John C. Panuska

Sediment and Phosphorus Losses in Snowmelt and Rainfall Runoff from Three Corn Management Systems

Limited comparisons of sediment and phosphorus (P) loss dynamics from agricultural fields under snowmelt and rainfall runoff conditions exist despite significant differences in underlying particle detachment and transport processes during these two periods. A systems approach was used on three hydrologically isolated hillslope tracts from which rainfall runoff and snowmelt data were collected over an 18-month period under different residue and manure management systems: corn-grain, corn-silage, and corn-silage with fall manure application. Particulate-bound P dominated overall losses for all the monitored events. While higher mass loads of sediments and P were exported in rainfall runoff, the concentrations of dissolved P forms and organic matter were higher in snowmelt. During the rainfall runoff period, both sediment and P losses were inversely related to the percent residue cover, with the highest coverage corn-grain site producing the lowest levels. In contrast, manure P input prior to melt events, rather than percent residue, dictated P loss patterns during the snowmelt period. Consequently, median P and volatile solids concentrations in snowmelt were higher for the manured site than from the non-manured sites. Importantly, the dissolved P load from the manured field was higher during the snowmelt period compared to the rainfall runoff period. Differences in organic matter sources (manure vs. crop residue) produced sediments with contrasting solids-P signatures, with those from manured corn-silage site enriched both in volatile solids and P. Interestingly, the residue cover-manure interplay and the mode of runoff generation had no significant effect on the sediment and P mass distribution in various particle size classes. Our results are expected to improve understanding of P loss pathways and facilitate development of better predictive tools for P transport by enhancing insight on sediment and P mass distribution in different size classes under contrasting row-crop production systems and from different modes of runoff generation.

Testing the Wisconsin Phosphorus Index with Year-Round, Field-Scale Runoff Monitoring

The Wisconsin Phosphorus Index (WPI) is one of several P indices in the United States that use equations to describe actual P loss processes. Although for nutrient management planning the WPI is reported as a dimensionless whole number, it is calculated as average annual dissolved P (DP) and particulate P (PP) mass delivered per unit area. The WPI calculations use soil P concentration, applied manure and fertilizer P, and estimates of average annual erosion and average annual runoff. We compared WPI estimated P losses to annual P loads measured in surface runoff from 86 field-years on crop fields and pastures. As the erosion and runoff generated by the weather in the monitoring years varied substantially from the average annual estimates used in the WPI, the WPI and measured loads were not well correlated. However, when measured runoff and erosion were used in the WPI field loss calculations, the WPI accurately estimated annual total P loads with a Nash-Sutcliffe Model Efficiency (NSE) of 0.87. The DP loss estimates were not as close to measured values (NSE = 0.40) as the PP loss estimates (NSE = 0.89). Some errors in estimating DP losses may be unavoidable due to uncertainties in estimating on-farm manure P application rates. The WPI is sensitive to field management that affects its erosion and runoff estimates. Provided that the WPI methods for estimating average annual erosion and runoff are accurately reflecting the effects of management, the WPI is an accurate field-level assessment tool for managing runoff P losses.

Positive Effects of Agricultural Land Use Changes on Coldwater Fish Communities in Southwest Wisconsin Streams

Abstract The Conservation Reserve Program (CRP) is a federal program that encourages the planting of cool- or warm-season grass cover on highly erodible croplands and along stream corridors. We sought to determine whether fish community structure in coldwater streams was associated with CRP and other agricultural land use changes in southwestern Wisconsin. We compared coldwater fish index of biotic integrity (IBI) scores and species richness in streams located in areas of relatively high (21.3% of land area; high-CRP area) versus relatively low (12.1% of land area; low-CRP area) CRP participation. All of the streams were sampled in the 1970s before implementation of the CRP and again at the same locations after implementation, from 2000 to 2005. Pre-CRP fish communities were characterized by a relatively high diversity of eurythermal species and low coldwater IBI scores. We found significant increases in coldwater IBI scores over time in streams within the high-CRP area relative to streams within the low-...

Effects of Hypolimnetic Releases on Two Impoundments and Their Receiving Streams in Southwest Wisconsin

Abstract The effects of bottom water withdrawals were evaluated within and below two southwestern Wisconsin impoundments. Like many man-made lakes in the unglaciated area of Wisconsin, Twin Valley Lake and White Mound Lake were constructed during the late 1960s for flood control and recreation. Both impoundments, which are located in large agricultural watersheds, were originally designed to release cold bottom water with the intended goal of managing trout below the dams. However, recent water quality monitoring results revealed that the streams became degraded due to frequent dissolved oxygen criterion violations and excessive filamentous bacteria growths. Organic loading from the bottom discharges is the likely reason that trout stream habitat was not successfully created below the dams as originally intended. Despite the accelerated phosphorus removal from the long-term withdrawals, blue-green algal blooms continued to be a problem in both impoundments. While maximizing total phosphorus export can improve lake water quality conditions, discharge rates from these impoundments were found to be excessively high, resulting in disturbance of their thermo-structure and entrainment of nutrients into the surface waters. In 2005, we blocked the bottom gate at Twin Valley Lake and monitored water quality and thermal responses. As a result, lake and stream water quality improved significantly while the downstream fish community structure did not change. The impoundment thermo-structure was restored with well-defined hypolimnion and epilimnion. We conclude that managing impoundments is often a balancing act between seemingly disparate goals of achieving optimum conditions above or below a dam, with undesirable consequences often occurring if the focus is disproportionately on a single goal.

Consequences of Selecting Incorrect Hydrologic Parameters When Using the Walker Pond Size and P8 Urban Catchment Models

ABSTRACT The Walker Pond Size Model is a widely used spreadsheet approach to the design of storm water treatment ponds. The user inputs the Soil Conservation Service curve number for the pervious areas and the fraction of impervious area. Input parameters must be correctly determined by the user. One commonly made error is to enter the weighted curve number in place of the pervious curve number, which can result in overestimation of dead storage volume by as much as 8.89 mm (0.35 inch). Using the weighted curve number only and not the models equation can underestimate the required volume as much as 19.56 mm (0.77 inch). Proper use of the Walker Model is essential to obtain a beneficial, cost-effective design.

Testing a two-scale focused conservation strategy for reducing phosphorus and sediment loads from agricultural watersheds

This study tested a focused strategy for reducing phosphorus (P) and sediment loads in agricultural streams. The strategy involved selecting small watersheds identified as likely to respond relatively quickly, and then focusing conservation practices on high-contributing fields within those watersheds. Two 5,000 ha (12,360 ac) watersheds in the Driftless Area of south central Wisconsin, previously ranked in the top 6% of similarly sized Wisconsin watersheds for expected responsiveness to conservation efforts to reduce high P and sediment loads, were chosen for the study. The stream outlets from both watersheds were monitored from October of 2006 through September of 2016 for streamflow and concentrations of sediment, total P, and, beginning in October of 2009, total dissolved P. Fields and pastures having the highest potential P delivery to the streams in each watershed were identified using the Wisconsin P Index (Good et al. 2012). After three years of baseline monitoring (2006 to 2009), farmers implemented both field- and farm-based conservation practices in one watershed (treatment) as a means to reduce sediment and P inputs to the stream from the highest contributing areas, whereas there were no out-of-the-ordinary conservation efforts in the second watershed (control). Implementation occurred primarily in 2011 and 2012. In the four years following implementation of conservation practices (2013 through 2016), there was a statistically significant reduction in storm-event suspended sediment loads in the treatment watershed compared to the control watershed when the ground was not frozen (p = 0.047). While there was an apparent reduction in year-round suspended sediment event loads, it was not statistically significant at the 95% confidence level (p = 0.15). Total P loads were significantly reduced for runoff events (p < 0.01) with a median reduction of 50%. Total P and total dissolved P concentrations for low-flow conditions were also significantly reduced (p < 0.01) compared to the control watershed. This study demonstrated that a strategy that first identifies watersheds likely to respond to conservation efforts and then focuses implementation on relatively high-contributing fields within those watersheds can be successful in reducing stream P concentrations and loads.

Identifying Sources of Suspended Sediment using Radionuclides in an Agricultural Watershed in South Central Wisconsin

Phosphorus (P) is an essential nutrient for plant and livestock growth. However, P loss in agricultural runoff can increase the frequency of toxic algal blooms and fish kills in receiving waters. Agricultural P loss occurs in both dissolved and particulate (sediment bound) forms. Suspended sediments play an important role in the transport of particulate P from fields to surface waters. Implementing appropriate management practices to control soil erosion and subsequent sediment delivery requires quantification of the relative contribution of sediment sources (e.g. stream bed, stream bank and upland areas under various land uses). Sediment fingerprinting using atmospheric fallout radionuclides can be used to apportion sediment sources, and thus provide valuable guidance for management decisions. Due to their long half-lives, the fallout radionuclides 137Cs and unsupported 210Pb are ideally suited for evaluating sediment transport processes that occur over long time scales. This fingerprinting method is independent of soil and rock type and can be used to differentiate between surficial and channel sources of suspended sediments. The objective of this study was to identify sources of in-stream suspended sediment in an agricultural watershed using the atmospheric fallout radionuclides 137Cs and 210Pb. The study was conducted in the non-glaciated region of southwestern Wisconsin in the Sugar Pecatonica River Basin, which is part of the Upper Mississippi River Basin. The watershed is approximately 5000 ha in size and contains primarily agriculture, forest, and grass land cover. The average watershed slope is about 11% with silt loam soils. Fieldwork included collection of both source materials (upland, streambed, and stream bank) and in-stream suspended sediments. In-stream suspended sediment samples were collected monthly for four months using passive time integrated in-stream tube samplers (Phillips et al., 2000). The samplers consist of a 10.2 cm diameter PVC tube with 0.4 cm diameter inlet and outlet, and collect a sample that is statistically representative of the grain size distribution in small streams. All source material samples were collected from the top 2.5 cm. Upland soil samples were collected from fields that represented various combinations of land use, soil type, and slope within the watershed. Upland samples were collected in a 20 m x 20 m grid with 5 m spacing and composited for analysis. Representative samples were also collected from the top 2.5cm of stream beds and eroding stream banks. All samples collected were stored at 40 °C and analyzed for organic matter content (percent volatile solids) and 137Cs and unsupported 210Pb. Radionuclide analysis was done through low background gamma counters. Over a four month period (mid-April through mid-August, 2010), results indicate that approximately two-thirds of in-stream suspended sediment originated from eroding stream banks and the remainder from upland areas. Within the upland categories (cultivated, pasture, woodland, grassland), cultivated lands followed by woodlands were significant contributors to in-stream sediments.