements (Cuffney and Wallace, 1988). The 0.45 µm to 4 mm size range used is larger than the typical 0.45 µm to 1 mm particle size and is necessary to accommodate the opening slot on the Coshocton proportional sampler. These data were compared with ongoing studies in the reference stream (C 53) where we have 18 years of continuous FPOM export. Water samples were taken bi-weekly at the headwater seep and at each flume in each stream to be measured for DOC. DOC was measured on a Shimadzu TOC-5000a and later on a Shimadzu TOC-V Total Organic Carbon Analyzer, using their platinum-catalyst aided combustion analysis technique, by the UGA Institute of Ecology’s Analytical Chemistry Laboratory (ACL). Instream generation of DOC was determined as the difference between concentrations at the flume and at the seep. We also have an extensive Project of biweekly DOC measurements before the study (12 years for the reference stream and 5 years for the treatment stream) for comparative analysis of annual and seasonal DOC production in the streams. Additionally, CTOM, FPOM, leaf pack leaves, benthic invertebrates, and periphyton were all analyzed for total C and total N on a monthly or seasonal basis. This analysis was performed with a Carlo Erba NA1500 micro-Dumas combustion elemental assay, also by the ACL.
Sampling Frequency:	Continuous: Stream stage and flow Leaf Pack decomposition Wood veneer decomposition Biweekly: Water Chemistry- nitrate, ammonia, phosphate, DOC, Temperature Fine Particulate Organic Matter- AFDM, C:N Coarse Transported Organic Material- AFDM, C:N Monthly: Benthic standing crop- FBOM, invertebrates, AFDM, C:N Leaf Litter standing crop- dry weight, AFDM, C:N Litter Fall- AFDM, C:N Microbial production, biomass, and respiration.
Data Columns:	Nutrients: all data expressed as milligrams per Liter Date WS54 - Watershed 54 NO₃ - Nitrate NH₄ - Ammonium SRP - Soluble Reactive Phosphate WS53 - Watershed 53 NO₃ - Nitrate NH₄ - Ammonium SRP - Soluble Reactive Phosphate WS54 upstream of drip - Watershed 54 upstream of drip NO₃ - Nitrate NH₄ - Ammonium SRP - Soluble Reactive Phosphate Temperature: degrees Celsius (C) Date WS53 WS54 Daily Mean Discharge: Liters and Liters per second Date WS53 mean daily_(L/s) ttl daily_(L) WS54 mean daily_(L/s) ttl daily_(L) Ratio
Publications:	None generated.
Data Restrictions:	Users must adhere to the Coweeta LTER Data Policy.
Metadata:	EML Format (XML Schema) \| Information about EML
Data Downloads:	Nutrients: Microsoft® Excel (.xls) \| Text Comma Delimited (.csv) \| DBase (.dbf) Temperature: Microsoft® Excel (.xls) \| Text Comma Delimited (.csv) \| DBase (.dbf) Daily Mean Discharge: Microsoft® Excel (.xls) \| Text Comma Delimited (.csv) \| DBase (.dbf)