Methods

See Oliver et al. (submitted) and Soranno et al. (2015) for details on sources of data, methods of collection, and derivation of parameters
Oliver S.K., Collins S.M., Soranno P.A., Wagner T., Stanley E.H., Jones J.R., Stow C.A., Lottig N.R. Unexpected stasis in a changing world: Lake nutrient and chlorophyll trends since 1990. Submitted to Global Change Biology.
Soranno P.A., Bissell E.G., Cheruvelil K.S., Christel S.T., Collins S.M., Fergus C.E., Filstrup C.T., Lapierre J.-F., Lottig N.R., Oliver S.K., Scott C.E., Smith N.J., Stopyak S., Yuan S., Bremigan M.T., Downing J.A., Gries C., Henry E.N., Skaff N.K., Stanley E.H., Stow C.A., Tan P.-N., Wagner T., and Webster K.E. 2015. Building a multi-scaled geospatial temporal ecology database from disparate data sources: fostering open science and data reuse. Gigascience 4: 28. doi: 10.1186/s13742-015-0067-4 .

Methods

The complete methods for this manuscript are described in the manuscript: Rose, K.C., S.R. Greb, M. Diebel, and M.G. Turner. Annual precipitation as a regulator of spatial and temporal drivers of lake water clarity. Ecological Applications.

Methods

Limnological water chemistry samples were collected through individual monitoring programs carried out or overseen by state agencies. Water chemistry analyses were performed using standard methods by individual labs. Methods for integrating the disparate state datasets are described in detail in Soranno et al. 2015 Building a multi-scaled geospatial temporal ecology database from disparate data sources: fostering open science and data reuse, GigaScience20154:28 DOI: 10.1186/s13742-015-0067-4

Methods

We developed a database of water quality interventions by government agencies in the Yahara Watershed. Interventions were included if they were a) publicly funded and implemented, b) land-based, c) implemented within the 5-year period 2007 to 2012, and d) aimed to reduce nutrient (phosphorus and nitrogen) and sediment runoff to surface waters as a primary or secondary goal. Our criteria excluded interventions implemented directly in the water. They also excluded work by non-profit watershed groups and for-profit companies.Interventions were categorized by type of conservation tool: regulation or standard; incentive (grant and cost-share programs); direct management (including public management actions and engineered practices); and acquisition (land conserved through fee simple acquisition or conservation easement). Interventions were next categorized by which government level (or multiple levels) of government were involved in rulemaking and implementation (Table 1). We defined the rulemaking level of government as that which created the standard or wrote the law, and the implementing level of government as that which made field-level decisions, negotiated with landowners, and monitored practices. If the intervention was a grant given to private recipients, the implementing agencies were considered those that supervised grant implementation.We mapped policy interventions in ArcGIS (version 10.1). The goal of mapping was to determine the extent and overlap of interventions throughout the watershed and the agency responsible for establishing and implementing the policies. Public acquisitions of conservation land were mapped and categorized by the government level acquiring the parcel or parcels. Incentive programs – grants and cost-share – were mapped for the parcels where the incentive program was applied from 2007-2012. For federal Farm Bill Natural Resources Conservation Service (NRCS) conservation programs, for instance, the farm parcels of the cost-share recipients were mapped with publicly available data or by matching recipient names with parcel ownership records. Regulatory programs were mapped according to each statutes definition. For example, Wisconsins shoreland zoning ordinances were mapped as the area in the 300 meter buffer around rivers or streams and 1000 meter buffer of lakes or ponds, using the Wisconsin DNR water body base layer. Regulations were represented by specific permit area when permit data were available, such as farms with county winter manure-spreading permits.Regional water quality experts validated the interventions list and map. Reviewers included a regional planner, two municipal administrators, a commissioner on the County Lakes and Watershed Commission, a County water conservationist, a lawyer for an environmental non-profit, and the director of a Wisconsin soil and water conservation organization. Through this process we added several interventions and clarified the mapping rules. Analyses were conducted on 35 of 41 interventions that could be represented spatially through publicly available data. The most significant unmapped intervention was nutrient management planning, for which the County office did not have spatial data. We estimated the percent land area covered by each intervention by subwatershed. The Yahara Watershed was divided into 300 subwatersheds based on a recent modeling effort that delineated 200 subwatersheds in the upper Yahara Watershed (Montgomery Assoc., 2011) and our delineation of 100 comparably-sized subwatersheds based on a Digital Elevation Model in the lower Yahara Watershed. We then calculated the percentage of each subwatershed covered by each of the 35 interventions. The percentage of land covered by every intervention within a subwatershed was then summed to get a cumulative percent coverage. This ranged from 0 to a possible 3,500 for each subwatershed. The total percent intervention coverage is shown in heat maps depicting low to high policy coverage by subwatershed, created in ArcGIS.We categorized subwatersheds as urban or rural in order to compare coverage of interventions. Subwatersheds were classified as urban if the developed land cover classes were 50percent or more of land area, based on 2010 National Land Cover Data, which resulted in 83 urban (28percent) and 217 rural (72percent) subwatersheds. We conducted a Welch 2-sample t-test to determine whether cumulative percent area of interventions differed significantly for urban and rural subwatersheds. The untransformed cumulative percent area data were consistent with assumptions of normality and were not improved by an ArcSin transformation (sometimes used with percentage data), so we report the t-test with untransformed data.We compared intervention locations with total phosphorus yields (kilograms phosphorus per hectare per year) for the 200 subwatersheds modeled for the year 2008 with the Soil and Water Assessment Tool (SWAT). The Montgomery and Associates SWAT model is widely used by policymakers in the watershed. The subwatersheds with the highest nutrient yields are consistent with earlier models and measurements conducted for conservation planning (Lathrop, 2007).A Pearsons product-moment correlation matrix compared interventions with modeled phosphorus yield by subwatershed, calculated in R (version 3.0.1). We correlated phosphorus yields with cumulative percent intervention coverage by municipal, county, state, and federal governments in both their rulemaking and implementation capacities. We also compared the correlation of phosphorus yields with intervention coverage for each type of intervention tool. Interventions were also grouped by whether they targeted agricultural or nonagricultural activities. These correlations give a proxy measure of whether public interventions target areas of concern for watershed nutrient reduction.

Methods

To control for sampling methodology and allow comparisons among native and invasive species, we only included data where both invasive and native species from a taxonomic group were sampled using the same methods across multiple sites. Exceptions were made to include rusty crayfish (Orconectes rusticus) in its native range and zebra mussel (Dreissena polymorpha) data. Native rusty crayfish data were obtained from (Jezerinac 1982). Zebra mussel data were mainly obtained from a meta-analysis (Naddafi et al. 2011) which compiled data from 55 European and 13 North American sites from 1959-2004. Additional densities from North America were compiled from multiple primary literature sources (Table S3). All zebra mussel records were presented as number per m2 and are from their invaded range; we did not include native mussel data.Crayfish data were obtained from multiple sources. Crayfish were collected in Wisconsin, USA during summers of 2002-2010 from lakes in the Northern Highlands Lake District following their protocol for crayfish collection. Crayfish were sampled in Wisconsin streams tributary to Lake Michigan from 2007-2010 using 10 gee-style minnow traps per site baited with chicken livers and set overnight. Swedish crayfish were sampled using 30 minnow traps baited with frozen fish in lakes and streams of southern Sweden from 2001-2003 as described in (Nystrom et al. 2006). Washington crayfish were collected from 100 lakes in the Puget Sound Lowlands region of Washington State, USA between 2007 and 2009 from mid-June to early October of each year. At each lake, the investigators set 20 minnow traps baited with fish-based dog food. Traps were deployed in four clusters of five traps each and recovered the following day. All crayfish densities are presented as number per trap per day, with the exception of native range rusty crayfish data, which were reported as number per site (Jezerinac 1982) and excluded from all comparisons that depend on sampling units.Wisconsin fish data were collected from streams throughout the state from 2005-2010 using either a backpack or towboat electrofisher with pulsed DC current in wadeable (less than1m depth) streams for a minimum of 15 minutes. For Wisconsin trout species, locations sampled within 10 years following a stocking event of that species were excluded. Lamprey data were collected from 2008-2010 from Great Lakes tributaries using backpack electrofishers following standardized methods as a part of the sea lamprey assessment program of the United States Fish and Wildlife Service and Department of Fisheries and Oceans, Canada. North American fish densities are presented as number per minute of sampling. Swedish fish data were collected using backpack electrofishing between 1980 and 2010 from streams in Vasterbotten county, northern Sweden, and were obtained from the Swedish Electrofishing REgister (SERS), www.fiskeriverket.se, and are reported as number per 100 m of stream.Snail data were collected in 2006 from lakes in the Northern Highlands Lake District in Wisconsin as described by (Solomon et al. 2010), and densities are presented as number per two m2. Aquatic plant data were collected using a systematic grid-based point-intercept sampling methodology to record macrophyte frequency of occurrence in 242 Wisconsin lakes from 2005-2008. Aquatic plant presence absence was recorded from a boat using a double-sided rake sampler at each point on a sampling grid as described in (Mikulyuk et al. 2010). Density data are presented as proportion of sites within lake littoral zone where a species was present.For all data, if multiple records existed from the same location, we used the most recent record. If replicate samples existed within the same site on the same sampling date, the mean value was used.

Methods

SamplingAt each location, the extent of the peatland basin was examined visually using the soils map and a long axis was defined as the longest linear stretch of peat, while a short axis was defined perpendicular to the long axis. The sample area of a given site was defined as the entire peatland basin (full basin site, N = 11) if the length of the long axis was 800 m or less. For larger peatlands, the site was defined as an area with width 150-200 m and length 400-600 m (partial basin site, N = 10) extending outward from one edge of the peatland and if possible crossing the entire short axis. Peat depth was measured throughout the area on a regular grid at intervals varying from 20-90 m depending upon the size of the site. In addition, vegetation was surveyed and peat pore water chemistry was sampled at 3 plots located at 25%, 50% and 75% of the length of the long axis of the sampling area. Peat cores were taken at the same plots for a subset of 5 sites described below, and slope at the upland-peatland interface at the edge of the site was also measured for all 11 full basin sites and 4 of the 10 partial basin sites.The depth of organic sediment (primarily peat) was measured to depth of contact with mineral surface (typically sand) throughout the sampling area using a stainless steel peat depth probe (PDP) on a regular grid at intervals varying from 20-90 m. Two different versions of the PDP were used, and intercalibrated to ensure consistency. The first consisted of 60 (1.83 m) sections of 3 800 (0.95 cm) diameter threaded steel rod, connected with hex-shaped coupling nuts. The second was a custom-made version with the same general design including length and diameter of sections, but consisted of a smooth stainless steel surface and contained an inset male and female threading system to avoid the protruding coupling nuts. The PDP was used only to determine depth to refusal and was not equipped to collect samples; thus it could not differentiate between peat and soft organic lacustrine sediment. In nearly all cases, the person using the PDP could feel contact with sand (typical glacial sediments) at depth to refusal.Peat CoresPeat cores were taken at 13 different locations, including the central plot for site 4n and each of the 3 plots for sites 7b, 9b, 12f, and 21b. At each core location, samples were taken using a Russian-style corer (50 cm length, 5cm diameter) at depths of 0.5, 1.0, 2.0, 4.0 and 6.0 m, up to the maximum depth (peat-sand interface). We examined peat color and degree of decomposition using the von Post scale in the field. Particular attention was paid to the presence/absence of gyttja at the peat-sand interface. Gyttja is a dark olive-green algae-derived gelatinous lacustrine sediment, which indicates the former presence of a clear-water lake at a given site.For each core sample (N = 45), a central 10 cm section was preserved and used to measure moisture content, bulk density, and organic matter (OM) content in the laboratory. The 10 cm section was halved vertically, and one half (between 50 and 150 g wet weight) was used for measurement of wet bulk density (rw =mw per V; where mw = wet weight and V = volume measured by water displacement). The other half was used to measure mass loss by oven-drying at 55C until the mass was stable (typically 5-10 days, measurement precision = or - 0.1 g). Volumetric moisture content was calculated as (mw md) per V and bulk density (rb)asmdper V where md =dry weight, mw = wet weight, and V = volume calculated as mw per rw. From the dried sample, a 1-3 g homogenized subsample was ashed in a muffle furnace at 440C for 8 h to determine ash-free dry weight (maf = md mash; precision plus or minus 0.01 g), and OM content (OM%) was calculated as maf per md. Finally, OM density (rOM) was calculated as rb OM%.For each of the 13 core locations, we estimated the total mass of OM by summing the product of rOM and volume over all measurement intervals. To estimate a continuous vertical distribution, rOM was interpolated linearly by depth between measurement points. The 0.25-0.5 m interval was assigned the same rOM as the 0.5 m value, while the 0-0.25 m surface interval was assigned a rOM of half of that measured at 0.5 m, to account for the lower bulk density in living recently dead Sphagnum in the acrotelm. The deepest measured rOM value was extrapolated down to a depth of 0.25 m above the base of the core, and the basal 25 cm of the core was assigned a rOM of 46 kg m3. This is equivalent to the mean value measured for gyttja, to account for the fact that the peat is grading into lower-OM gyttja and or sand at the interface with glacial till. Vertically averaged mean rOM was calculated as the total mass of OM in the core divided by the total core volume.Edge Slope in the FieldBecause many peatlands in this region formed from in-filling of lakes, we hypothesized that local geomorphology, specifically slope at the peatland margin (peatland-upland interface), might be a good indicator of peatland depth. At a subset of 15 sites (including all 11 full basin sites) we measured slope at the peatland-upland interface (Edge Slope in the Field, ESF). At full basin sites, ESF readings were taken at 8 peatland-upland interface locations distributed evenly around the edge of the site. At partial basin sites, measurements were only taken at those site edges that were adjacent to upland, resulting in fewer than 8 locations at each site. At each location, a Suunto clinometer was used to measure slope (%) from the peatland-upland interface oriented up the steepest upland slope at a distance of 5 m, 10 m, 20 m and 30 m, and these four values were averaged to give a single slope value for each location. The precision of individual measurements was plus or minus 0.3% slope (mean SD of replicate measurements). The values used for statistical analysis were site mean (ESFmean) and maximum (ESFmax) of location slopes.VegetationWe hypothesized that the surface vegetation characteristics might be related to peat depth, either directly (due to differential contributions of plant species to decomposition rates and water-holding capacity), or indirectly by responding to local environmental characteristics (e.g., water table, groundwater flow) that also influence peat formation. Vegetation was surveyed following a modification of the U.S. Forest Service Forest Inventory and Analysis (FIA) protocol. At each plot, a circular sampling area with 7.3 m radius was laid out, with 3 linear transects extending from the center to the perimeter at 0, 120 and 240 degrees from compass north. Within the circular plot, all trees with diameter at breast height (DBH) of 2.5 to 4.9 cm were counted as saplings and their species recorded, while species and DBH were recorded for all trees with DBH at 5 cm. Basal area for each tree species in units of m2 ha-1 was calculated by summing the DBH of all individual trees and normalizing by plot area. The mean height and intersection length of each of 4 categories of shrub (alder, bog birch, ericaceous or tree seedling) were recorded for woody vegetation of height greater than 50 cm (but DBH less than 2.5 cm) that intersected any of the linear transects. Shrub percent cover was estimated for each category by dividing the intercepted length by the total transect length. Coarse woody debris (CWD) with length greater than 1 m that intersected any of the linear transects with diameter greater than 5 cm was tallied; small end diameter (down to 5 cm), large end diameter, and length of each piece of CWD was recorded and used to calculate volume as described by Waddell [2002]. Finally, three ground-layer quadrats (1 m2) were laid out, 1 each adjacent to the 3 linear transects spanning a distance of 4 to 5 m from the center point. Within each quadrat, percent cover was recorded for each of 8 commonly occurring ground cover types: bare ground, ericaceous shrubs, ferns, forbs, graminoids, Sphagnum mosses, other mosses, and other woody vegetation (tree seedlings). Values for the 3 quadrats were averaged to give a plot mean cover of each ground cover type.

Methods

1. Dataset: sr1 - Surface Water Resource Inventory (SWRI) Wisconsin. Temporal coverage: 1960-1980. Original description found in the preface of each Wisconsin Department of Natural Resources (WDNR) SWRI report, published by county.Data manipulation for incorporation into database: Original source of data is WDNR SWRI printed reports. An electronic version (MS Excel spreadsheet) of the data was available (the origin of this spreadsheet was unknown) and was used in preparation of this database. Some discrepancies observed between printed version and electronic version of the dataset: 1) in the printed reports, alkalinity is expressed either as methyl orange or methyl purple; varies from county to county. The electronic format does not contain any metadata or explanation regarding alkalinity. 2) in the printed reports, sometimes max depth provided, sometimes known depth, and sometimes Secchi depth- these values seem to have been transcribed as Secchi depth in the electronic dataset. 3) values of area, conductivity, alkalinity, and depth in electronic format have been rounded up from values in the books. 4) a field in the spreadsheet named "Cl" has no match in books and was not included in the final dataset. 5) color code was not defined in electronic format. It was deciphered and checked against a few lakes from different counties in the printed reports. Final color codes: 1 - Light brown. 2 - Medium brown. 3 - Dark brown. 4 - Clear. 5 - TurbidIssues specific to the electronic format: 13822 records originally. After eliminating all records without WBICs (Water Body Identification Code) or with duplicate WBICs, the dataset reduced to 12638 records with unique WBICs. Of these, 151 records (with area >10 acres) had no or zero data for some chemical parameters. Checked these records using WDNR SWRI reports. Eliminated any record that couldn't be resolved using the books and WDNR WBICs file.. Most records contain both alkalinity and conductivity data, although some do not contain both parameters. Final dataset sr1 has 12383 records2. Dataset: sr2 - Pieter Johnson. Temporal coverage: not specified. Original description: Combination of WDNR Register of Waterbodies (ROW) file, Wisconsin Lakes Book (wilk), and SWRI. Selected lakes with areas >= 10 acres, and lakes in at least 2 of the 3 datasets. Lakes with missing WBIC were not included. Lakes with missing surface area were not included.Data manipulation for incorporation into database: Received original dataset from Jake Vander Zanden (UW-Madison, Center Data manipulation for incorporation into database: Received original dataset from Jake Vander Zanden (UW-Madison, Center for Limnology). The dataset was used in the following publication: Johnson, P.T., J.D. Olden, M.J. Vander Zanden. 2008. Dam invaders: impoundments facilitate biological invasions in freshwaters. Frontiers in Ecology and the Environment 6:357-363. Original dataset contained 5213 records; . Eliminated 8 records without WBIC, legal (TRS) description, and no values for lake characteristics. Note: Many records are repeated from sr1 dataset. Final dataset sr2 has 5205 records.3. Dataset: sr3 - Biocomplexity Project. Temporal coverage: 2001-2004. Original description: Data Set Title: Biocomplexity; Coordinated Field Studies: Chemical Limnology. Investigators: Steve R. Carpenter, Jim Kitchell, Timothy K. Kratz, John J. Magnuson. Contact:NTL LTER Information Manager; Center for Limnology, 680 N Park St, Madison, WI, 53706-1492, USA;(phone) 608-262-2573;(fax) 608-265-2340;(email) infomgr@lter.limnology.wisc.edu; 62 Vilas County lakes were sampled from 2001-2004 (approximately 15 different lakes each year)Data manipulation for incorporation into database: Original dataset had 62 records. Replicate samples per lake averaged to single measurements. Two records represented a single lake (Little Rock, North and South basins); these were merged into one record. Final dataset sr3 has 61 records.4. Dataset: sr4 - Landscape Position Project. Temporal coverage: 1998. Original description: Data Set Title: Landscape Position Project: Chemical Limnology. Investigators: Ben Greenfield, Thomas Hrabik, Timothy K. Kratz, David Lewis, Amina Pollard, Karen Wilson. Contact: NTL LTER Information Manager; Center for Limnology, 680 N Park St, Madison, WI, 53706-1492, USA;(phone) 608-890-3446;(fax) 608-265-2340;(email) infomgr@lter.limnology.wisc.edu; Parameters characterizing the chemical limnology and spatial attributes of 51 lakes were surveyed as part of the Landscape Position Project.Data manipulation for incorporation into database: WBICs added. Ward Lake removed from data. Parameters values over multiple sampling events were averaged. Info regarding depth at which samples were taken was not retained. Final dataset sr4 has 50 records.5. Dataset: sr5 - Lillie and Mason. Temporal coverage: 1979. Original description: printed report WI DNR Technical Bulletin no.138. 1983. Limnological characteristics of Wisconsin LakesData manipulation for incorporation into database: Original file containing 667 records received from Paul Garrison (WDNR). 88 records lacked WBICs but 65 of these were assigned using WDNR lakes shapefile, matching names and areas of lakes. Final 23 records without WBICs were removed. Note: Since lake / impoundment classification doesn't seem to match Johnson's dataset (sr2), it was not included. Note from Richard Lathrop (WDNR): total P measurements are probably unreliable due to method used not being sensitive enough. Final dataset sr5 has 644 records.6.Dataset: sr6 - EPA- Eastern Lakes Survey (1984): Temporal coverage: 1984. Original description: Data Set Title: National Surface Water Survey: Eastern Lake Survey-Phase I. The Eastern Lake Survey-Phase I (ELS-I), conducted in the fall of 1984, was the first part of a long-term effort by the U.S. Environmental Protection Agency known as the National Surface Water Survey. It was designed to synoptically quantify the acid-base status of surface waters in the United States in areas expected to exhibit low buffering capacity. The effort was in support of the National Acid Precipitation Assessment Program (NAPAP). The survey involved a three-month field effort in which 1612 probability sample lakes and 186 special interest lakes in the northeast, southeast, and upper Midwest regions of the United States were sampled.Data manipulation for incorporation into database: Original dataset, downloaded from EPA website, has over 100 parameters. Only a small subset of interest was retained. Original documentation for full dataset available is available. Dataset includes 285 Wisconsin lakes. WBICs were assigned using geographic coordinates from dataset. WBIC for one lake could not be determined and was excluded.. Note regarding conductivity parameter: value represents calculated conductivity, as the sum of concentrations of each major cation and anion. It is not a parameter measured in the field or lab. Actual formula used to calculated conductivity was not discovered. Final dataset sr6 has 284 records7. Dataset: sr7 - Environmental Research Lab Duluth (ERLD). Temporal coverage: 1979-1982. Original description: ERLD Lake Survey. Contact(s): NTL LTER Information Manager; Center for Limnology, 680 N Park St, Madison, WI, 53706-1492, USA;(phone) 608-262-2573;(fax) 608-265-2340;(email) infomgr@lter.limnology.wisc.edu; Chemical survey of 832 lakes in Minnesota, Michigan, Wisconsin and Ontario conducted by ERL-Duluth and UMD between 1979 and 1982 for evaluation of trophic state and sensitivity to acid deposition Glass, G.E. and Sorenson, J.A. (1994) USEPA ERLD-UMD acid deposition gradient-susceptibility database. U.S. EPA Environmental Research Laboratory - Duluth and University of Minnesota at Duluth, MN.Data manipulation for incorporation into database: Dataset included 428 Wisconsin records for which WBICs were included. Note: Original dataset had several errors in WBIC assignment: 1179900 was assigned to three different water bodies; correct WBICs are: 1503000, 1502400, 1481100; also 1515800 changed to 1516000. Lake Clara had 5 different stations for most parameters sampled. First station that had values for all parameters was included in final dataset. Final dataset sr7 has 428 records.8. Dataset: sr8 - Birge-Juday Historical Dataset. Temporal coverage: 1925-1941. Original description: Birge-Juday Historical Lake Data. Investigator(s): Edward A. Birge, Chauncy Juday. Contact: NTL LTER Information Manager; Center for Limnology, 680 N Park St, Madison, WI, 53706-1492, USA;(phone) 608-262-2573;(fax) 608-265-2340;(email) infomgr@lter.limnology.wisc.edu; Data collected by Birge, Juday, and collaborators, mostly in north-central Wisconsin, from 1925 through 1941; generally one sample per lake during the summer, but on some lakes, especially around Trout Lake Station, samples were taken on several successive years. Note that not all variables were measured on all lakes (scarce data for nutrients and ions). Documentation: Johnson, M.D. (1984) Documentation and quality assurance of the computer files of historical water chemistry data from the Wisconsin Northern Highland Lake District (the Birge and Juday data).WDNR Technical Report. Number of sites: 608 (generally one sampling point per lake; occasionally, several sampling points per lake on multibasin, large lakes).Data manipulation for incorporation into database: Original dataset downloaded from UW-Madison, Center for Limnology LTER website. Values averaged for lakes with multiple samples. WBICs assigned to 577 lakes via GIS spatial join using site coordinates and WDNR lake shapefile. Note from Johnson, M.D. (1984): the units for alkalinity (fixed CO2) changed from cc/L to mg/L sometime between Aug 1926 and May 1927. 17 entries were originally cc/l. Thus there might be inconsistencies in the alkalinity data. Final dataset sr8 has 577 records.9. Dataset: sr9 - USGS National Water Inventory System (NWIS). Temporal coverage: 1969-2009. Original description: U.S. Geological Survey. This file contains selected water-quality data for stations in the National Water Information System water-quality database (http://nwis.waterdata.usgs.gov/nwis/). Explanation of codes found in this file are followed by the retrieved data. The data you have secured from the USGS NWIS Web database may include data that have not received Director's approval and as such are provisional and subject to revision. The data are released on the condition that neither the USGS nor the United States Government may be held liable for any damages resulting from its authorized or unauthorized use.Data manipulation for incorporation into database: Data downloaded for 240 lakes for the following parameters: calcium, conductivity, alkalinity, pH. Original parameter codes (USGS NWIS schema): p00915 p00095 p00400 p00916 p29801 p39086 p90095. Data are averaged for multiple measurements. WBICs assigned via GIS spatial join using site coordinates and WDNR lake shapefile. Final dataset sr9 has 240 records.10. Dataset: sr10 - WI Department of Natural Resources (WDNR) Temporal coverage: 1969-2009 Original description: available at http://dnr.wi.gov/org/water/swims/Data manipulation for incorporation into database: Original data received from Jennifer Filbert (WDNR). Data were extracted from WDNR Surface Water Integrated Monitoring System (SWIMS) database (http://dnr.wi.gov/org/water/swims/). Lakes represented had one or more of the following parameters: Secchi depth, calcium, conductivity, alkalinity, pH, total P, turbidity,, chlorophyll a. Data were averaged where multiple measurements were available. Final dataset sr10 has 53 records.The Data Source data table contains a summary of the 10 data sources with information on temporal coverage and record counts. It also includes information on the availability of calcium and conductivity data from the data sources.

Methods

Our first goal was to understand the relationship between primary productivity and species richness for several groups of freshwater organisms. By species richness, we mean the number of species observed in a lake over a number of years. It is useful to have several years of observations because the number of species observed varies from year to year. We chose the total list of species (the asymptote of the collectors curve) as our index of species richness. The lakes studied as part of the U.S. Long-Term Ecological Research (LTER) Program are particularly valuable because they have been studied for two decades, and complete species lists exist for many kinds of organisms in these systems. LTER lake sites occur in northern and southern Wisconsin and northern Alaska (Toolik Lake). However, because there are fewer than 15 LTER lakes (and only seven with measured rates of primary productivity), we increased sample size by including data from additional well-studied lakes of similar size, but which span a greater range of primary productivity (see Table 1). These lakes have been studied for several years, and estimates of annual primary productivity exist for each lake. Some well-studied lakes were not included, such as those which lacked much of the crucial data, or lakes that were unusually turbid or saline. For example, Lake Okeechobee (Florida, USA) is turbid and exhibits a wide range of productivity levels, depending on the part of the lake sampled, while Marion Lake (British Columbia, Canada) has a flushing rate of only a few days (W. E. Neill, personal communication). Sampling design and protocol are not standardized among studies of lakes. For example, species identifications were done by different people, sampling period was quite variable, and the number of samples per lake was variable. Such heterogeneity reduces the accuracy and precision of relationships between productivity and species richness.Primary productivity.—Pelagic primary productivity (PPR) can be measured by the 14C method (Vollenweider 1974). This method gives a close approximation to gross primary productivity (GPP), but because some of the fixed carbon is respired quickly, the value obtained is less than GPP (Fee et al.1982). Point values of PPR are then integrated by depth and area to produce estimates of whole-lake annual primary productivity per cubic meter or square meter.Lake primary productivity is fundamentally different than productivity measured in other biomes (e.g., grasslands, forests). The 14C method measures available (gross) primary productivity more than utilized (net) production, which is what is normally measured in terrestrial systems. The 14C method is also a fairly direct measure of productivity, compared to the proxy methods (e.g., nutrient loading, biomass, climate, soil fertility) used in many studies.Sampling protocols for aquatic organisms.—Sampling protocols differed among taxonomic groups and lakes (e.g., Downing and Rigler 1984). For example, phytoplankton samples are taken by capturing (at most) a few liters of lake water, either from a specific depth or with a sampler that integrates water across a range of depths. Zooplankton are usually sampled by vertical tows (i.e., raising a net through the water column). Both zooplankton and phytoplankton samples are typically taken from the center of the lake, although replicate samples at different locations may be taken from larger lakes. Planktonic organisms are much smaller than the sampling device, and hundreds to hundreds of thousands of organisms are typically captured in a single sample. In contrast, aquatic macrophytes are sampled using quadrats and rake samples, or simply based on a walk around the lake, while fish are sampled using a variety of nets andoror electroshocking equipment. Criteria for species lists.—Species lists for fish, macrophytes, and pelagic phytoplankton, rotifers, cladocerans, and copepods were obtained from the literature and from unpublished data. We avoided lists restricted to only dominant or common species, and thus included only lists that were exhaustive. Few lakes had species lists for all six groups of organisms. However, we included any lake that had an estimate of the average annual primary productivity and had lists for at least three taxa.We standardized this database by developing criteria for inclusion of species in analyses. Phytoplankton lists included all prokaryotic and eukaryotic photosynthetic phytoplankton for which there were abundances of more than one organism per milliliter (a criterion also used by Lewis 1979). We included all nonsessile species caught in open water as pelagic rotifers. For the crustacean zooplankton (cladocerans and copepods), we followed the criteria of Dodson (1992). Species lists of macrophytes included all submerged, floating, or emergent species of flowering plants, including Typha, sedges, grasses, and duck weed. We did not include Isoetes or macroalgae such as Chara and Nitella as macrophytes. The fish list included all species reported from the lake, including introduced taxa. Fish species reported to occur in the watershed, but not in the lake (as in Pearse1920) were not considered part of the lakes biota.

Methods

Dodson, S. I. and R. A. Lillie. 2001. Zooplankton communities of restored depressional wetlands in Wisconsin, USA. Wetlands 21:292-300.

Methods

Schell, Jeffery M., Carlos J. Santos-Flores, Paula E. Allen, Brian M. Hunker, Scott Kloehn, Aaron Michelson, Richard A. Lillie, and Stanley I. Dodson. 2001. Physical-chemical influences on vernal zooplankton community structure in small lakes and wetlands of Wisconsin, U.S.A. Hydrobiologia 445:37-50