Stream Water Quality to Support HUC 12 Prioritization in the Lake Wister Watershed, Oklahoma: August 2017 Through May 2019

University of Arkansas, Fayetteville ScholarWorks@UARK

Technical Reports Arkansas Water Resources Center

11-2019

Stream Water Quality to Support HUC 12 Prioritization in the Lake Wister Watershed, Oklahoma: August 2017 through May 2019

Bradley J. Austin University of Arkansas, Fayetteville

Brina A. Smith University of Arkansas, Fayetteville

Brian E. Haggard University of Arkansas, [email protected]

Follow this and additional works at: https://scholarworks.uark.edu/awrctr

Part of the Fresh Water Studies Commons, and the Water Resource Management Commons

Citation Austin, Bradley J.; Smith, Brina A.; and Haggard, Brian E.. 2019. Stream Water Quality to Support HUC 12 Prioritization in the Lake Wister Watershed, Oklahoma: August 2017 through May 2019. Arkansas Water Resource Center, Fayetteville, AR. MSC389. 29 https://scholarworks.uark.edu/awrctr/359

This Technical Report is brought to you for free and open access by the Arkansas Water Resources Center at ScholarWorks@UARK. It has been accepted for inclusion in Technical Reports by an authorized administrator of ScholarWorks@UARK. For more information, please contact [email protected]. Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

WATERSHED INVESTIGATIVE SUPPORT TO THE POTEAU VALLEY IMPROVEMENT AUTHORITY

Stream Water Quality to Support HUC 12 Prioritization in the Lake Wister Watershed, Oklahoma: August 2017 through May 2019

2019 November Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Stream Water Quality to Support HUC 12 Prioritization in the Lake Wister Watershed, Oklahoma: August 2017 through May 2019

Bradley J. Austin¹, Brina A. Smith², and Brian E. Haggard3

¹Postdoctoral Research Associate, Arkansas Water Resources Center, University of Arkansas 2Water Quality Laboratory Technician, Arkansas Water Resources Center, University of Arkansas 3Director of the Arkansas Water Resources Center; Professor, Department of Biological Engineering, University of Arkansas

INTRODUCTION Lake Wister is on Oklahoma’s 303(d) list for im- Nonpoint source pollution associated with human paired water quality, including excessive algal biomass, land use (agriculture and urbanization) is one of the pH, total phosphorus (TP), and turbidity (ODEQ, 2014). leading causes of impairment to waterways in the Unit- To address these water quality issues, the Poteau Valley ed States (EPA 2000). The primary pollutants associat- Improvement Authority (PVIA) released its “Strategic ed with agricultural and urban land use are sediment Plan to Improve Water Quality and Enhance the Lake and nutrients which enter nearby streams during rain Ecosystem” in 2009. The strategic plan breaks down the events and are then carried downstream. These sedi- restoration efforts into three zones of action to focus ments and nutrients may result in water quality issues on the watershed, the full lake, and Quarry Island Cove. in the downstream water bodies like increased algal This study is a continuation of the study completed in growth or decreased water clarity (e.g. Smith et al., June of 2017 (Austin, Smith, et al. 2018). The purpose 1999). of this project was to monitor stream water quality Best management practices (BMPs) are often used during base flow conditions at or near the outlets of to mitigate the effects of nonpoint source pollution in the subwatersheds, in the Oklahoma portion of the the watershed. Practices such as riparian buffers in- Lake Wister Watershed (LWW). The Oklahoma Non- stalled along the edge of field and conservation tillage point Source Management Program Plan suggests that (e.g., no-till, spring-till, and cover crops) slow overland monitoring and assessment at the HUC 12 subwater- flow, reducing erosion and nutrient loss from the land- shed scale is the most effective means to identify- wa scape (Schoumans et al. 2014). Installing BMPs through- ter quality problems associated with nonpoint source out the entire watershed would have the greatest effect pollution (NPS Management Program Plan, 2014). The at reducing nonpoint source pollution; however, this is primary goal of this monitoring was to assist PVIA and not socially or economically feasible. Targeting critical other stakeholders in identifying the HUC 12 subwater- source areas or priority watersheds for BMPs installa- sheds where implementation of BMPs could be priori- tion, optimizes the benefits while reducing the overall tized to address sediment and nutrient transport from (Sharpley et al. 2000). the landscape. One way of targeting priority watersheds for im- plementing BMPs is through water quality monitoring STUDY SITE DESCRIPTION during base flow (McCarty and Haggard 2016, Austin, The LWW covers an area of 2,580 km2 (~640,000 Patterson, et al. 2018, McCarty et al. 2018). The prem- acres) and makes up the southern half (52%) of the en- ise is that stream water quality during base flow condi- tire Poteau River sub-basin (HUC 11110105; Figure 1). tions reflects the influence of nonpoint source pollution The LWW is divided into 10-digit hydrologic unit code across the watershed. Stream nutrient concentrations or HUC 10 watersheds. The headwaters of the Poteau generally increase with the proportion of agricultural River watershed is entirely within Arkansas and was and urban land use in the watershed (Haggard et al. not part of this study. The Black Fork Poteau River and 2003, Giovannetti et al. 2013, Cox et al. 2013, Austin, the Poteau River watersheds traverse the state line Patterson, et al. 2018). Thus, stream water quality can between Oklahoma and Arkansas, and the Middle Po- be related to human development (i.e., percent urban teau River and Fourche Maline watersheds are entirely and agriculture land cover) across a target watershed within Oklahoma. The HUC 10 watersheds that make and this relationship can be used to highlight priority up the LWW range in size from 377 to 675 km2 (93,300 subwatersheds. to 166,800 acres). The primary land use and land cover

2 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 1: Sampling sites for the routine monitoring (red symbols) and special studies (gray symbols) across the Oklahoma portion of the Lake Wister Watershed.

(LULC) across the Oklahoma portion of the LWW is 72% METHODS forest, 19% agriculture, and 4% urban; the LULC for the 845 km2 (~209,000 acres) portion of the LWW in Arkan- Sample Collection and Analysis sas is similar with 71% forest, 20% agriculture, and 5% Water samples were collected at the 26 sites at ap- urban. proximately monthly intervals during base flow condi- Within the Oklahoma portion of the LWW there are tions from August 2017 through May 2019 (following 26 HUC 12 subwatersheds that range in size from 42 to the approved quality assurance project plan). Water 125 km2 (10,300 to 30,800 acres; Table 1). Forest is the samples were not collected in January 2018 because dominant LULC across the HUC 12s, ranging from 45 to several stream reaches were dry. The samples were 95% of the watershed. The proportion of human devel- collected from the vertical centroid of flow where the opment (i.e., agriculture plus urban) was less than half water is actively moving either by hand or with an Al- of the LULC across the HUC 12s (4−48%; Table 1). Addi- pha style horizontal sampler lowered from the bridge. tionally, across the LWW there are 7 EPA national pol- Water samples were split, filtered, and acidified in the lutant discharge elimination system (NPDES) permitted field based on the specific storage needs for each an- point sources, including waste water treatment plants alyte. Field duplicate water samples were collected at (WWTPs), sewage systems, and a poultry processing 10% of the sites within each monthly sampling event; plant (Table 2). these field duplicates were collected in the same fash- For this study we selected 26 sampling sites near ion as the original water sample. Additionally, a field the outflow of 23 of the HUC 12’s in the Oklahoma por- blank was collected during each sampling event. A sum- tion of the LWW (Figure 1; Table 3). The LULC for the mary of field quality assurance and quality control (QA/ catchments upstream of the 26 sample sites ranged QC) data can be found in Appendix 1. All samples were from 49−95% forest, <1−37% agriculture, and <1−10% stored on ice until delivered to the Arkansas Water urban. While these sampling sites are located near the Resources Center certified Water Quality Labs (AWRC outflow of many of the HUC 12s within the Oklahoma WQL). portion of the LWW, they represent the catchment area In addition to the routine monthly sampling, wa- upstream of them and not specifically the HUC 12s. ter samples were collected within select HUC 12 sub-

3 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Table 1: Hydrologic unit code (HUC) 12 subwatersheds in the Oklahoma portion of the and processed in the same manner Lake Wister Watershed and corresponding LULC data, organized at the HUC 10 watershed as routine monthly samples. scale. All water samples, field dupli- HUC 12 Huc 12 Name Area (km2) % F1 % AG2 % U3 %HDI4 cates, and field blanks were ana- HUC 10-1111010502: Black Fork Poteau River lyzed for anions (Cl and SO4), am- 111101050201Big Creek 111.7 90 4 5 9 monia-nitrogen (NH4-N), nitrate-N plus nitrite-N (hereinafter, NO3-N), 111101050202Upper Black Fork 124.5 87 7 2 9 total N (TN), soluble reactive phos- 111101050203Haws Creek 73.5 91 3 2 5 phorus (SRP), total P (TP), turbidity, 111101050204Shawnee Creek 50.2 87 2 6 9 total suspended solids (TSS) and 111101050205Cedar Creek 49.8 95 1 3 4 sestonic chlorophyll-a (chl-a) fol- 111101050206Lower Black Fork 100.4 81 14 3 17 lowing standard methods (Table HUC 10-1111010503: Poteau River 4). The analytical techniques, re- 111101050303Cane Creek 70.4 68 20 4 24 porting limits and method -detec tion limits are provided (Table 4), 111101050304Sugar Creek 71.3 68 27 3 30 and additional information about 111101050305Hontubby Creek 55.5 63 25 9 34 the certified labs are available at: HUC 10-1111010504: Fourche Maline https://arkansas-water-center. 111101050401Cunneo Creek-Fourche Maline 55.5 83 13 1 14 uark.edu/water-quality-lab.php 111101050402Coon Creek-Fourche Maline 74.4 80 13 4 17 (date acquired 9/22/2019). 111101050403Bandy Creek 61.6 48 38 10 48 111101050404Little Fourche Maline 61.8 68 25 3 28 Data Analysis 111101050405Clear Creek-Fourche Maline 56.1 53 40 4 44 All LULC data for the LWW, HUC 12s within the LWW, and catch- 111101050406Red Oak Creek 73.2 54 37 6 42 ments upstream of each sampling 111101050407Upper Long Creek 103.8 83 12 2 13 location were compiled using - Ge 111101050408Lower Long Creek 78.2 77 16 1 17 odataCrawler http://www.geo- 111101050409Pigeon Creek-Fourche Maline 110.5 53 35 2 37 datacrawler.com/ (Leasure 2013) HUC 10-1111010505: Middle Poteau River and Model My Watershed https:// 111101050501Coal Creek- Poteau River 83 74 19 4 23 app.wikiwatershed.org/ (date ac- 111101050502Upper Holson Creek 77.7 60 25 3 28 quired 1/31/2018). Within this LULC data, forest is defined as the 111101050503Coal Creek- Fourche Maline 41.6 68 19 4 22 sum of deciduous, evergreen, and 111101050504Middle Holson Creek 73 94 3 2 5 mixed forest, agriculture is the 111101050505Lower Holson Creek 59.1 89 5 4 9 sum of pasture/hay, row crop, and 111101050506Cedar Creek-Fourche Maline 59.5 82 12 3 14 grassland, and urban is the sum of 111101050507Baker Branch-Fourche Maline 97.6 65 23 5 28 barren, developed open, and low, 111101050508Wister Lake Dam 75.5 45 16 3 19 medium, and high intensity development. Previous, studies from 1 % Forest, includes deciduous, evergreen and mixed forest; 2 % Agriculture, includes crops, grassland, and pasture/hay; 3 % Urban, includes barren, developed-open space, low, me- northwest Arkansas have found dium, and high intensity development; 4 % Human Development Index is the sum of % stream nutrient concentrations to Agriculture and % Urban. increase with increasing percent agriculture and urban area up- watersheds to further understand the spatial variabil- stream (Haggard et al. 2003, Giovannetti et al. 2013). ity in water quality and potential sources of nutrients Because of this, a simple human development index within them. Additional sites were sampled within the (HDI) was calculated as the total percent agriculture Little Fourche Maline HUC 12 subwatershed and the and urban land use for the catchment upstream of each Oil Branch subwatershed, tributary of the Poteau River sample site and for each subwatershed (Tables 1 & 3). (Figure 1). The sites were sampled a total of three times All water quality data collected from August 2017 during the monthly routine samplings in January, Feb- through May 2019 can be found in the data report ruary, and March of 2019. All samples were collected “DR-WQ-MSC389” (last accessed). For the purpose of

4 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Table 2: National Pollution Discharge Elimination System (NPDES) permitted sites within the concentrations likely increase. For Lake Wister Watershed in Arkansas and Oklahoma. constituent changepoint analyses NPDES Code Location Source the overall geomeans from each site were plotted against the HDI OK0038407 Heavener, OK WWTP1 for each site. OK0031828 U.S. Forest Service - Cedar Lake, near Hodgen, OK Sewage systems OK0022951 Jim E. Hamilton Correctional Center, near Hodgen, OK WWTP RESULTS AND DISCUSSION OK0021881 Wilburton, OK WWTP OK0031631 Red Oak Public Works Authority, Red Oak, OK Sewage systems Nitrogen AR0038482 Tyson Poultry, Waldron, AR Poultry Processing Annual geomean concentra- AR0035769 Waldron, AR WWTP tions of NH4-N across the streams -1 1 WWTP = Waste water treatment plant ranged from 0.01 to 0.17 mg L over the course of the study. NH4-N did not vary annually, and analysis in this report, each parameter’s MDL was sub- annual geomean concentrations were not different stituted for measured values below the MDL and each from the project geomean (F3,100=0.767; P=0.575). parameters RL was substituted for all measured values While NH4-N was not variable between years, it did vary greater than the MDL but less than the RL, with the seasonally (F3,300=4.30; P=0.005), with concentrations exception of turbidity, TSS, CHL a, where all measured greatest in the summer and least in the winter (Figure values were left uncensored 3E). This difference in NH4-N concentrations between The geometric mean (hereinafter geomean) of con- seasons was driven by Site 23, which consistently ex- stituent concentrations at each site was used inthe ceeded 0.2 mg L-1 during the summer months. Overall, data analysis, because it is less sensitive to extreme low we would not expect to see relatively high NH4-N con- and high values than arithmetic means. The geomean centrations, except maybe downstream from effluent is typically a good estimate of the central tendency or discharges, as is the case with Bandy Creek (Merbt et al. middle of the data. Seasonal, annual, and overall proj- 2011) because it is quickly nitrified in streams (Haggard ect (all three years) geomeans were calculated for the et al. 2005). water quality parameters at each site. Parameter geo- Nitrate concentrations were relatively low across means and ranges at each site and for each project year the streams sampled, where annual geomean concen- can be found in Appendix 2. trations of NO3-N varied from 0.01 to 0.55 mg L-1. There The overall project geomeans from each site were were no clear annual or seasonal patterns in NO3-N related to HDI using simple linear regressions, generat- across the streams sampled (F3,100=0.831; P=0.480, ing a linear model for each parameter. This statistical F3,300=0.420; P=0.739; respectively) possibly because analysis shows how geomean concentration increases NO3-N was less than 0.1 mg L-1 at most sites through- across a gradient of human development within the out each year. While there were no seasonal trends in watershed. The predictive equation, associated with NO3-N, sites with elevated NH4-N in summer tended to the linear regression, may have some merit in set- have elevated NO3-N in the summer as well, likely due ting achievable water quality targets across the LWW. to increased rates of nitrification from increased4 NH We cannot expect a stream with relatively high HDI availability ((Kemp and Dodds 2002)). to have constituent concentrations reflective of near The majority of TN in the flowing waters was in the background conditions. However, it may be feasible to particulate form, where dissolved inorganic N (DIN: expect streams with constituent concentrations well NH4-N plus NO3-N) was on average less than 30% of the above the regression line to be reduced to near or be- total. Annual geomean concentrations for TN ranged low the line. from 0.07 to 1.59 mg L-1. This range in TN is fairly con- Changepoint analysis is another way to examine sistent across all three years (F3,100=0.506; P=0.679; how HDI might influence constituent concentrations Figure 2A), and there was no real seasonal pattern in streams. Changepoint analysis looks for a threshold (F3,300=1.341; P=0.270; Figure 3A). Overall, nitrogen in the geomean concentration and HDI relation, where concentrations tended to be within the range nutrient the mean and variability in the data changes. This sta- supply threshold concentrations needed to promote tistical analysis is not dependent on data distributions, algal growth and cause shifts in algal community com- and it gives a threshold in HDI where the geomean position (0.27-1.50 mg L-1; (Evans-White et al. 2013),

5 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Table 3: Sample sites and land cover within the Lake Wister Watershed organized by HUC 10s. The number in the HUC 12 column is the final two digits associated with the HUC10 number listed at the top of each group of sites. Site # HUC 12 Stream Name Area (km2) %F1 %AG2 %U3 % HDI4 Latitude Longitude HUC10-1111010503: Upper Poteau River *1 3 Poteau River 694 66 25 5 30 34.880 -94.483 *2 4 Poteau River 768 66 25 5 30 34.859 -94.566 *3 5 Poteau River 1335 74 18 5 22 34.858 -94.629 HUC10-1111010505: Middle Poteau River 4 2 Conser Creek 34 95 3 2 5 34.867 -94.704 5 4 Holson Creek 73 94 3 2 5 34.807 -94.838 6 5 Holson Creek 132 92 4 3 7 34.823 -94.876 7 6 Holson Creek 182 91 5 3 7 34.879 -94.853 8 2 Rock Creek 11 67 30 2 32 34.843 -94.636 9 3 Coal Creek 27 72 19 2 21 34.951 -94.890 HUC10-1111010502: Black Fork Poteau River 10 2 Black Fork 122 88 6 2 9 34.760 -94.490 11 1 Big Creek 112 90 3 5 8 34.769 -94.499 12 3 Black Fork 323 89 5 3 8 34.793 -94.526 *13 4 Shawnee Creek 48 88 1 6 8 34.768 -94.628 14 5 Cedar Creek 48 95 1 4 4 34.779 -94.640 *15 6 Black Fork 509 88 6 4 9 34.843 -94.625 *26 4 Shawnee Creek 23 93 1 5 6 34.789 -94.628 HUC10-1111010504: Fourche Maline 16 8 Long Creek 180 80 13 1 15 34.908 -94.980 17 7 Long Creek 77 83 12 1 13 34.851 -95.066 18 7 Long Creek tributary 20 87 8 3 12 34.840 -95.054 *19 9 Fourche Maline 417 63 28 4 32 34.929 -94.981 *20 6 Red Oak Creek 71 54 37 6 43 34.936 -94.981 21 4 Little Fourche Maline 55 70 23 3 26 34.927 -95.163 *22 5 Fourche Maline 313 67 26 4 30 34.912 -95.156 *23 3 Bandy Creek 59 49 37 10 47 34.902 -95.261 24 2 Fourche Maline 72 81 12 4 16 34.933 -95.319 25 1 Cunneo Creek 45 90 7 >1 7 34.942 -95.298 1 %Forest, includes deciduous, evergreen and mixed forest; 2 %Agriculture, includes crops, grassland, and pasture/hay; 3 % Urban, includes barren, developed-open space, low, medium, and high intensity development; 4 %Human Development Index is the sum of %agriculture and %urban; and * indicates sites downstream of EPA NDPES permitted point sources. potentially creating nuisance algal conditions. The relationships with stream N concentrations The geomean concentrations of the N species var- and HDI have been observed across the region (e.g. ied across the LWW, reflecting changes in nutrients see Haggard et al. 2003; Migliaccio & Srivastava 2007; sources and land uses within the drainage areas. The Giovannetti et al. 2013). The regression lines provide geomean N concentrations increase with the propor- a possible water-quality target to where N concentration of agriculture and urban development (Figures 4A, tions might be reduced at a given HDI. The sites, or C, &E), i.e., HDI values, in the watershed, explaining: streams, with concentrations well above this line might • 46% of the variability in NH4-N, be of specific interest for management, e.g. Site 23. • 35% of the variability in NO3-N, and The geomean concentrations of the N species also • 70% of the variability in TN. showed changepoints or threshold responses to in-

6 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Table 4: Laboratory parameters with specific EPA approved analyt- increase algal growth and drive shifts is algal commu- -1 ical procedures nity composition in streams (0.007 – 0.100 mg L ; (Ev- Parameter Method Units RL MDL ans-White et al. 2013) and potentially cause nuisance algal conditions; although, two sites with values much NO -N EPA 353.2 mg L-1 0.01 0.005 3 higher than this range were directly downstream of ef- NH -N EPA 351.2 mg L-1 0.01 0.01 3 fluent discharges (Bandy Creek and Shawnee Creek at Cl EPA 300.0 mg L-1 0.5 0.06 Hwy 59). -1 SO4 EPA 300.0 mg L 0.5 0.11 Geomean P concentrations varied across the SRP EPA 365.1 mg L-1 0.005 0.004 streams draining the LWW, showing that over 70% of TP APHA 4500PJ mg L-1 0.02 0.003 the variability in SRP and TP concentrations was ex- TN APHA 4500PJ mg L-1 0.05 0.01 plained by HDI (Figures 4B & D). These relationships between stream P concentrations and HDI, like N species, Chl a APHA 10200 H1&2C µg L-1 -- --

-1 have been observed across the region (e.g. see (Hag- TSS EPA 160.2 mg L 4 -- gard et al. 2003, Cox et al. 2013), reflecting the poten- Turbidity EPA 180.1 NTU -- -- tial P sources such as poultry litter applied to pastures (DeLaune et al. 2004, Cox et al. 2013). The regression lines provide a realistic water quality target to where P creasing HDI; that is, the average and deviation of the concentrations might be reduced and show sites that geomeans increased above an HDI value. The change- deviate greatly from concentrations at a given HDI. points were relatively similar across the N species at The geomean concentrations of the P species also 28% HDI (Figures 5A, C, &E). The average of the data showed changepoint responses to increasing HDI. The above the changepoint was generally 2 to 3 times changepoints for P species were slightly lower than for greater than the data below that HDI value. Subwater- N species at 24% HDI. In both cases mean values to the sheds with measured values greater than the site av- right (above) of the threshold were 3 times greater than erage above the changepoint should be considered as the mean values to the left (below) of the threshold. higher priority than sites below the site average of sites Site 23 consistently shows elevated P and N concentra- beneath the changepoint. tions relative to other sites across the LWW, suggesting nutrient sources upstream might need to be investigat- Phosphorus ed (Figure 5B & D). Geomean concentrations of SRP across the streams -1 ranged from less than 0.005 to 0.189 mg L , with 45% Chlorophyll a of the values measured less than the lab’s reporting -1 Annual geomean concentrations of sestonic Chl-a limit (0.005 mg L ). Geomean concentrations of SRP did (algal biomass in the water column) ranged from 0.2 not vary between project years (F3,100=0.465; P=0.707 to 18.5 µg L-1 across the streams in the LWW. Over- Figure 2D) or between seasons (F3,300=2.059; P=0.106 all, CHL-a was less than 5 µg L-1 in nearly 80% of the Figure 3D). Overall, SRP concentrations across the samples collected from July 2016 through May 2019. streams of the LWW were low with nearly 75% of sites -1 Geomean Chl-a concentrations were consistent across having geomean concentrations less than 0.015 mg L . project years (F3,100=2.318; P=0.080; Figure 2F), but Geomean concentrations for TP ranged from -1 showed some variability between seasons with concen- 0.013 to 0.265 mg L ; much of which was in the par- trations greatest in the summer and least in the spring ticulate form, where the dissolved form (SRP) typically (F3,300=2.761; P=0.042; Figure 3F). made up less than 33% of the measured TP. TP con- The geomean concentrations of Chl-a increased centrations were not different between project years with the proportion of human development in the wa- (F3,100=0.350; P=0.789; Figure 2B), whereas TP con- tershed (i.e., HDI values), where HDI explained 54% of centrations in the summer were greater than concen- the variability in sestonic Chl-a (P<0.001; Figure 4F). trations in the fall and winter (F3,300=3.400; P=0.018; This strong relationship was surprising, because many Figure 3D). The increase in TP across the streams during physical, chemical, and biological factors influence algal the summer corresponded with slight increases in TSS growth in streams (Evans-White et al. 2013). However, and Chl-a in the water column (discussed later). Like in steams hydrology (e.g. discharge; Honti et al. 2010) TN, TP concentrations tended to be within the range is one of the most important factors since most algal or nutrient supply threshold concentrations needed to growth would be on substrates not generally in the

7 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 2: Box and whisker plots of constituents showing medians (horizontal line within each box), range (error bars show the 5th and 95th percentiles), and outliers (points above and below error bars) for each of the constituents analyzed at the Oklahoma sites in the Lake Wister Watershed. The full project (FP) median and range is shown to the right of the individual project years.

8 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 3: Box and whisker plots of seasonal variability in constituents showing medians (horizontal line within each box), range (error bars show the 5th and 95th percentiles), and outliers (points above and below error bars) for each of the constituents analyzed at the Oklahoma sites in the Lake Wister Watershed.

9 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 4: Simple linear regression of geomean constituent concentrations verse human development index (HDI) values for the Oklahoma portion of the Lake Wister Watershed. The site number in red is Shawnee Creek at highway 59 downstream of effluent discharge, thus it was not used in the statistical analysis.

10 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority water column (i.e., sestonic). It is likely that this cor- relation is driven by the increased nutrient concentra- Anions tions found at sites with higher HDI values. Geomean Annual geomean concentrations of Cl ranged from Chl-a concentrations across these sites were strongly 2 to 17 mg L-1, site Cl geomeans for individual project (positively) related to both TP (r2=0.47; P<0.001) and years were not significantly different from the overall TN (r2=0.79; P<0.001). The sites with elevated Chl-a project geomean (Figure 2I). However, Cl geomeans had increased total nutrient concentrations and sup- varied significantly between seasons, with Cl greatest ply available, as seen in other systems (Chambers et in the winter (Figure 3I), this was likely due to great- al. 2012, Haggard et al. 2013). Not suprisngly, based on er groundwater inputs during the winter. Also, great- its relationship with TN and TP, sestonic Chl-a showed er Cl concentrations during the winter may be due to a threshold at an HDI value (28%) similar to that ob- the use of road deicers during icy road condition as has served with nutrient concentrations (Figure 5F). been found elsewhere (Sun et al. 2014). Despite having greater concentrations in the winter, Cl was consis- Suspended Sediments and Turbidity tently below EPA secondary drinking water standards of Annual geomeans for turbidity and TSS were from 250 mg L-1 across all sites sampled. Relatively few stud- 4 to 57 NTU and from 1 to 31 mg L-1, respectively. These ies have focused on toxicity of Cl on freshwater fish. two constituents were strongly correlated (r=0.90; However, the reported values in this study for Cl were P<0.001) and show similar seasonal patterns, with 2 to 3 orders of magnitude less than those reported to greater values in the spring and lesser values in the fall have chronic toxicity effects on fat head minnows and (Figures 3G & H). Low values in the fall, for both constit- rainbow trout [704 mg L-1 and 1174 mg L-1, respectively uents, may be explained by the drier conditions during (Elphick, Bergh, et al. 2011)]. the fall. The less frequent rainfall events producing run- Annual geomean concentrations of SO4 ranged from off, reduces erosion from the landscape and within the 2 to 40 mg L-1. Like Cl, annual geomeans for SO4 were fluvial channel, and the lower flows throughout this not different from the overall project SO4 geomean (Fig- season have less power to erode the channel and keep ure 2J), but showed seasonal variability, with increased particulates in the water column (Morisawa 1968). The concentrations during the winter (Figures 3J). This was more frequent storms and elevated base flow during likely due to a combination of increased groundwater spring and early summer likely keep TSS and turbidity inputs and the use of road deicers (Sun et al. 2014). Sul- elevated in streams (relative to fall) across the LWW. fate concentrations were consistently below EPA sec- Many factors influence turbidity and the amount ondary drinking water standards of 250 mg L-1 across of particulates in the water column of streams, includ- all sites sampled. Chronic toxicity of SO4 on aquatic or- ing rainfall-runoff, discharge, channel erodibility, and ganisms varies in relation to the water hardness, with even algal growth to some degree. The myriad of fac- greater SO4 toxicity under soft water conditions (hard- tors that influence turbidity (and particulates) in water ness<80 mg L-1 measured as CaCO3) which is common in are also influenced by human activities, which is likely sandstone dominated systems such as the LWW. Sulfate why HDI explained more than 50% of the variability in values measured were lower than suggested standards geomeans of turbidity and TSS across the streams of for protecting aquatic life in soft water systems [129 mg the LWW (Figure 4F & H). These relations are not well L-1 SO4 (Elphick, Davies, et al. 2011)]. defined regionally but where data is available similar The geomean concentrations of both Cl and SO4 observations have been made (Price and Leigh 2006). were both positively related to the HDI gradient with- Turbidity and TSS often are positively correlated to TP in the Oklahoma portion of the LWW, explaining 42% in streams (Stubblefield et al. 2007), which was also the of the variability for Cl and 59% of the variability for case across the streams in the LWW (r=0.70; P<0.001 SO4 (P<0.001; Figure 4I & J). The geomean concentra- and r=0.73; P<0.001; respectively).In addition to the tions of these two anions also showed changepoint re- significant linear relationships, there was also a signifi- sponses to increasing HDI, which were similar between cant threshold response in turbidity and TSS at 30% HDI constituents (15−18% HDI) but slightly less than other (Figure 5F & H), with mean values above the threshold parameters. The average value for the data above the 2.5 to 3.5 times greater than the mean value below the changepoint tended to be 2 to 3 times greater than the threshold. It is interesting that turbidity and TSS, during average of the values below the changepoint line (Fig- base flow conditions were so strongly correlated to HDI ure 5I & J). across these sites.

11 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 5: Change point analysis of geomean concentrations verse human development index (HDI) value for sites in the Oklahoma portion of the Lake Wister Watershed. The vertical line represents the change point values specific to each constituent. The gray box shows the 90% confidence interval about the changepoint. Horizontal bars represent the mean of the data points to the left and right of the change point. The site number in red is Shawnee Creek at highway 59 downstream of effluent discharge was not used in the statistical analysis.

12 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Special Studies Based on work completed during the first year, a few of the subwatersheds or sampling sites were of specific interest: the Little Fourche Maline (site 21) and Oil Branch, one of the four Poteau River tributaries sampled in the first special studies. We sampled additional sites within these subwatersheds to help determine where potential nutrient and sediment sources might be or to confirm the influence of a known specific source. The additional sampling was short-term (n=3) sampled during routine monitoring from January through March 2019 sample periods. During the first project year, measured water chemistry highlighted the Little Fourche Maline site as a medium level priority HUC 12. For the little Fourche Maline subwatershed we wanted to see where among the tributaries were constituent concentrations greatest. So, we sampled four tributaries and one additional site along the main stem upstream of the main site on the Little Fourche Maline (Figure 1). These additional data showed: • Total N, P, and suspended solids (SS), along with SO4 increased with %HDI in the Little Fourche Maline watershed. • Tributaries LFM-2, 3, and 4 tended to have greater constituent concentrations than sites along the main stem (LFM-5 and Site 21; Figure 6). • The Little Fourche Maline watershed north of Hwy 270 is predominantly forested (i.e., 87%) and had relatively low constituent concentrations (Figure 6). These data suggest that the majority of total nutrients and suspended solids enter the Little Fourche Maline watershed between Hwy 270 and the Little Fourche Maline’s confluence with the main stem of the Fourche Maline. Implementation of BMP’s within this portion of the watershed would likely have the greatest effect at reducing constituent concentrations in the Little Fourche Maline. The second subwatershed we focused on was Oil Branch, a tributary of the main stem of the Poteau Riv- er. Previously we had sampled one location in this watershed; site OB-5 which is just upstream of Oil Branch’s confluence with the Poteau River. Previous work at this site found some of the greatest concentrations of dissolved and total N and P across all sites sampled during the first project year. This site and watershed were also Figure 6: Geomean concentrations across the sites within the Little of special interest to PVIA and stakeholders due to the Fourche Maline Watershed sampled from January through March of 2019 for the special study. With geomean concentrations of TN (A); Heavner WWTP (Table 2) which discharges effluent into TP (B); TSS (C); and CHL-a (D). this watershed. Four additional sites were sampled, two

13 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority upstream of the WWTP and two tributaries that flowed into the main stem of Oil Branch after the WWTP (Fig- ure 1). Data from these additional sites show: • Total N and P, and chl-a were greatest at the site below the Heavner WWTP (Figure 7). • TSS was greatest at OB-4 and OB-5, suggesting that the OB-4 tributary is likely a source of suspended solids to the watershed (Figure 7). • There were no significant relationships between the constituent concentrations and %HDI when the point source influenced site (OB-5) is included in the analysis. • With OB-5 excluded, Cl and SO4 are positively related to %HDI; while turbidity, TP, NO3, and NH4 are all negatively related to %HDI. Finer scale sampling of this watershed suggests that the Heavner WWTP is the dominant source of nutrients to the watershed. Across the tributaries of Oil Branch creek only Cl and SO4 related positively with %HDI whereas TP and dissolved N were inversely related to %HDI. Further sampling may be needed to determine why these sites relate to %HDI differently than other sites.

Criteria for Selecting Priority HUC 12s Changepoint analysis is a powerful statistical tool, and one of its most useful aspects is that it gives a threshold, i.e., specific value on the x−axis. In this case, the changepoint gives an HDI value or the proportion of the watershed that is agriculture and urban. This is the point where watershed land use has an influence on water quality, increasing the constituent concentrations. Thus, this information can be used to help design a process from which PVIA and its stakeholders could establish which HUC 12s or smaller subwatersheds are priorities for NPS management. The following sections provide some guidance on how this might be done. In the absence of water quality data at all subwatersheds, specific HDI thresholds can be used to help identify which HUC 12s or smaller watersheds might be a priority for NPS management. The HUC 12s could be prioritized and separated into categories based on the example (Figure 8A). The hypothetical categories could include: • Preservation: HDI<15%; these subwatersheds would be background or reference sites as established by the lower end of the 90th percentile confidence interval about the change- Figure 7: Geomean concentrations across the sites within the Oil points. Branch Watershed sampled from January through March of 2019 for the special study. With geomean concentrations of TN (A); TP • Low priority: HDI from 15-25%; these subwa- (B); TSS (C); and CHL-a (D). The vertical dashed line represents the tersheds would be a low priority for NPS man- Heavener waste water treatment plant.

14 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

agement as established by the lower end of the 90th percentile confidence interval about the changepoint and the changepoint. • Medium priority: HDI from 25-30%; these subwatersheds would be a medium priority for NPS management as established by the changepoint and the upper end of the 90th percentile confidence interval about the changepoint. • High priority: HDI>30%; these subwatersheds would be a high priority for NPS management as established by the upper end of the 90th percentile confidence interval about the changepoint. Based on the LWW stream data, sites with HDI values less than 90th percentile confidence interval about the changepoint had low constituent concentrations (Fig- ure 8A). The goal here would be to keep or preserve these HUC 12s to maintain existing water quality conditions. On the opposite end of the spectrum, streams with HDI values greater than the 90th percentile confidence interval around the change point generally had greater constituent concentrations. So, PVIA and stakeholders might want to focus efforts on HUC 12s with HDI values above 30% when establishing NPS management priorities. If we just use the LULC for each individual HUC 12 (Table 1), then following this classification scheme, HUC 12s along the Fourche Maline and Poteau River would be ranked as medium to high priority and HUC 12s along the southern border of the Wister Wa- Figure 8: Potential methods using changepoints to identify water- tershed would be classified as preservation (Figure 9). sheds for nonpoint source management. Categorization of HUC In the absence of water quality data, this option can be 12s based on their human development index (HDI) value only a good method for selecting HUC 12s when developing (A); separation of HUC 12s based on measured water quality data (B). Linear models (regression line) represent realistic targets for the watershed management plan. improving water quality within a HUC 12 of a given HDI value (C).

Figure 9: Potential prioritization of hydrologic unit code (HUC) 12 subwatersheds based on the threshold response of constituent concentration to the human development index (HDI) as shown in Figure 8A; the priority for nonpoint source management varies from lightest (preservation) to darkest (highest priority). HUC 12 subwatersheds are labeled with the last four digits of their HUC 12 code.

15 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Figure 10: Potential prioritization of HUC 12 subwatersheds when chemical concentrations are available in streams. Using specific constituents to meet specific management needs, or using a cumulative approach, where priorities are added across multiple constituents. For each constituent shown and for the cumulative map the priority for nonpoint source management varies from lightest (low priority) to darkest (highest priority). Each subwatershed is labeled with the last four digits of their HUC 12 code.

When water quality data is available, thresholds As stated earlier, constituent concentrations below the can be used differently to select HUC 12s based on thresholds were generally low. The horizontal dashed measured constituent concentrations as opposed to line provides a realistic bench mark for separating low predicted values that are based on human develop- and medium priority watersheds, as it represents the ment (Figure 8B). This method focuses on the average upper limits of baseline conditions for the constituents constituent concentrations on either side of the thresh- analyzed in this study. This method could be carried out old. The HUC 12s could be prioritized and separated for each constituent of interest, resulting in the selec- into categories based on the example in Figure 8B. The tion of constituent specific HUC 12s (Figure 10). hypothetical categories could include: A weight of evidence approach may be used to • Low priority: HUC 12s with constituent con- combine HUC 12 priorities developed for individual centrations less than average constituent con- constituents. Low, medium, and high priorities can be centration below the threshold plus 2 standard ranked 1, 2, and 3, respectively, for each constituent. deviations (horizontal dashed line; Figure 8B). Rankings for each constituent can then be added to- • Medium priority: HUC 12s with constituent gether to form a cumulative rank for each HUC 12. The concentrations greater than the horizontal cumulative ranks across all HUC 12s within the Oklaho- dashed line but less than the average constit- ma portion of the LWW were divided into 5 categories uent concentration above the threshold (upper where the subwatersheds labeled as the highest priori- solid line; Figure 8B) ty had the highest rank (Figure 10). • High Priority: HUC 12s with constituent con- With this approach you must be mindful of the nest- centrations greater than upper solid line. ed nature of the LWW in that several subwatersheds

16 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority are down river of one or more other subwatersheds. cific equipment does not constitute a guarantee by It is possible that water quality in an upstream subwa- the USDA and does not imply its approv. J Environ tershed may result in higher than expected constituent Qual. 33:2183–2191. concentrations based on the level of human develop- Elphick JRF, Bergh KD, Bailey HC. 2011. Chronic toxicity ment. In this case, it may be beneficial to compare sub- of chloride to freshwater species: Effects of hard- watershed priorities identified by both methods. ness and implications for water quality guidelines. Constituent concentrations change with land use, Environ Toxicol Chem. 30:239–246. where the relation can often be described with a sim- Elphick JR, Davies M, Gilron G, Canaria EC, Lo B, Bai- ple linear model (Figure 4). Once subwatersheds have ley HC. 2011. An aquatic toxicological evaluation been prioritized, the goal should be to move the higher of sulfate: The case for considering hardness as a priority HUC 12s below the linear regression which rep- modifying factor in setting water quality guidelines. resents the average conditions for a given HDI (Figure Environ Toxicol Chem. 30:247–253. 8C). The methods should follow previous routine mon- EPA U. 2000. Atlas of America’s polluted waters. EPA itoring methods used to develop these relationships, 840-B00-002. Washington D.C. where 12 monthly base flow samples should be used Evans-White MA, Haggard BE, Scott JT. 2013. A Review to determine an annual geomean concentration data of Stream Nutrient Criteria Development in the point. The data point should be plotted against the United States. J Environ Qual. 42:1002–1014. most current land use information available, to reflect Giovannetti J, Massey LB, Haggard BE, Morgan RA. the changing LULC and HDI gradient. Once the data 2013. Land use effects on stream nutrients at Bea- point shifts from above the line to below the line, then ver Lake Watershed. J Am Water Work Assoc. 105 this site has reached its target concentration as defined Haggard BE, Moore Jr PA, Chaubey I, Stanley EH. 2003. by the original regression. However, it would be wise Nitrogen and phosphorus concentrations and - ex to make sure the HUC 12s have consistently changed port from an Ozark Plateau catchment in the Unit- priority categories (e.g., moved from high to low) over ed States. Biosyst Eng. 86:75–85. multiple years before assuming the end point has been Haggard BE, Scott JT, Longing SD. 2013. Sestonic chloro- met. phyll-a shows hierarchical structure and thresholds with nutrients across the Red River Basin, USA. J En- REFERENCES viron Qual. 42:437–445. Austin BJ, Patterson S, Haggard BE. 2018. Water Chem- Haggard BE, Stanley EH, Storm DE. 2005. Nutrient re- istry During Baseflow Helps Inform Watershed tention in a point-source-enriched stream. J North Management: A Case Study of the Lake Wister Wa- Am Benthol Soc. 24:29–47. tershed, Oklahoma. :42–58. Honti M, Istvánovics V, Kovács ÁS. 2010. Balancing be- Austin BJ, Smith BA, Haggard BE. 2018. Watershed In- tween retention and flushing in river networks— vestigative Support to the Poteau Valley Improve- optimizing nutrient management to improve tro- ment Authority: Stream Water Quality to Support phic state. Sci Total Environ. 408:4712–4721. HUC 12 Prioritization in the Lake Wister Watershed, Kemp MJ, Dodds WK. 2002. The influence of ammoni- Oklahoma. AWRC MSC385.:1–35. um, nitrate, and dissolved oxygen concentrations Chambers PA, McGoldrick DJ, Brua RB, Vis C, Culp JM, on uptake, nitrification, and denitrification rates Benoy GA. 2012. Development of Environmen- associated with prairie stream substrata. Limnol tal Thresholds for Nitrogen and Phosphorus in Oceanogr. 47:1380–1393. Streams. J Environ Qual. 41:7–20. Leasure DR. 2013. Geodata Crawler. Cox TJ, Engel BA, Olsen RL, Fisher JB, Santini AD, Ben- McCarty JA, Haggard BE. 2016. Can We Manage Non- nett BJ. 2013. Relationships between stream phos- point-Source Pollution Using Nutrient Concentra- phorus concentrations and drainage basin charac- tions during Seasonal Baseflow? Ael. 1:0. teristics in a watershed with poultry farming. Nutr McCarty JA, Matlock MD, Scott JT, Haggard BE. 2018. Cycl Agroecosystems. 95:353–364. Risk Indicators for Identifying Critical Source Areas DeLaune PB, Moore PA, Carman DK, Sharpley AN, Hag- in Five Arkansas Watersheds. Trans ASABE. gard BE, Daniel TC. 2004. Development of a Phos- Merbt SN, Auguet J-C, Casamayor EO, Marti E. 2011. phorus Index for Pastures Fertilized with Poultry Biofilm recovery in a wastewater treatment Litter—Factors Affecting Phosphorus Runoff Men- plant-influenced stream and spatial segregation of tion of a trade name, proprietary product, or spe- ammonia-oxidizing microbial populations. Limnol

17 Arkansas Water Resources Center | Publication MSC389 Funded by the Poteau Valley Improvement Authority

Oceanogr. 56:1054–1064. novative Measures for the Control of Agricultural Migliaccio KW, Srivastava P. 2007. Hydrologic compo- Phosphorus Losses to Water: An Overview. J Envi- nents of watershed-scale models. Trans ASABE. ron Qual. 29:1–9. 50:1695–1703. Smith VH, Tilman GD, Nekola JC. 1999. Eutrophication: Morisawa M. 1968. Streams: Their dynamics and mor- impacts of excess nutrient inputs on freshwater, phology. McGraw-Hill Company, New York, NY marine, and terrestrial ecosystems. Environ Pollut. Price K, Leigh DS. 2006. Comparative Water Quality 100:179–196. of Lightly- and Moderately-Impacted Streams in Stubblefield AP, Reuter JE, Dahlgren RA, Goldman CR. the Southern Blue Ridge Mountains, USA. Environ 2007. Use of turbidometry to characterize sus- Monit Assess. 120:269–300. pended sediment and phosphorus fluxes in the Schoumans OF, Chardon WJ, Bechmann ME, Gas- Lake Tahoe basin, California, USA. Hydrol Process. cuel-Odoux C, Hofman G, Kronvang B, Rubæk GH, 21:281–291. Ulén B, Dorioz J-M. 2014. Mitigation options to- re Sun H, Alexander J, Gove B, Pezzi E, Chakowski N, Husch duce phosphorus losses from the agricultural sec- J. 2014. Mineralogical and anthropogenic controls tor and improve surface water quality: A review. Sci of stream water chemistry in salted watersheds. Total Environ. 468–469:1255–1266. Appl Geochemistry. 48:141–154. Sharpley A, Foy B, Withers P. 2000. Practical and In-

18 APPENDIX 1: QA/QC report

Appendix 1: QA/QC summary for each constituent, including field blanks and field duplicates.

Field Blanks Field Duplicates Average Constituent Units RL Average % Pass % Pass %RPD TN mg L-1 0.05 0.01 100 10 93 TP mg L-1 0.020 0.003 100 5 98 -1 NO3-N mg L 0.01 0.003 90 12 84* -1 NH4-N mg L 0.01 0.001 100 10 93 SRP mg L-1 0.005 0.001 100 7 98 Turbidity NTU NA 0.1 100 4 97 TSS mg L-1 4 0.1 100 6 93 Chl-a µg L-1 NA 0.0 100 17 84* Cl mg L-1 0.50 0.1 100 2 100 -1 SO4 mg L 0.5 0.9 65 7 95 *Constituents with field duplicates that did not pass the defined criteria in the

QAPP. All samples with a high %RPD (>30%) for both NO3-N and TSS had measured values below the MDL, which can make it difficult to attain a %RPD<30.

APPENDIX 2: Summary Data The complete data set for August 2017 – May 2019 is available in the data report (DR-WQ-MSC388)

Appendix 2A: Total Nitrogen geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 0.673 0.324 1.745 0.691 0.353 1.532 0.680 0.461 1.745 0.642 0.324 1.208 2 0.703 0.356 3.123 0.794 0.368 3.123 0.644 0.471 1.472 0.648 0.356 1.025 3 0.463 0.148 4.943 0.449 0.148 4.943 0.525 0.353 0.759 0.427 0.221 0.861 4 0.103 0.050 0.248 0.104 0.050 0.227 0.116 0.050 0.248 0.090 0.058 0.137 5 0.097 0.050 0.228 0.119 0.064 0.228 0.101 0.050 0.169 0.072 0.050 0.148 6 0.113 0.050 0.253 0.127 0.072 0.176 0.138 0.056 0.253 0.078 0.050 0.177 7 0.168 0.056 0.325 0.209 0.137 0.325 0.197 0.123 0.262 0.110 0.056 0.297 8 0.553 0.222 1.745 0.681 0.420 1.257 0.544 0.254 1.206 0.446 0.222 1.745 9 0.228 0.061 1.045 0.308 0.158 1.045 0.247 0.143 0.903 0.150 0.061 0.466 10 0.159 0.081 0.301 0.157 0.085 0.282 0.179 0.111 0.278 0.145 0.081 0.301 11 0.184 0.059 0.291 0.170 0.059 0.268 0.212 0.115 0.291 0.176 0.127 0.258 12 0.205 0.079 0.570 0.187 0.079 0.570 0.246 0.171 0.363 0.192 0.120 0.301 13 0.153 0.060 0.957 0.194 0.097 0.957 0.167 0.087 0.244 0.109 0.060 0.153 14 0.124 0.051 0.344 0.131 0.051 0.329 0.119 0.060 0.228 0.123 0.061 0.344 15 0.254 0.127 0.450 0.235 0.127 0.329 0.300 0.217 0.450 0.238 0.136 0.438 16 0.299 0.110 0.715 0.375 0.213 0.715 0.302 0.163 0.535 0.225 0.110 0.456 17 0.340 0.100 0.951 0.379 0.176 0.951 0.359 0.143 0.776 0.283 0.100 0.448 18 0.147 0.050 0.777 0.182 0.073 0.761 0.145 0.056 0.636 0.114 0.050 0.777 19 0.530 0.314 0.804 0.547 0.421 0.804 0.523 0.314 0.724 0.514 0.337 0.734 20 0.565 0.274 1.534 0.610 0.274 1.534 0.512 0.305 1.355 0.568 0.334 1.388 21 0.387 0.124 0.750 0.503 0.308 0.750 0.364 0.171 0.718 0.300 0.124 0.501 22 0.590 0.281 1.064 0.672 0.408 1.064 0.564 0.345 0.820 0.515 0.281 0.754 23 1.455 0.693 2.639 1.497 0.759 2.639 1.590 1.093 2.401 1.281 0.693 2.330 24 0.479 0.141 1.120 0.548 0.322 0.763 0.517 0.341 1.120 0.367 0.141 0.670 25 0.335 0.127 0.817 0.379 0.249 0.569 0.389 0.241 0.817 0.248 0.127 0.515 26 0.469 0.122 4.368 0.457 0.240 1.270 0.810 0.261 4.368 0.278 0.122 1.064 19

Appendix 2B: Total Phosphorus geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 0.062 0.022 0.194 0.064 0.027 0.160 0.052 0.022 0.145 0.072 0.036 0.194 2 0.068 0.029 0.399 0.079 0.029 0.399 0.055 0.030 0.114 0.069 0.039 0.130 3 0.051 0.022 0.856 0.058 0.022 0.856 0.051 0.024 0.093 0.042 0.022 0.104 4 0.021 0.003 0.053 0.018 0.003 0.048 0.019 0.003 0.053 0.027 0.020 0.037 5 0.018 0.003 0.041 0.020 0.003 0.037 0.014 0.003 0.041 0.021 0.020 0.022 6 0.018 0.003 0.059 0.018 0.003 0.038 0.015 0.003 0.059 0.021 0.020 0.023 7 0.026 0.020 0.049 0.027 0.020 0.047 0.027 0.020 0.049 0.023 0.020 0.030 8 0.082 0.027 0.472 0.093 0.047 0.314 0.072 0.027 0.189 0.082 0.038 0.472 9 0.031 0.020 0.183 0.033 0.020 0.059 0.034 0.020 0.183 0.027 0.020 0.049 10 0.020 0.003 0.043 0.023 0.020 0.035 0.016 0.003 0.043 0.022 0.020 0.024 11 0.018 0.003 0.028 0.019 0.003 0.023 0.019 0.003 0.028 0.018 0.003 0.021 12 0.019 0.003 0.168 0.023 0.003 0.168 0.016 0.003 0.036 0.018 0.003 0.024 13 0.019 0.003 0.037 0.020 0.003 0.034 0.013 0.003 0.031 0.026 0.020 0.037 14 0.020 0.003 0.077 0.023 0.020 0.031 0.015 0.003 0.020 0.025 0.020 0.077 15 0.024 0.020 0.045 0.025 0.020 0.044 0.025 0.020 0.045 0.022 0.020 0.025 16 0.033 0.020 0.067 0.036 0.020 0.067 0.033 0.020 0.064 0.031 0.022 0.043 17 0.035 0.020 0.070 0.035 0.020 0.067 0.035 0.020 0.070 0.034 0.020 0.054 18 0.024 0.003 0.044 0.024 0.003 0.035 0.020 0.003 0.035 0.028 0.020 0.044 19 0.072 0.026 0.141 0.074 0.026 0.141 0.075 0.049 0.111 0.068 0.049 0.097 20 0.052 0.020 0.509 0.065 0.027 0.509 0.053 0.026 0.147 0.038 0.020 0.068 21 0.056 0.024 0.161 0.069 0.030 0.161 0.055 0.024 0.123 0.043 0.026 0.087 22 0.070 0.033 0.117 0.072 0.033 0.116 0.076 0.047 0.117 0.062 0.047 0.098 23 0.178 0.089 0.670 0.209 0.089 0.670 0.183 0.132 0.245 0.138 0.098 0.195 24 0.053 0.027 0.111 0.056 0.027 0.104 0.061 0.033 0.111 0.044 0.028 0.075 25 0.033 0.020 0.084 0.036 0.020 0.062 0.034 0.020 0.084 0.028 0.020 0.044 26 0.152 0.020 1.028 0.191 0.020 0.790 0.265 0.069 1.028 0.069 0.024 0.178

Appendix 2C: Nitrate-N geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 0.10 0.01 1.11 0.09 0.01 0.71 0.09 0.01 1.11 0.13 0.01 0.37 2 0.06 0.01 0.76 0.04 0.01 0.63 0.04 0.01 0.76 0.15 0.01 0.46 3 0.08 0.01 0.37 0.06 0.01 0.36 0.05 0.01 0.37 0.16 0.04 0.34 4 0.02 0.01 0.16 0.03 0.01 0.16 0.02 0.01 0.15 0.02 0.01 0.03 5 0.01 0.01 0.05 0.01 0.01 0.05 0.01 0.01 0.03 0.01 0.01 0.02 6 0.02 0.01 0.05 0.02 0.01 0.05 0.03 0.01 0.05 0.02 0.01 0.04 7 0.02 0.01 0.07 0.02 0.01 0.07 0.01 0.01 0.05 0.01 0.01 0.05 8 0.02 0.01 0.27 0.01 0.01 0.20 0.03 0.01 0.27 0.04 0.01 0.18 9 0.03 0.01 0.64 0.04 0.01 0.64 0.03 0.01 0.15 0.02 0.01 0.03 10 0.03 0.01 0.09 0.02 0.01 0.06 0.02 0.01 0.09 0.03 0.01 0.07 11 0.05 0.01 0.18 0.06 0.01 0.14 0.03 0.01 0.18 0.08 0.02 0.18 12 0.05 0.01 0.20 0.04 0.01 0.11 0.04 0.01 0.20 0.07 0.01 0.13 13 0.06 0.01 0.89 0.09 0.02 0.89 0.06 0.01 0.18 0.04 0.02 0.10 14 0.02 0.01 0.17 0.02 0.01 0.17 0.02 0.01 0.12 0.02 0.01 0.10 15 0.05 0.01 0.20 0.04 0.01 0.15 0.04 0.01 0.20 0.09 0.03 0.17 16 0.03 0.01 0.20 0.03 0.01 0.18 0.02 0.01 0.16 0.05 0.01 0.20 17 0.05 0.01 0.51 0.03 0.01 0.51 0.03 0.01 0.47 0.09 0.01 0.28 18 0.02 0.01 0.45 0.02 0.01 0.41 0.02 0.01 0.44 0.01 0.01 0.45 19 0.04 0.01 0.31 0.03 0.01 0.16 0.03 0.01 0.27 0.08 0.01 0.31 20 0.05 0.01 0.95 0.04 0.01 0.34 0.02 0.01 0.26 0.15 0.02 0.95 21 0.02 0.01 0.13 0.03 0.01 0.09 0.02 0.01 0.12 0.03 0.01 0.13 22 0.09 0.01 0.50 0.10 0.01 0.50 0.05 0.01 0.31 0.14 0.06 0.28 23 0.33 0.01 1.53 0.19 0.01 1.37 0.45 0.20 1.06 0.55 0.21 1.53 24 0.02 0.01 0.10 0.01 0.01 0.06 0.01 0.01 0.07 0.03 0.01 0.10 25 0.03 0.01 0.21 0.04 0.01 0.09 0.03 0.01 0.09 0.02 0.01 0.21 26 0.19 0.05 2.28 0.18 0.05 0.65 0.28 0.06 2.28 0.13 0.06 0.79

Appendix 2D: Ammonium-N geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 0.02 0.01 0.09 0.03 0.01 0.09 0.02 0.01 0.05 0.02 0.01 0.04 2 0.02 0.01 0.10 0.02 0.01 0.10 0.02 0.01 0.03 0.01 0.01 0.05 3 0.02 0.01 0.06 0.02 0.01 0.06 0.02 0.01 0.03 0.01 0.01 0.03 4 0.01 0.01 0.03 0.01 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.02 5 0.01 0.01 0.03 0.01 0.01 0.03 0.01 0.01 0.03 0.01 0.01 0.02 6 0.01 0.01 0.03 0.01 0.01 0.02 0.01 0.01 0.03 0.01 0.01 0.03 7 0.01 0.01 0.03 0.02 0.01 0.03 0.01 0.01 0.02 0.01 0.01 0.01 8 0.03 0.01 0.17 0.03 0.01 0.09 0.03 0.01 0.12 0.02 0.01 0.17 9 0.02 0.01 0.11 0.02 0.01 0.06 0.02 0.01 0.11 0.01 0.01 0.03 10 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 11 0.01 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.01 0.01 12 0.01 0.01 0.06 0.02 0.01 0.06 0.01 0.01 0.02 0.01 0.01 0.03 13 0.01 0.01 0.03 0.01 0.01 0.03 0.01 0.01 0.03 0.01 0.01 0.01 14 0.01 0.01 0.05 0.01 0.01 0.05 0.01 0.01 0.01 0.01 0.01 0.03 15 0.01 0.01 0.04 0.01 0.01 0.04 0.01 0.01 0.04 0.01 0.01 0.03 16 0.02 0.01 0.04 0.02 0.01 0.04 0.02 0.01 0.04 0.01 0.01 0.03 17 0.02 0.01 0.05 0.02 0.01 0.05 0.02 0.01 0.05 0.02 0.01 0.05 18 0.01 0.01 0.05 0.01 0.01 0.03 0.01 0.01 0.05 0.01 0.01 0.02 19 0.03 0.01 0.13 0.03 0.01 0.13 0.03 0.01 0.06 0.02 0.01 0.05 20 0.03 0.01 0.15 0.03 0.01 0.08 0.03 0.01 0.11 0.03 0.01 0.15 21 0.03 0.01 0.24 0.05 0.01 0.16 0.04 0.01 0.24 0.02 0.01 0.11 22 0.03 0.01 0.13 0.04 0.01 0.13 0.03 0.01 0.10 0.02 0.01 0.08 23 0.12 0.01 1.09 0.17 0.02 1.01 0.12 0.01 1.09 0.09 0.03 0.41 24 0.03 0.01 0.32 0.04 0.01 0.21 0.03 0.01 0.32 0.02 0.01 0.12 25 0.02 0.01 0.07 0.03 0.01 0.07 0.02 0.01 0.07 0.01 0.01 0.03 26 0.02 0.01 0.26 0.02 0.01 0.06 0.03 0.01 0.26 0.01 0.01 0.03

Appendix 2E: Soluble Reactive Phosphorus geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 0.019 0.004 0.123 0.019 0.004 0.086 0.016 0.004 0.032 0.023 0.004 0.123 2 0.014 0.004 0.080 0.015 0.004 0.080 0.011 0.004 0.022 0.017 0.004 0.069 3 0.012 0.004 0.093 0.013 0.004 0.093 0.010 0.004 0.025 0.013 0.004 0.064 4 0.006 0.004 0.011 0.006 0.004 0.011 0.006 0.004 0.007 0.006 0.004 0.010 5 0.005 0.004 0.007 0.005 0.004 0.007 0.006 0.004 0.007 0.005 0.004 0.005 6 0.005 0.004 0.009 0.005 0.004 0.007 0.006 0.004 0.009 0.005 0.004 0.005 7 0.005 0.004 0.007 0.005 0.004 0.007 0.006 0.004 0.007 0.005 0.004 0.005 8 0.026 0.004 0.235 0.027 0.007 0.130 0.022 0.004 0.101 0.031 0.010 0.235 9 0.008 0.004 0.081 0.010 0.004 0.021 0.009 0.004 0.081 0.006 0.004 0.011 10 0.005 0.004 0.007 0.006 0.004 0.007 0.006 0.004 0.007 0.005 0.004 0.006 11 0.005 0.004 0.006 0.005 0.004 0.006 0.005 0.004 0.004 0.005 0.004 0.005 12 0.006 0.004 0.052 0.007 0.004 0.052 0.005 0.004 0.006 0.006 0.004 0.007 13 0.006 0.004 0.010 0.006 0.004 0.007 0.006 0.004 0.009 0.007 0.004 0.010 14 0.005 0.004 0.008 0.005 0.004 0.007 0.006 0.004 0.008 0.005 0.004 0.005 15 0.006 0.004 0.007 0.006 0.004 0.007 0.006 0.004 0.006 0.006 0.004 0.007 16 0.007 0.004 0.015 0.008 0.004 0.015 0.007 0.004 0.013 0.007 0.004 0.012 17 0.007 0.004 0.020 0.008 0.004 0.018 0.008 0.004 0.020 0.007 0.004 0.011 18 0.006 0.004 0.012 0.006 0.004 0.012 0.005 0.004 0.007 0.006 0.004 0.008 19 0.019 0.004 0.064 0.020 0.007 0.064 0.025 0.004 0.054 0.015 0.007 0.032 20 0.013 0.004 0.116 0.019 0.004 0.116 0.014 0.004 0.046 0.008 0.004 0.024 21 0.017 0.004 0.063 0.021 0.008 0.063 0.021 0.004 0.057 0.010 0.004 0.022 22 0.020 0.004 0.067 0.023 0.012 0.038 0.026 0.004 0.067 0.012 0.005 0.027 23 0.033 0.004 0.173 0.042 0.008 0.173 0.033 0.004 0.104 0.025 0.010 0.052 24 0.014 0.004 0.048 0.013 0.005 0.048 0.018 0.004 0.041 0.010 0.004 0.017 25 0.007 0.004 0.023 0.008 0.004 0.023 0.007 0.004 0.010 0.006 0.004 0.007 26 0.097 0.004 0.715 0.126 0.004 0.703 0.189 0.046 0.715 0.037 0.014 0.130

Appendix 2F: Turbidity geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in NTU.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 10.6 2.9 41.0 9.6 3.5 39.1 7.3 2.9 26.6 17.6 4.7 41.0 2 14.6 4.7 33.0 14.7 5.2 33.0 12.9 6.8 27.1 16.5 4.7 24.4 3 14.2 5.6 30.7 15.4 6.9 30.7 14.5 7.3 23.5 12.5 5.6 17.8 4 10.4 1.2 33.9 6.0 1.2 33.9 9.8 3.0 25.8 21.2 9.0 31.6 5 7.6 1.7 26.1 5.7 1.7 26.1 7.1 2.7 16.6 11.7 5.4 16.6 6 7.8 2.1 28.2 5.7 2.1 28.2 7.0 2.4 18.9 12.7 4.8 18.9 7 9.5 3.5 29.1 7.3 3.5 29.1 9.3 4.5 19.9 13.5 5.9 22.7 8 15.7 5.1 168.0 13.5 5.1 168.0 13.2 5.7 33.5 22.2 10.0 31.6 9 9.0 1.7 146.0 6.8 1.7 25.5 9.6 1.7 146.0 11.5 4.0 24.8 10 7.8 2.2 17.6 8.3 3.3 14.1 5.7 2.2 15.1 9.9 2.8 17.6 11 5.5 2.1 19.4 5.8 3.4 9.5 4.0 2.1 6.3 7.2 3.6 19.4 12 6.3 1.3 84.1 6.9 2.0 84.1 4.3 1.3 9.9 8.2 2.2 13.8 13 9.1 2.4 25.9 6.6 2.4 13.7 6.9 3.0 15.9 17.1 11.1 25.9 14 10.1 4.3 22.6 7.9 4.4 22.6 8.1 4.3 14.7 16.4 11.1 21.2 15 8.5 3.7 18.5 7.9 3.7 18.0 7.4 4.1 15.9 10.5 6.2 18.5 16 14.1 4.7 49.6 12.6 4.7 49.6 11.9 5.4 27.0 19.1 9.2 28.0 17 11.6 2.4 50.3 9.3 2.4 50.3 9.1 4.1 27.3 19.2 6.2 29.6 18 11.9 2.1 39.6 12.3 2.1 39.6 8.2 2.6 24.8 16.6 6.8 28.1 19 29.3 10.1 104.0 31.6 10.5 104.0 27.0 16.5 52.7 28.7 10.1 43.6 20 18.0 3.2 431.0 19.9 3.6 431.0 16.3 3.2 45.7 17.6 6.7 30.5 21 15.4 5.3 47.1 18.5 7.4 47.1 11.1 5.3 21.1 17.2 9.2 26.6 22 21.2 4.2 73.2 21.5 4.2 73.2 19.6 7.4 32.7 22.5 7.1 36.4 23 52.7 20.5 154.0 57.1 21.8 154.0 50.9 20.5 87.6 48.9 23.5 71.8 24 13.7 6.1 28.4 12.3 6.2 27.1 13.7 6.1 21.5 16.1 8.1 28.4 25 17.4 5.9 46.0 17.2 5.9 46.0 17.6 10.3 28.6 17.4 11.2 22.0 26 19.2 2.4 35.7 17.9 2.4 32.2 19.2 9.6 31.2 20.8 14.3 35.7

Appendix 2G: Total Suspended Solids geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 5.3 1.1 21.8 5.2 1.1 15.6 4.7 1.8 10.3 6.4 1.5 21.8 2 7.8 3.0 22.8 8.5 3.9 16.4 7.7 3.6 13.8 7.1 3.0 22.8 3 6.4 1.1 17.2 7.3 1.1 17.2 8.1 2.8 13.3 4.2 2.0 15.1 4 1.5 0.4 10.8 1.2 0.5 3.0 1.7 0.4 3.4 1.8 0.9 10.8 5 1.2 0.0 3.5 1.5 0.7 3.4 1.6 0.7 3.5 0.7 0.0 2.9 6 1.8 0.3 5.1 1.6 0.3 3.4 2.1 1.1 5.1 1.9 1.1 2.8 7 2.4 0.2 6.2 3.0 0.8 5.1 3.3 1.7 6.2 1.3 0.2 4.1 8 5.0 2.0 48.2 6.8 2.7 48.2 4.6 2.5 11.0 3.9 2.0 10.8 9 2.3 0.2 20.9 2.8 0.6 5.5 2.8 0.5 20.9 1.6 0.2 3.1 10 1.9 0.6 4.9 1.9 0.6 4.9 1.8 0.6 3.9 2.1 1.1 3.3 11 1.6 0.4 4.0 1.6 0.8 4.0 1.8 0.7 2.7 1.4 0.4 3.5 12 1.7 0.4 51.6 2.2 0.9 51.6 1.6 0.9 3.4 1.2 0.4 3.6 13 1.9 0.6 5.1 1.9 0.8 5.1 2.1 0.9 4.8 1.7 0.6 3.2 14 1.6 0.4 5.5 1.8 1.1 3.0 2.0 1.1 5.5 1.1 0.4 3.1 15 2.6 0.6 6.8 2.8 0.6 6.8 3.0 1.6 4.4 2.0 0.7 6.4 16 3.5 1.2 14.8 4.7 2.2 14.8 3.3 1.2 7.1 2.6 1.6 4.5 17 2.9 0.7 14.7 3.7 1.6 14.7 2.4 0.7 5.0 2.8 1.3 5.0 18 2.0 0.4 12.5 1.8 0.4 7.6 2.5 1.0 12.5 1.8 0.5 6.4 19 11.9 4.2 80.8 14.4 5.1 80.8 9.0 4.2 17.0 11.9 4.3 25.3 20 5.8 1.1 158.1 6.0 1.1 158.1 6.0 1.5 22.9 5.4 2.4 10.6 21 5.4 0.5 20.8 8.4 4.0 20.8 4.0 0.5 10.4 4.4 2.0 8.4 22 8.9 2.7 58.1 11.2 4.2 58.1 7.2 2.7 13.4 8.0 3.6 18.2 23 28.9 8.5 100.1 30.7 9.8 100.1 28.8 8.5 58.9 26.5 10.3 57.5 24 6.3 2.0 15.3 6.9 2.0 15.3 6.5 4.0 8.9 5.4 2.4 12.5 25 5.3 1.7 14.7 5.6 1.7 14.7 6.3 3.3 10.9 4.2 2.4 9.2 26 5.3 1.4 22.4 6.1 1.4 12.3 7.1 2.4 22.4 3.3 1.6 5.9

Appendix 2H: Chlorophyll-a geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in µg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 4.4 0.5 35.3 5.8 0.5 26.4 5.4 1.1 35.3 2.5 1.2 17.9 2 7.1 0.5 46.7 10.5 2.0 46.7 10.0 1.3 34.8 3.0 0.5 17.8 3 4.0 0.7 27.2 4.3 0.7 27.2 8.8 3.3 18.0 1.6 0.7 3.7 4 0.5 0.1 3.2 0.6 0.4 1.5 0.4 0.1 1.0 0.4 0.1 3.2 5 0.5 0.1 2.7 0.8 0.3 1.6 0.5 0.2 2.7 0.2 0.1 1.9 6 0.7 0.1 2.7 1.0 0.4 2.4 0.9 0.4 2.7 0.3 0.1 1.3 7 1.4 0.1 10.2 2.9 0.6 7.4 2.2 0.3 10.2 0.4 0.1 7.5 8 3.9 0.5 38.8 9.3 1.6 36.9 4.3 0.8 38.8 1.4 0.5 28.7 9 0.9 0.0 16.5 1.7 0.4 16.5 1.3 0.2 9.4 0.3 0.0 2.1 10 0.8 0.2 2.4 0.9 0.3 2.4 0.9 0.3 1.9 0.5 0.2 1.3 11 0.5 0.0 6.3 0.5 0.1 3.6 1.0 0.2 6.3 0.2 0.0 0.8 12 0.6 0.2 2.9 0.7 0.3 2.9 0.8 0.2 1.7 0.4 0.2 1.2 13 0.6 0.1 1.5 0.7 0.3 1.5 0.6 0.4 1.0 0.6 0.1 1.1 14 0.6 0.1 7.8 0.7 0.4 7.8 0.5 0.1 2.1 0.6 0.2 1.5 15 1.3 0.2 13.3 1.4 0.5 13.3 2.1 0.5 6.0 0.7 0.2 2.8 16 1.7 0.2 11.6 2.7 0.9 11.1 2.8 0.6 11.6 0.6 0.2 4.0 17 1.7 0.2 7.5 2.8 1.5 5.1 2.1 0.2 5.2 0.8 0.2 7.5 18 0.8 0.1 23.5 1.4 0.2 23.5 1.1 0.3 8.6 0.3 0.1 19.4 19 2.3 0.4 12.8 3.3 0.4 9.8 2.8 0.7 12.8 1.1 0.7 2.1 20 2.1 0.4 11.1 3.1 1.1 11.1 1.8 0.7 4.0 1.5 0.4 10.0 21 1.7 0.3 24.5 2.8 0.8 10.3 1.7 0.4 9.2 0.9 0.3 24.5 22 3.3 0.5 48.8 6.5 1.4 48.8 2.9 0.5 12.7 1.5 0.7 15.7 23 13.4 0.7 113.5 12.6 0.7 113.5 18.5 1.5 91.7 10.7 2.7 62.5 24 3.1 0.6 38.2 6.4 1.5 38.2 2.6 0.6 20.5 1.4 0.6 4.6 25 2.1 0.2 62.8 1.9 0.8 5.0 3.3 0.5 62.8 1.5 0.2 20.9 26 3.7 0.6 32.3 2.9 0.6 18.8 6.0 1.7 32.3 3.1 1.2 9.9

Appendix 2I: Chloride geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 8.6 2.2 83.9 13.3 6.7 83.9 8.5 3.4 20.6 4.7 2.2 10.0 2 8.1 2.2 59.9 12.5 6.1 59.9 7.5 3.4 19.9 4.8 2.2 10.3 3 4.8 2.0 19.3 5.1 2.9 19.3 5.9 2.6 16.4 3.8 2.0 7.3 4 2.7 1.6 4.2 3.0 1.6 4.2 2.9 2.3 3.4 2.2 1.7 2.7 5 3.1 2.1 6.6 3.7 2.2 6.6 3.1 2.5 3.6 2.6 2.1 5.1 6 3.1 1.9 6.3 3.5 1.9 6.3 3.2 2.6 3.9 2.7 2.3 5.2 7 3.4 1.7 7.4 3.7 1.7 7.4 3.5 2.8 4.6 3.0 2.4 7.1 8 6.3 2.0 16.2 8.4 4.5 16.2 6.9 3.2 14.0 4.1 2.0 8.9 9 8.8 4.2 88.4 14.1 5.0 88.4 7.8 4.5 12.4 5.8 4.2 9.5 10 2.3 1.3 3.2 2.2 1.3 2.9 2.4 1.9 3.2 2.3 1.7 3.0 11 2.2 1.6 3.3 2.2 1.8 2.7 2.4 1.7 3.3 2.0 1.6 2.3 12 2.3 1.4 3.4 2.3 1.4 2.9 2.5 1.9 3.4 2.3 1.7 2.6 13 4.9 2.4 14.2 6.5 4.1 14.2 5.4 3.3 9.8 3.3 2.4 4.0 14 2.8 1.6 6.3 3.2 1.6 6.3 3.0 2.2 4.1 2.2 1.9 2.6 15 2.6 1.7 4.6 2.5 1.7 3.0 2.8 2.0 4.6 2.5 1.7 3.1 16 3.2 1.7 6.2 3.3 1.7 6.2 3.4 2.7 5.3 2.9 2.4 3.9 17 3.3 1.5 8.0 3.2 1.5 5.8 3.6 2.8 8.0 3.2 2.4 5.8 18 4.7 2.4 11.2 5.0 3.1 11.2 5.5 4.2 7.3 3.7 2.4 5.1 19 8.9 3.0 21.6 9.8 3.0 21.6 9.6 5.2 17.9 7.3 6.0 10.7 20 6.5 3.5 14.6 6.1 3.5 14.6 7.7 4.7 11.8 5.8 4.7 8.5 21 12.6 0.1 36.4 13.4 3.3 36.4 10.6 0.1 28.9 14.1 9.2 33.4 22 10.7 3.2 32.3 10.7 3.2 21.6 12.0 6.7 32.3 9.5 6.6 14.3 23 10.7 0.5 27.0 9.9 0.5 27.0 13.8 6.2 20.0 9.4 6.1 25.0 24 14.7 3.5 153.4 17.2 3.5 153.4 15.6 5.2 37.9 11.0 5.5 48.5 25 7.1 2.3 20.8 8.0 2.3 20.8 6.9 4.3 10.3 6.2 4.3 14.1 26 5.4 2.2 28.3 6.3 3.7 16.6 8.1 2.9 28.3 3.1 2.2 4.8

Appendix 2J: Sulfate geometric mean, minimum, and maximum values for each site in the Lake Wister Watershed for the overall project (all three project years), and each individual project year. All values are reported in mg L-1.

Overall Year 1 Year 2 Year 3 Site no. Geomean Min Max Geomean Min Max Geomean Min Max Geomean Min Max 1 11.1 6.5 34.1 13.1 6.6 34.1 10.5 6.8 18.4 9.4 6.5 12.5 2 14.0 7.8 36.7 15.9 9.2 36.7 14.6 8.0 32.4 11.3 7.8 15.1 3 10.8 0.1 89.9 10.1 4.5 39.0 10.4 0.1 89.9 12.5 7.5 25.0 4 6.5 4.4 16.0 6.9 4.4 16.0 5.9 5.3 7.3 6.7 4.7 7.6 5 5.3 3.5 16.1 5.4 3.5 16.1 4.4 3.8 5.1 6.2 4.5 8.9 6 5.7 3.4 8.5 5.5 3.4 8.5 5.4 4.4 7.1 6.4 5.4 8.4 7 6.2 2.8 11.0 5.4 2.8 11.0 6.1 4.9 7.2 7.6 5.2 9.0 8 8.2 2.9 22.7 7.0 2.9 21.6 9.0 4.7 22.7 9.0 6.2 13.8 9 28.4 14.3 292.2 39.3 14.8 292.2 25.2 14.3 45.6 22.3 16.3 35.4 10 4.0 2.3 7.3 3.6 2.6 5.6 3.9 3.3 5.1 4.7 2.3 7.3 11 2.9 1.6 5.2 2.4 1.6 3.0 2.7 2.2 3.8 4.0 2.8 5.2 12 3.4 1.9 6.0 2.9 1.9 3.9 3.0 2.4 4.2 4.6 3.2 6.0 13 8.7 5.9 86.1 7.7 5.9 9.2 10.0 6.8 86.1 8.5 6.5 9.4 14 4.7 3.1 6.9 4.3 3.1 6.2 4.6 3.4 5.7 5.3 3.5 6.9 15 4.1 2.7 6.7 3.3 2.7 4.1 4.5 3.2 6.4 5.0 3.9 6.7 16 5.8 2.8 9.9 4.3 2.8 8.5 5.9 3.8 8.6 8.1 4.9 9.9 17 5.4 1.5 14.2 3.8 1.5 8.3 5.3 3.0 14.2 8.5 5.6 11.8 18 10.0 4.5 17.4 10.6 7.6 17.4 9.9 6.5 12.9 9.5 4.5 12.5 19 15.3 6.8 61.5 13.4 6.8 23.5 17.8 9.3 61.5 15.8 10.5 20.3 20 37.3 15.2 76.5 35.7 16.7 76.5 36.8 15.2 55.5 40.0 26.8 60.8 21 11.8 3.0 25.1 9.3 3.0 24.7 12.6 5.3 25.1 14.8 9.3 18.9 22 16.6 6.6 32.6 16.1 6.6 32.6 16.1 11.4 28.2 17.7 14.5 22.4 23 24.4 0.1 45.3 28.5 11.7 45.3 18.4 0.1 40.1 26.3 22.2 32.1 24 7.4 1.7 31.9 7.1 1.7 31.9 5.8 2.1 13.9 9.7 6.9 13.2 25 8.6 3.7 15.9 8.6 3.7 15.9 7.9 5.0 11.5 9.1 6.3 11.6 26 8.0 4.8 34.3 7.1 4.9 17.0 10.5 5.0 34.3 7.0 4.8 8.3