E. John Sadler

Long-term agroecosystem research in the central Mississippi river basin: goodwater creek experimental watershed flow data.

Knowledge of weather, particularly precipitation, is fundamental to interpreting watershed and hydrologic processes. The long-term weather record in the Goodwater Creek Experimental Watershed (GCEW) complements hydrologic and water quality data in the region. The GCEW also is the core of the Central Mississippi River Basin (CMRB) node of the Long-Term Agroecosystem Research network. Our objectives are to (i) describe the climatological context of the GCEW and CMRB settings, (ii) document instrumentation and the data collection, quality assurance, and reduction processes; (iii) provide examples of the data obtained and descriptive statistics; and (iv) document the availability of and access methods to obtain the data from the web-based data access portal at . These objectives support an overall goal to make these long-term data available to the public for use in further analyses and modeling in support of research and public policy on watershed management.

Vapor pressure deficit calculations and their effect on the combination equation

Abstract Of the several models used to calculate potential evapotranspiration (PET), many researchers use the combination method because of its theoretical basis. This model can be affected by random errors in the input parameters (net radiation, air temperature, wind speed, and daily average vapor pressure deficit, ▿) and sensitivity analyses have described the impact of these errors. However, a more subtle non-random error may be introduced in PET estimates by changing the form by which the ▿ term is specified. At least 12 different ways to present ▿ have been published; the primary differences among them are the measured humidity parameter and the algebra used to compute ▿. The effect of all applicable published computational methods on monthly and seasonal PET values for a range of locations differing in evaporative demand was examined in this study. Related methods of computing ▿ resulted in little difference between PET values. The range of summer PET means obtained from the extreme methods was 8–17% of the best estimate method over all locations. Although this range approximates the expected accuracy of the combination method, it must be stressed that the net effect of the systematic and random errors may constitute a bias and, therefore, should be evaluated as such. Apparently innocuous computational differences can significantly affect PET results and, therefore, degrade confidence in the resulting values.

Nutrient Sources and Transport from the Goodwater Creek Experimental Watershed

The Goodwater Creek watershed has been monitored for flow since 1971 and for dissolved nutrients since 1991 for 3 nested watersheds (12.1, 31.5 and 73.0 km2 drainage area). This watershed includes row crop land (76%), grassland (14%), woodland (6%) and a small town at the upper end (4%). The objectives of this paper are to analyze nutrient loadings at the 3 gauging stations from 1991 to 2004. Daily, monthly and annual loadings and flow-weighted concentrations of ammonium-nitrogen, nitrate-nitrogen, dissolved phosphorus and atrazine were calculated and analyzed using the non parametric tests for differences, homogeneity, and trends. Atrazine was included in the analysis as one compound not implicated with point source discharges in the watershed. Dissolved phosphorus and ammonium-nitrogen concentrations and loads at the upstream weir were significantly greater than those at the two downstream weirs, which suggest wastewater was a potential source of these nutrients. Possible explanations for these differences were drawn from our knowledge of the watershed and tested with a SWAT model of the watershed. These findings provide insight to what should be included in a complete analysis of the nutrient sources in the watershed and how stream processes affect nutrient loadings.

Estimating Water Quality with Airborne and Ground-Based Hyperspectral Sensing

Remotely sensed estimates of water quality parameters would facilitate efforts in spatial and temporal monitoring. In this study we collected hyperspectral water reflectance data with airborne and ground-based sensing systems for multiple arms of Mark Twain Lake, a large manmade reservoir in northeast Missouri. Water samples were also collected and analyzed in the laboratory for chlorophyll, nutrients, and turbidity. Wavelength-selection (i.e., stepwise multiple regression) methods and previously reported indices were used to develop relationships between spectral and water quality data. Within the single measurement date of this study, all measured water quality parameters were strongly related (R2 > 0.6) to reflectance data from the ground system. Relationships between water quality parameters and airborne reflectance data were generally somewhat lower, but still with R2 > 0.6. Previously developed narrow-band reflectance indices also worked well to estimate chlorophyll concentration. Wide-band, multispectral reflectance, simulating Landsat data, was strongly related only to turbidity and those other parameters (e.g., phosphorus) highly correlated to turbidity in this dataset. Thus, hyperspectral sensing, coupled with calibration sampling, can be used to estimate lake water quality differences, and appears to have advantages over multispectral sensing in this application.

Optimal Levels of Irrigation in Corn Production in the Southeast Coastal Plain

ABSTRACT Water is a precious resource and is used in many competing industries. To use water efficiently in crop production, knowledge about crop responses to irrigation water, or the production function, is essential. In this paper, we estimated six production functions, two N-fertilizer treatments for each year in 1999, 2000, and 2001, for corn production using the data from experimental plots in Florence, South Carolina, USA. Optimal levels of irrigation and gross margins under profit-maximizing and yield-maximizing strategies were computed. The results indicate that at the current prices of corn and water in South Carolina, the profit-maximizing strategy conserved more irrigation water and produced larger gross margins than the yield-maximizing strategy. The differences in optimal levels of irrigation water and gross margins between the two strategies became even more significant when the relative water/corn price ratios increased. To find out how demand for irrigation water responds to changes in water prices, demand functions for water were derived and demand elasticities of water were computed. At the current prices of water and corn, the demand elasticities were inelastic, which means that irrigation is not very responsive to changes in the price of water. As the price of water increased, demand for irrigation became more responsive to changes in water prices.

Long-Term Agroecosystem Research in the Central Mississippi River Basin: Hyperspectral Remote Sensing of Reservoir Water Quality

In situ methods for estimating water quality parameters would facilitate efforts in spatial and temporal monitoring, and optical reflectance sensing has shown potential in this regard, particularly for chlorophyll, suspended sediment, and turbidity. The objective of this research was to develop and evaluate relationships between hyperspectral remote sensing and lake water quality parameters-chlorophyll, turbidity, and N and P species. Proximal hyperspectral water reflectance data were obtained on seven sampling dates for multiple arms of Mark Twain Lake, a large man-made reservoir in northeastern Missouri. Aerial hyperspectral data were also obtained on two dates. Water samples were collected and analyzed in the laboratory for chlorophyll, nutrients, and turbidity. Previously reported reflectance indices and full-spectrum (i.e., partial least squares regression) methods were used to develop relationships between spectral and water quality data. With the exception of dissolved NH, all measured water quality parameters were strongly related ( ≥ 0.7) to proximal reflectance across all measurement dates. Aerial hyperspectral sensing was somewhat less accurate than proximal sensing for the two measurement dates where both were obtained. Although full-spectrum calibrations were more accurate for chlorophyll and turbidity than results from previously reported models, those previous models performed better for an independent test set. Because extrapolation of estimation models to dates other than those used to calibrate the model greatly increased estimation error for some parameters, collection of calibration samples at each sensing date would be required for the most accurate remote sensing estimates of water quality.

A methodology to reduce uncertainties in the high-flow portion of the rating curve for Goodwater Creek Watershed

Flow monitoring at watershed scale relies on the establishment of a rating curve that describes the relationship between stage and flow and is developed from actual flow measurements at various stages. Measurement errors increase with out-of-bank flow conditions because of safety concerns and difficulties in measuring flow velocity in the flood plain. With increasing frequency of large rain events associated with climate change, it becomes critical to have an accurate rating curve for out-of-bank flows. We propose a methodology based on the Manning formula on one hand and physical and mathematical principles that govern the hydrograph and the rating curve on the other hand to test and develop the high flow limb of a rating curve. The methodology was developed and tested using flow data from the Goodwater Experimental Watershed, a 72-km2 watershed in Northeast Missouri. A final water balance on flow events was performed using measured precipitation, and evapotranspiration and soil water estimated from a SWAT model. Preliminary results indicate that flow values that correspond to out-of-bank stage were overestimated by 30% in the original rating curve. This method provides additional and innovative means to verify the accuracy of the high limb of a rating curve.

Using Precision Agriculture Methods to Predict Soil Suitability for Rainfed Corn Production

The need for additional corn to meet the high demand for bio-fuels in the US will likely lead to more rainfed corn production in the Mid-South. The high degree of soil variability in the area suggests that some fields would be better suited to such a production system than others. Methods are being developed to use precision agriculture data to indicate suitable fields to reduce the risk of crop failure. A field study was conducted at the University of Missouri Delta Research Center Marsh Farm at Portageville to relate information obtained with precision agriculture methods to rainfed corn yields. Initially soil apparent electrical conductivity (ECa) and surface elevation were investigated along with soil mapping units. Yield differences did not correspond well to county soil survey map units. Including relative elevation (RE) in a quadratic equation of ECa provided only a slightly better equation than the ECa terms only. The range of yield values that could be predicted from ECa was not sufficient to adequately describe the observed yields. The study is continuing in 2008 and additional information will be collected.

Precision Management Systems for Conservation and Profitability

The uses of site-specific information (e.g., yield maps, soil sampling) can extend beyond variable-rate management of inputs to managing parts of a field in completely different ways (e.g., cropping system, conservation measures). In this paper, we discuss the development and initial evaluation of such a precision management system for a typical claypan-soil field in Missouri. For more than a decade (1991-2003), we intensively monitored crop, soil, and water quality information on this field, finding considerable spatial variation in crop yield and many other properties. We used maps of these properties to develop a crop management plan addressing site-specific problems. In 2004, we implemented the Precision Agriculture System (PAS) plan, using precision information to determine what production and conservation measures were needed, and where they should be placed. For example, shallow topsoil areas of the field were no longer planted to corn receiving soil-applied herbicides. Instead, these areas were planted to wheat and a cover crop, usually without herbicides. The goal of PAS is to improve profitability and protect soil and water resources through management of sub-field areas based on multiple spatial datasets. Preliminary results after three years of implementation show progress toward this goal, such as a reduction in soil loss from erosive sideslope areas. Although definitive results will depend on more years of data collection, it appears that the PAS approach may help producers to improve both profitability and conservation.