Search USDA's Current Research Information System (CRIS) Database

Item No. 1 of 13

INVESTIGATOR: Schultz, R. C.; Isenhart, T. M.; Raich, J. W.; Simpkins, W. W.

PERFORMING INSTITUTION:
FORESTRY
IOWA STATE UNIVERSITY
AMES, IOWA 50011

IMPROVING SOIL AND WATER QUALITY WITH A MULTI-SPECIES RIPARIAN BUFFER STRIP

OBJECTIVES: 9501579. Our specific project objectives are: 1) to demonstrate the effectiveness of a constructed multi-species riparian buffer strip system (CMRPBS) in retaining and/or transforming nitrate in vadose zone and shallow ground waters; 2) to evaluate mechanisms of nitrate retention and transformation within the CMRBS by: a) assessing rates of denitrification and the potential for denitrification as affected by vegetation and soil properties, b) assessing the importance of plant immobilization of nitrogen in different vegetation types, and c) assessing the importance and potential of microbial immobilization of nitrate. 3) to quantify soil quality parameters potentially regulating water movement, microbial activities, and overall rates of nitrate transformation, and to investigate their relationships to vegetation cover.

APPROACH: These goals will be achieved through intensive sampling and experimental approaches focused on artificial buffer strips in central Iowa. These sites have been the focus of research by the Iowa State Agroforestry Research Team for the past four years and background data on water and agricultural chemical fluxes, and on rates of plant establishment in the buffer strips, provides a valuable information base upon which we will build. We will measure rates of denitrification, potential denitrification activity, N immobilization in above-and belowground plant materials, potential rates of microbial nitrate immobilization, and soil quality characteristics (Bulk density, infiltration rates, rooting depths, organic matter contents, C:N ratios) in each of three replicates of each of three transects in each of two different types of streamside buffer strips. The CMRBs consists of fast-growing trees placed closest to the stream, then several rows of shrubs, and a wide strip of native grasses established next to the agricultural field. This 20 m-wide buffer strip is being compared to a cool season grass buffer strip of the same width. Study results will be integrated using the soil-plant ecosystem model CENTURY.

PROGRESS: 1995/09 TO 1999/08
A sixteen-meter-wide multi-species riparian buffer was planted in 1990 consisting of 5 rows of trees adjacent to the stream, 2 rows of shrubs adjacent to the trees and a 7-meter-wide strip of native warm-season grass between the shrubs and cropfield. Seven years after establishment the buffer removed 95% of the sediment, 94% of total-N, 85% of nitrate nitrogen, 91% of total-P and 80% of phosphate phosphorus from surface runoff. Total soil carbon increases since the buffer establishment were 123% under trees, 85% under switchgrass and 61% under cool-season grasses. Particulate organic matter showed similar increases. Fine root and microbial biomass were three times higher in buffer soil than in crop soil. As a result, soil respiration rates were twice as high in the buffer soil as in the crop soil. Soil infiltration rates under the buffer were four times as high as in crop soils. Nitrate in the soil water of the unsaturated zone was reduced by up to 90% as it crossed through the buffer. Four years after establishment, denitrification rates in the soil under trees were significantly higher than in crop soil, but rates under switchgrass were not different from the crop soil. After seven years the denitrification rates under both trees and switchgrass were more than four times higher than in the crop soil and not significantly different from those in soil under cool-season grass. Removal of nitrate from the groundwater depends on the hydrogeologic setting of the aquifer. In this project, 10-15% of the nitrate was removed from the groundwater, mainly near the water table. In thinner aquifers, up to 90% of the nitrate might be removed. Nitrate removal seems limited by available carbon in the groundwater, and work has been started to assess the production of dissolved organic carbon under different plant communities. People in the Bear Creek watershed 1) want clean surface and ground water, 2) perceive that multi-species riparian buffers are acceptable and functional best management practices (BMPs), and 3) are willing to pay for improvements in surface water quality. Landowners in the Bear Creek watershed rated the 1993 surface water quality at a 6.0 (on a 0 to 10 water quality ladder scale with 10 representing best quality for human use). They wanted to improve the surface water quality to 8.3. This represents an improvement of over 30%. To achieve this improvement they perceive that > 50% reductions in sediment, fertilizer and herbicide pollutants are required. In 1993, landowners (farmers and non-farmers) were willing to pay ~ $48 per year to improve surface water quality. A 1997 focus group of farmers indicated a willingness to pay (WTP) of $25 per year to improve surface water quality. Bear Creek landowners believe that a mix of upland (field) BMPs and riparian buffers will improve water quality the most. Considering the Bear Creek watershed and a 20-year WTP present value (@5% and $48 per year) of $610,000, a 10 to 20 m wide buffer could be placed along the entire stream on both sides. Such a buffer would provide additional non-market goods and services such as hunting and aesthetics, which are highly valued by landowners.

IMPACT: 1995/09 TO 1999/08
Buffers developed in this project have served as a template for the NRCS Riparian Forest Buffer Conservation Standard - 392 that is being applied along streams nationwide. The research has demonstrated that the buffers can significantly reduce non-point source pollution from cropfields and pastures. Landowners value these buffers and have shown strong support for their installation.

PUBLICATIONS: 1995/09 TO 1999/08
1. Johnston, D.A. 1998. Hydrogeology and geology of three restored riparian buffers near Roland, Iowa. M.S. Thesis. Iowa State University.
2. Andress, R.J. 1999. Fate and transport of nitrate in groundwater within a riparian buffer in the Bear Creek watershed. M.S. Thesis. Iowa State University.
3. Lee, K.H. 1999. Effectiveness of a multi-species riparian buffer system for sediment and nutrient removal. Ph.D. Dissertation. Iowa State University.
4. Lee, K.H., Isenhart, T.M., Schultz, R.C. and Mickelson, S.K. 1999. Nutrient and sediment removal by switchgrass and cool-season grass filter strips in Central Iowa, USA. Agroforestry Systems 44:121-132.
5. Marquez, C.O., Cambardella, C.A., Isenhart, T.M., and Schultz, R.C. 1999. Assessing soil quality in a riparian buffer by testing organic matter fractions in central Iowa, USA. Agroforestry Systems 44:133-140
6. Pickle, J. 1999. Microbial nitrate immobilization in a multi-species riparian buffer. M.S. Thesis. Iowa State University.
7. Szymanski, M. and Colletti, J. 1999. Combining the socio-economic-cultural implications of community owned agroforetry: the Winnebago Tribe of Nebraska. Agroforestry Systems 44:227-239.
8. Tufekcioglu, A., Raich, J.W., Isenhart, T.M., and Schultz, R.C. 1999. Root biomass, soil respiration, and root distribution in crop fields and riparian buffer zones. Agroforestry Systems 44:163-174.
9. Schultz, R.C., Isenhart, T.M., Colletti, J.P., and Marquez, C.O. 2000. Integrated riparian management systems to protect water quality. Chapter 7 in B. Rietveld and G. Garrett (eds.) Agroforestry: An integrated Science and Practice. American Society of Agronomy, Inc., Madison, WI. USA (In press).

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 2 of 13

INVESTIGATOR: Tim, U. S.; Kanwar, R. S.; Batchelor, W.; Babcock, B. A.; Mallarino, A.

PERFORMING INSTITUTION:
AGRI & BIOSYSTEMS ENGINEERING
IOWA STATE UNIVERSITY
AMES, IOWA 50011

INTEGRATED ASSESSMENT OF ENVIRONMENTAL AND ECONOMIC IMPLICATION OF PRECISION FARMING ON CROP PRODUCTION

OBJECTIVES: 9703966. Our research goals will be evaluated by the following specific aims: to collect data for evaluating the effects of variable rate application of fertilizer and atrazine herbicide on water quality; to develop methods to quantify the spatial distribution of cup yields and the impact of variable rate application on chemical losses to surface and groundwater; to evaluate the economic consequences of precision-farming and variable rate technology so that farmers can make informed management decisions.

APPROACH: This study will collect soil and water quality data at the Northeast Research Farm in Nashua to facilitate determination of variable rate application of nitrogen (N) and atrazine on leaching and runoff losses, uptake of N, and crop yields. Methods will be developed to characterize the spatial distribution of crop yields and the quantitative assessment, the Root Zone Water Quality model will be linked with a GIS to enable characterization and display of the spatial distribution of crop yield and chemical losses under site-specific chemical management. The study will also develop trade-off frontiers and measure the economic inputs to more site-specific and precise management, net farm returns change, and the need to identify the critical amounts of variability that justify a given level of investement invariable rate technology increases. Overall, the approaches used and data collected in this project will make a significant contribution to improved understanding of the economic and egronomic, and water quality benefits of precision farming.

PROGRESS: 1997/11 TO 2002/11
In this project, a systems approach was used to evaluate the agronomic, environmental, and economic implications of precision agriculture. A comprehensive problem-solving and decision-support system that improves analyses, simulation and visualization of field-scale impacts of precision agriculture practices (e.g, variable rate nutrient and pesticide management) on environmental quality and productivity has been developed and tested. The system combines biophysical modeling provided by RZWQM and CEREES-Maize models, (RZWQM 98), S-PLUS, and ArcView GIS. Components of the system have been validated and AgLink for Windows SSToolbox and many other programs. The problem solving environment decision support system also features an economic analysis component that provides risk-based estimates and trade-offs of effects of variable rate nutrient application on productivity and profitability of the farm operations. The problem-solving environment/decision support system is used to assess different options for implementing site-specific and nutrient and pesticide management practices. The modeling environment has been used to assess different combinations of climate, landscape, and management regimes on the agronomic and environmental benefits of precision agriculture.

IMPACT: 1997/11 TO 2002/11
The significance of this research to production agriculture is threefold. First, it enables us to move beyond the lumped treatment of ecological processes in agricultural fields and creates an integrated approach needed to facilitate the evolution towards site-specific management of crop production inputs. Second, it addresses the research needs articulated in the 1997 National Research Council on the scientific basis of precision agriculture. Third, the integrated decision support system significantly improves the use of computer models and the evaluation of "What if" scenarios to elucidate the optimal combination of soil, crops chemicals, terrain, and weather that enhances farm productivity and reduces off-site environmental impacts on a site-specific basis.

PUBLICATIONS: 1997/11 TO 2002/11
1. Tim US and X Wang. 1999. Integrated Spatial Decision Support System for Precision Resource Management. Proceeding of Conference on Geosolutions: Integrated Our World, GIS '99, March 1-4, Vancouver, BC, Canada.
2. Wang X and US Tim. 2003. Mining Factor Effects on Spatial Structure of Corn Yield. Journal of Agricultural, Biological and Environmental Statistics (in press).
3. Wang X and US Tim. 2003. Evaluating the environmental and agronomic implications of variable rate nitrogen management. Transactions of the ASAE (in press).
4. Tim US. 2003. Precision Agriculture and Water Quality. Encyclopedia of Agricultural, Food and Biological Engineering. New York, NY: Marcel Dekker, Inc.
5. Warnemuende EA and RS Kanwar. 2002. Effects of swine manure application on bacterial quality of leachate from intact soil columns. Transactions of the American Society of Agricultural Engineers 45(6):1849-1857
6. Bakhsh A, RS Kanwar, TB Bailey, CA Camberdella, DL Karlen and TS Colvin. 2002. Cropping systems effects on NO3-N loss with subsurface drainage water. TRANSACTIONS of the American Society of Agricultural Engineers 45(6):1789-1797.
7. Chung S, PW Gassman, R Gu and RS Kanwar. 2002. Evaluation of EPIC for assessing tile flow and nitrogen losses for alternative agriculture management systems. Transactions of the American Society of Agricultural Engineers 45(4):1135-1146.
8. Ella VB, SW Melvin, RS Kanwar, L Jones and R Horton. 2002. Inverse three-dimensional groundwater modeling using finite difference method for recharge estimation in a glacial till aquitard. Transactions of the American Society of Agricultural Engineers 45(3):703-715.
9. Chinkuyu AJ, RS Kanwar, JC Lorimor, H Xin and TB Bailey. 2002. Effects of laying hen manure application rate on water quality. Transactions of the American Society of Agricultural Engineers 45(2):299-308.

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 3 of 13

INVESTIGATOR: Isenhart, T. M.; Parkin, T. B.; Simpkins, W. W.; Schultz, R. C.

PERFORMING INSTITUTION:
NATURAL RESOURCE ECOLOGY & MANAGEMENT
IOWA STATE UNIVERSITY
AMES, IOWA 50011

ASSESSMENT AND PREDICTION OF THE FATE OF NITRATE IN RE-ESTABLISHED RIPARIAN BUFFERS

OBJECTIVES: The goal of this work is to obtain a process level understanding of the mechanisms of nitrate transport and transformation in riparian buffers re-established on previously cropped or pastured land. Specific objectives of this work are: 1) to utilize a groundwater and vadose zone monitoring network to define the spatial and temporal heterogeneity and evaluate mechanisms of nitrate attenuation in re-establied riparian buffers, 2) to assess a chronosequence of re-established riparian buffer vegetation of 0 to 9 years of age for nitrate attenuation and compare these with riparian zones under crops, cool-season grasses, and mixed timber, and 3) to develop a decision-aid for assessing the efficacy of re-established riparian buffers to attenuate nitrate.

APPROACH: This research is part of an ongoing project to develop a riparian management system model for agricultural watersheds in the Midwestern US. The focus of the riparian restoration is the Bear Creek Watershed where we have already re-established over 6.5 km of streamlength to riparian buffer since 1990. This research will utilize existing groundwater and vadose zone monitoring networks to assess nitrate flux in these buffers and will determine the relative importance of dilution, denitrification, or plant and microbial uptake in subsurface nitrate attenuation. Nitrate flux will also be assessed within a chronosequence of re-established buffers and riparian zones in crops, cool-season grasses and mixed timber using multi-level piezometers and direct-push groundwater monitoring. N transformations will be related to surrogate variables such as soil type, vegetation cover, depth to water table, soil organic C and N content, pH, and topography to develop a decision-aid tool to a priori assess the efficacy of established buffers in removing nitrate.

PROGRESS: 1998/12 TO 2002/12
This research is part of an ongoing project to develop a riparian management system model for agricultural watersheds in the Midwestern US. The focus of the riparian restoration is the Bear Creek Watershed in Central Iowa, where riparian buffer has been re-established on over 6.5 km of stream length since 1990. The sum of this research suggests that riparian buffers re-established on previously cropped or pastured land have tremendous potential to remediate nonpoint source pollution in agricultural ecosystems. One component of this research was to determine the effectiveness of an established multi-species buffer in trapping sediment, nitrogen, and phosphorus from cropland runoff. Results during natural rainfall events demonstrate that a 16.3 m wide switchgrass/woody buffer removed 97% of the sediment, 94% of the nitrate-nitrogen, 91% of the total phosphorus, and 80% of the phosphate-phosphorus. A second component of the research has been to assess the cascading impacts of the composition and age of the re-established riparian vegetation on soil quality and nitrogen biogeochemistry within riparian soils. When using soil-water infiltration as an index, the established multi-species buffer vegetation improved soil quality after six years, with average 60-min cumulative infiltration five times greater under the buffers than under cultivated fields and pasture. Soil bulk densities under the multi-species buffer vegetation were also significantly smaller than in the crop fields and pasture. Soil respiration, microbial biomass, aggregate stability and potential denitrification rates were significantly greater within re-established riparian buffers than in cropped fields, indicating greater soil biological activity within the buffers. Riparian buffers have been shown to remove nitrate from groundwater, but the processes controlling removal are not well documented. Previous research in the Bear Creek Watershed suggests that geology influences groundwater velocity, residence time, denitrification rate, and ultimately how well the buffer functions. A multi-electrode electrical resistivity imaging system was used to characterize the extent and distribution of alluvial materials beneath buffers. Seven locations were selected for further groundwater investigations based on resistivity data and buffer maturity. Hydraulic gradient and hydraulic conductivity data from multilevel piezometers were used to assess controlling factors on nitrate removal in buffers. Several buffers consistently removed over 95 percent of nitrate. However, a few sites often had little effect on nitrate removal. Nitrate removal was favored in locations with available dissolved organic carbon and low groundwater velocities (long residence time). Lack of dissolved oxygen in these locations suggests denitrification as the removal mechanism. Based on data from the six sites, the water quality benefits of buffers are most dependent on geology, groundwater residence time and geochemical environment. Five graduate students were partially or wholly supported on this project.

IMPACT: 1998/12 TO 2002/12
This research demonstrates that riparian buffers re-established on previously cropped or pastured land have tremendous potential to remediate nonpoint source pollution in agricultural ecosystems. The research also provides insight into the geomorphic, hydrologic, and biologic factors controlling the effectiveness of streamside buffers. This information is being used to improve the performance of buffers in the field and help realize the promise of conservation buffer technology.

PUBLICATIONS: 1998/12 TO 2002/12
1. Lee K, TM Isenhart and RC Schultz. 2003. Sediment and nutrient removal in an established multi-species riparian buffer. Journal of Soil and Water Conservation 58(1):1-8.
2. Bharati L, K-H Lee, TM Isenhart and RC Schultz. 2002. Riparian zone soil-water infiltration under crops, pasture and established buffers. Agroforestry Systems 56:249-257.
3. Dornbush ME, TM Isenhart and JW Raich. 2002. Assessing the importance of fine roots: and improved alternative to litterbags. Ecology 83(11):2985-2990.
4. Simpkins WW, TR Wineland, RJ Andress, DA Johnston, GC Caron, TM Isenhart and RC Schultz. 2002. Hydrogeological constraints on riparian buffers for reduction of diffuse pollution: examples from the Bear Creek Watershed in Iowa, USA. Water Science and Technology 45(9):61-68.
5. Wineland TR. 2002. Assessing the role of geology for nitrate fate and transport in groundwater beneath riparian buffers. M.S. Thesis. Iowa State University. 122 pp.
6. Dornbush ME. 2001. Fine root decay: a comparison among three species. M.S. Thesis. Iowa State University. 109 pp.
7. Marquez CO. 2001. Soil aggregate dynamics and aggregate-associated carbon under different vegetation types in riparian soils. Ph.D. Dissertation. Iowa State University. 214 pp.
8. Tufekcioglu A. 2000. Biomass, carbon, nitrogen, and soil respiration within riparian buffers and adjacent crop fields. Ph.D. Dissertation. Iowa State University. 104 pp.
9. Webber D. 2000. Comparing estimated surface flowpaths and sub-basins derived from digital elevation models of Bear Creek watershed in central Iowa. M.S. Thesis. Iowa State University. 82 pp.
10. Andress RJ. 1999. Fate and transport of nitrate in groundwater within a riparian buffer in the Bear Creek Watershed. M.S. Thesis. Iowa State University. 93 pp.
11. Hameed S. 1999. Spatio-temporal modeling in an agricultural watershed. Ph.D. Dissertation. Iowa State University. 237 pp.
12. Lee KH. 1999. Effectiveness of a multi-species riparian buffer system for sediment and nutrient removal. Ph.D. Dissertation. Iowa State University. 139 pp.
13. Pickle JE. 1999. Microbial biomass and nitrate immobilization in a multi-species riparian buffer. M.S. Thesis. Iowa State University. 86 pp.

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 4 of 13

INVESTIGATOR: Russell, A. E.

PERFORMING INSTITUTION:
BOTANY
IOWA STATE UNIVERSITY
AMES, IOWA 50011

NITROGEN FERTILIZATION EFFECTS ON SOIL ORGANIC MATTER DYNAMICS IN A CORN-BELT AGROECOSYSTEM

OBJECTIVES: The purpose of this research is to quantify the effects of increasing rate of N fertilization on C and N inputs from crops to the soil and losses from soil organic matter (SOM). The main objective is to couple field measurements of plant detrital inputs to the soil with measures of CO2 loss from the soil to the atmosphere and N mineralization rates under different N fertilization levels in a Corn-Belt agroecosystem. This researcher will determine in 4-yr-old experimental plots 1) the quantity of crop annual net primary productivity (ANPP), both above- and belowground components, 2) the quality of the crop components (as determined by tissue C, N and lignin content), 3) the decomposability of the crop residues, 4) soil respiration of total soil and soil minus roots, and 5) net N mineralization rates. The final objective is to use a process-based model, the CENTURY SOM model, to investigate in an experimental mode the various mechanisms by which N fertilization influences SOM dynamics.

APPROACH: The approach will be to make complementary use of an ongoing field experiment, laboratory analyses, and process-based simulation modeling. Experimental plots were established four years ago (Ames, IA) in an annual corn-soybean rotation under three levels of N fertilization: 60, 120 and 180 kg/ha. Root production in these systems will be estimated by measuring total root biomass to a depth of 60 cm at the end of the growing season (before harvest). Total aboveground crop production will be measured by harvesting all aboveground crop material from quadrats of known area at three times during the growing season. Crop tissue quality will be assayed by collecting, drying and grinding crop tissues (leaves and roots), and analyzing for C and N by dry combustion methods, using a Carlo Erba NA 1500 N/C/S Elemental Analyzer. Another subsample will be used for determination of lignin content using van Soest's Acid Detergent Fiber method. Decomposition rates of crop residues will be determined by direct measurements of weight losses of crop residue from litter bags placed on the soil surface. To test the combined effects of both substrate quality and site-treatment factors, leaves will be decomposed in their N treatment of origin. To examine the effect of substrate quality alone, leaves from all three N fertilization levels will be decomposed in a single N fertilization treatment, 134 kg/ha. CO2 efflux rates from soils will be measured using the soda-lime technique for capturing CO2 from the soil surface. Russell will couple the measurements of CO2 fluxes with measurements of net N mineralization and nitrification using the in situ buried bag technique. Analysis of variance tests (ANOVAs) will be used to test for differences in N treatment effect on above- and belowground production, tissue quality and decomposability, and N mineralization rates. ANOVA for a repeated measures design will be used to analyze for differences in CO2 fluxes. Modeling using CENTURY version 4.0 will complement data collection and analysis in an iterative process in the effort to understand the mechanistic bases by which N fertilization influences soil detrital processes under two crop types.

NON-TECHNICAL SUMMARY: Nitrogen fertilization is a management practice that can potentially influence carbon cycling on a large scale over the Corn-belt region. The objectives of this research are to investigate the effects of increasing rate of N fertilization on the quantity and quality of carbon inputs from crops to, and CO2 losses from, soil.

PROGRESS: 1999/12 TO 2001/04
Plots where research was conducted were established in 1995. The original design consisted of 3 levels of N fertilizer application, 67, 135 and 202 kg/ha in three replicates. Two other unreplicated N levels, 0 and 280 kg/ha were added in 2000. Variability was assessed by sampling multiple transects within replicates. The study site soils are classified as fine-loamy, mixed, mesic Typic Haplaqualls and the field was moldboard plowed. The crop system is corn in rotation with soybean; the field has been in corn in 1996, 1998 and 2000. Soil respiration (SR) and in situ net N mineralization, along with soil temperature and moisture, were measured on a monthly basis from Feb to Nov 2000. Bulk density was measured in July. Total corn biomass was measured in May, three wks after planting, and in Sept, at the black-layer stage. Biomass was determined for the following plant parts: leaves, stalks, fruit, grain, tassles, detritus, prop roots, and fine and coarse roots (live and dead) to a depth of 110 cm. Soil samples and plant components were analyzed for total C and N by dry combustion. End-of-season stalk nitrate increased with N fertilizer level, from 670-7882 ug/g, exceeding the optimal fertilization level (1000 ug/g) in all but the 0 and 67 kg/ha N treatment. Total aboveground biomass increased significantly with rate of fertilizer addition, from 797 to 1133 g/m2. In contrast, root biomass tended to decrease with increasing N fertilizer, from 49 to 35 g C/m2. Root biomass comprised <6% of total corn biomass under all N treatments. Soil organic C in the top 10 cm of soil ranged from a high in May of 3590 g/m2 to a low of 3200 g/m2 in Aug. There were no differences in SOC among the three replicated N treatments (67, 135 and 202 kg/ha). Soil respiration peaked during June-Sept, ranging from 3.5 to 6.1 g C/m2/day. Soil respiration was significantly positively correlated with fine root biomass (r2 = 0.94). These data suggest that SR was driven by root inputs and that older soil organic matter (SOM) decay contributed relatively little to soil respiration. Net N mineralization during the growing season ranged from 0.49 to 2.4 g N/m2/30 day, and tended to be lowest under the 0 and 280 kg/ha N treatment. This suggests that microbes may have been limited, perhaps by N in the 0-N treatments and by C in the 280-N treatment. More root inputs in the 0-N treatment may have provided more substrate leading to greater microbial activity, hence greater net N immobilization by microbes. Decomposition experiments had just been initiated when the grant was terminated, so no results are possible for that study. Similarly, the grant was terminated before the scheduled modeling work using the CENTURY SOM model was allowed to take place at Colorado State Univ. Nitrogen fertilization is generally believed to increase SOM stocks by increasing plant production and thus detrital inputs to soils. Results indicate that under higher-than-optimal N fertilization levels, corn allocates less biomass to roots, leading to low soil biological activity, indicating poor conditions for SOM accumulation. There was no evidence of carbon sequestration under any of the N fertilizer levels.

IMPACT: 1999/12 TO 2001/04
N fertilization is generally believed to increase SOM stocks by increasing plant production. However, these results indicate that under higher-than-optimal N fertilization levels, corn allocates less biomass to roots, leading to low soil respiration and net N mineralization rates, all of which indicate poor conditions for SOM accumulation, hence carbon sequestration in these soils under corn-soy rotation.

PUBLICATIONS: 1999/12 TO 2001/04
No publications reported this period

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 5 of 13

INVESTIGATOR: Herriges, J. A.; Zhao, J.

PERFORMING INSTITUTION:
ECONOMICS
IOWA STATE UNIVERSITY
AMES, IOWA 50011

INFORMATION EXCHANGE IN FARMERS' ADOPTION OF CONSERVATION TILLAGE PRACTICES

OBJECTIVES: This project has three primary objectives. Objective 1: Develop a theoretical model of conservation tillage that incorporates information exchange among farmers. While there is a sizeable amount of literature on technology adoption, this literature generally ignores the exchange of information among farmers. We will develop a theoretical model of conservation tillage adoption that incorporates information exchange and examine its impact on the path of adoption. Objective 2: Construct econometric models to test the predictions of the theoretical model and estimate the role of information exchange. We propose to employ a series of complementary econometric models to capture the diffusion of technology both spatially and temporally, and to test the implications of the theoretical framework developed under Objective 1. Objective 3: Examine the attributes of effective government policies that promote conservation tillage adoption. Given information exchange and neighbor effect, government policies that promote adoption should be 'community oriented' rather than 'individual farm oriented.' Based on the estimation results, we will examine efficient policies that would promote conservation tillage adoption.

APPROACH: In a recent paper, Jinhua Zhao developed a theoretical model of the impact that information exchange among farmers can have on technology adoption. In this project, we propose to test whether the information mechanism suggested by this model is important in actual technology adoption decisions and diffusion processes. We will undertake several complementary approaches to capture the diffusion of technology both spatially and temporally. Approach 1: A Reduced Form Model of Technology Adoption with Spatial/Temporal Correlation - The first step in our empirical analysis will be to generalize the existing literature on the adoption of conservation tillage by allowing for spatial and temporal correlation in a reduced form discrete choice model. Spatial correlation patterns can be incorporated into a technology adoption model by assuming that a spatial autoregressive process (SAR) generates the errors. SAR assumes that the stochastic nature of an individual's adoption decision depends not only on stochastic factors directly affecting him/her, but also on similar stochastic factors impacting his/her neighbors. There are several limitations to the standard SAR error specification. First, they typically assume that any error or shocks impact only the period in which they occur or with a single lag. Second, the spatial effects are assumed to arise only in terms of the intercept of the expected profit function. We will employ the mixed logit framework both to apply the SAR model to the adoption of conservation tillage practices and to extend it to allow for inter-temporal correlation and spatial correlation that is not limited to the intercept term. Approach 2: Simultaneous Equations Model - The reduced form specification in the previous section is limited in that it simply acknowledges correlation among the decision-makers over time, not the actual neighbor decision. An alternative specification is the mixed regressive, spatial autoregressive model. In this case, an individual's expected profits are assumed to be a function of both his/her farm and socio-economic characteristics and the expected profits of his/her neighbors. Unlike in the reduced form models described previously, it is the neighbors' expected profits (and not simply their error components) that directly influences a given farmer's decision. In this research project, we will apply the simultaneous equation systems approach to modeling the adoption of conservation tillage practices and attempt to generalize the structure to allow for inter-temporal, as well as spatial, correlation. Approach 3: Structural Estimation - Finally, we develop a structural estimation model that builds directly on Zhao's theoretical model but ignores the strategic aspect. In particular, rather than estimating explicitly the 'game' played by the farmers, we model the dynamic adoption decision of a representative farmer who takes as given the equilibrium diffusion of the technology. The structural model allows us to estimate a farmer's profit function, discount rate, and the transition matrix of the state variables (or information sets).

NON-TECHNICAL SUMMARY: Conservation tillage practices reduce soil erosion, energy consumption, fertilizer use and water pollution and can even sequester a substantial amount of carbon. Yet, conservation tillage has not been widely adopted in the U.S. The goal of this research is to address theoretically and empirically the under-adoption of conservation tillage and the appropriate government policies to promote adoption.

PROGRESS: 2002/01 TO 2002/12
This project focuses on the adoption of conservation tillage in agricultural production, with an emphasis on the exchange of information among farmers in the adoption decision, both spatially and intertemporally. The environmental benefits of adopting conservation tillage practices are well known. Compared with conventional tillage, low till and no till practices reduce soil erosion, energy consumption, fertilizer use and water pollution, and can even sequester a substantial amount of carbon. Yet, conservation tillage has been only slowly adopted in the United States. The goal of the proposed research is to address theoretically and empirically: (i) the under-adoption of conservation tillage, and (ii) the appropriate government policies to promote adoption. Our efforts this past year have focused on two areas. First, we continue to gather and analyze the requisite data for our empirical analysis. The primary data sources for this project are the Natural Resource Inventory (1982, 1987, 1992, and 1997) and the Area Studies (AS) data sets collected by the U.S. Department of Agriculture. As reported last year, we encountered considerable difficulty in acquiring these data and linking the two data sources. We also discovered a discrepancy between the types of tillage practice being used on a farm as reported by the NRI data versus the Area Studies data. After a series of discussions with USDA on the data, we were told that the Area Studies information on tillage practice was the more reliable and appropriate for our analysis. Unfortunately, we found that virtually all of the factors that were statistically significant in explaining conservation tillage adoption in Iowa using the NRI tillage variable were no longer statistically using the Area Studies tillage variable. We have had more success with the Nebraska Area studies data and are in the process of modeling tillage adoption in that state. Second, we have begun development of a discrete choice model that accounts not only for observed characteristics of a field and farm in their adoption of conservation tillage, but also controls for likely correlation in unobserved characteristics over space. This model employs the mixed logit framework of McFadden and Train (1999). The major difficulty in implementing this model is that we require specific location information, so as to measure the proximity of farms to one another. We are currently relying on crude spatial location information (e.g., county and crop reporting districts), but have requested exact longitude and latitude information for the points in our data. This should greatly enhance our ability to capture spatial correlations among farm level decisions.

IMPACT: 2002/01 TO 2002/12
First, the results of the project will assist government agencies in understanding the adoption process and designing efficient policies that promote adoption of conservation tillage practices. We expect that conservation tillage will continue to be important in contributing to the sustainability of U.S. agriculture, especially if Kyoto Protocol will allow agricultural soil as eligible carbon sinks. Second, both the theoretical and econometric models of the project can be easily applied to other regions of the U.S. and other adoption decisions, such as drip irrigation, new seed variety, etc. Finally, the results will be useful for other researchers, in both economics and sociology, who are studying farmer interaction and community development.

PUBLICATIONS: 2002/01 TO 2002/12
No publications reported this period

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 6 of 13

INVESTIGATOR: Liebman, M.; Buhler, D. D.

PERFORMING INSTITUTION:
AGRONOMY
IOWA STATE UNIVERSITY
AMES, IOWA 50011

IMPACTS OF COMPOSTED SWINE MANURE ON WEED SEED SURVIVAL, SEEDLING EMERGENCE, GROWTH, AND COMPETITIVE ABILITY

OBJECTIVES: This project will (1) identify how application of an organic amendment to soil (composted swine manure) alters weed population dynamics and weed-crop interactions; and (2) determine the influence of weed seed size on weed responses to soil conditions. Information generated by this project will provide insight into the possibility of developing "weed suppressive soils" through the use of manure management practices that foster efficient nutrient recycling, diminish odor problems, and reduce opportunities for water contamination.

APPROACH: Effects of composted swine manure on weed seeds and seedlings will be studied by creating artificial weed seed banks in field plots amended or not amended with composted manure. Seeds of three common Midwestern weed species (Abutilon theophrasti, Amaranthus rudis, and Setaria faberi) will be added to soil contained in PVC rings and then recovered at various times after placement. Seed survival and germination will be determined. Soil and weed seeds collected from field plots will processed in a microbiology laboratory and analyzed for the presence and impacts of seed-inhabiting bacteria and fungi. To determine whether composted swine manure alters weed-crop competition, seeds of A. theophrasti, A. rudis, and S. faberi will be sown with corn in field plots that have or have not received composted manure. Weed seedlings will be thinned shortly after emergence to fixed densities. A weed-free, corn-only control treatment will also be established. Weed and corn growth, nutrient uptake, final biomass, and seed production will be measured. Laboratory bioassays will be used to determine the effects of composted swine manure on 10 weed species (Ipomoea hederacea, Abutilon theophrasti, Brassica kaber, Solanum ptycanthum, Amaranthus rudis, Sorghum bicolor, Eriochloa villosa, Setaria faberi, Echinochloa crus-galli, and Panicum dichotomiflorum) that differ in seed weight. Pre-germinated seeds will be placed on soil drawn from field plots amended or not amended with composted manure, and seedling length and disease incidence will be measured after incubation. A glasshouse experiment will be conducted to determine the effects of composted swine manure on seedling growth of the same 10 weed species over a five-week period.

NON-TECHNICAL SUMMARY: Midwestern agriculture is currently characterized by heavy reliance on herbicides and concentrated livestock production, both of which can threaten environmental quality. This project examines the impacts of composted swine manure on weeds in corn and soybean fields. Because of their seed bank and fixed root habits, annual weeds are highly responsive to soil conditions. Thus application of composted swine manure to soil is likely to alter weed dynamics. The compost used in this study will be produced in hoop structures that are increasingly popular because of low capital investment costs, lower risks of water contamination, and reduced odor emissions. Two field experiments, a laboratory bioassay experiment, and a glasshouse experiment will be conducted to examine compost effects on seed and seedling survival, establishment, growth, and competitive ability of ten weed species commonly infesting Midwestern U.S. corn and soybean fields. We will use species spanning a range of seed weights and seedling emergence times because we believe those are key factors affecting the structure and function of weed communities. The project will test whether the use of an organic matter amendment to soil, such as composted swine manure, can foster the development of weed-suppressive soil conditions, through changes in physical, chemical, and biological characteristics.

PROGRESS: 2002/01 TO 2002/12
Field plot, growth chamber, and glasshouse experiments were conducted from 1999 to 2002 to determine soil, weed, and crop responses to compost made from swine manure and corn stalks. Application of compost (at 8 Mg C per hectare) to Clarion (Typic Hapludolls) and Nicollett (Aquic Hapludolls) loam soils in Boone, IA, increased soil moisture content, and soil nitrate-N, P, K, and organic matter levels. In plots in which different weed species grew with corn, compost increased common waterhemp (Amaranthus rudis) height and biomass in each of three years, and increased velvetleaf (Abutilon theophrasti) height and biomass in one year. Compost had no effect on the height or biomass of giant foxtail (Setaria faberi) grown with corn. Seed production by common waterhemp, velvetleaf, and giant foxtail was proportional to biomass production. Compost consistently increased corn height, leaf K concentration, and stalk nitrate-N concentration, but typically did not increase corn grain yield under weed-infested conditions. In some years, yield of weed-infested corn was lower with compost than without compost, indicating that compost could increase weed competition against corn. Results from plots in which soybean grew with common waterhemp sown at different times indicated that the impact of compost depended on the timing of weed emergence. When common waterhemp emerged before soybean reached the second-node stage, compost increased the weed's height, biomass, competitive ability, and seed production. In contrast, when common waterhemp emerged after the soybean second-node stage, weed growth and competitive impact on soybean were negligible, regardless of compost application. To evaluate the effect of composted swine manure on crop and weed germination in a growth chamber, seeds of corn, soybean, wheat, common waterhemp, velvetleaf, and giant foxtail were placed in moistened germination paper with 100 g of soil amended with compost at rates equivalent to 0, 3.5, 7.0, or 10.7 g of C per kg of soil. Compost inhibited germination in a manner that was concentration dependent and inversely proportional to seed size: larger seeded species (the three crops) were less inhibited than the smaller seeded species (the three weeds). To evaluate the impact of compost on emergence and growth of the three crops and three weeds in a glasshouse, seeds were sown in pots containing compost at the concentrations described above, and seedling density and biomass were determined periodically. Compost did not affect crop emergence, but reduced weed emergence. Without compost, relative growth rates of the crop and weed species were similar, whereas with compost, relative growth rates of the weeds were higher than those of the crops. Overall, results indicate that compost can reduce weed emergence, but can also increase weed growth, seed production, and competitive ability against crops. Seed size appears to be an important variable determining differential responses to compost among crop and weed species.

IMPACT: 2002/01 TO 2002/12
Deep-bedded hoop structures are becoming increasingly popular for swine production in Iowa and other Midwestern states because they can reduce capital investment costs, odor emissions, and water contamination risks associated with conventional swine production systems. Hoop buildings used for swine production are typically bedded with corn stalks or cereal straw, and mixtures of the bedding and manure can be composted readily and spread on farm fields as soil amendments. Results of this project indicate that composted swine manure has beneficial effects on soil characteristics and crop nutrition, and that it can reduce weed emergence. However, once weeds emerge successfully, this type of compost can promote weed growth, competitive ability, and seed production. Consequently, effective weed management strategies should be in place when composted swine manure is used as a soil amendment.

PUBLICATIONS: 2002/01 TO 2002/12
1. Liebman M, T Richard, DN Sundberg, DD Buhler and FD Menalled. 2002. Impacts of composted swine manure on maize and three annual weed species. Proceedings of the 5th Workshop of the European Weed Research Society Working Group on Physical and Cultural Weed Control (11-13 March 2002, Pisa, Italy), p. 173. Institut de Malherbologie, Ste. Anne de Bellevue, Quebec, Canada.
2. Menalled FD, M Liebman and DD Buhler. 2002. Differential response of crop and weed seeds to composted swine manure applications. Ecological Society of America 2002 Meeting Abstracts: http://199.245.200.45/pweb/?SOCIETY=esa&YEAR=2002&ID=5483.
3. Menalled FD, M Liebman and DD Buhler. 2002. Impact of composted swine manure on crop and weed establishment and growth. Proceedings of the 5th Workshop of the European Weed Research Society Working Group on Physical and Cultural Weed Control (11-13 March 2002, Pisa, Italy), p. 183. Institut de Malherbologie, Ste. Anne de Bellevue, Quebec, Canada.

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 7 of 13

INVESTIGATOR: Birrell, S. J.

PERFORMING INSTITUTION:
AGRI & BIOSYSTEMS ENGINEERING
IOWA STATE UNIVERSITY
AMES, IOWA 50011

REAL-TIME SOIL NITRATE ANALYSIS SYSTEM FOR PRECISION NITROGEN APPLICATION

OBJECTIVES: The overall goal this research is to develop a real-time soil nutrient analysis system, based on ion-selective field-effect transistors (ISFETs). The work concentrates on the development of nitrogen sensors, due to the economic importance of these fertilizers and the potential environmental effects of excess fertilizer applications. However, the proposed analysis system could be adapted to sense potassium, phosphate, soil pH, soil micro-nutrients and used for the simultaneous analysis of multiple nutrients. The objectives are: 1) Develop a real-time soil sampler and extraction system for soil nutrient analysis. The system must minimize reagent consumption to allow for continuous sampling, and be capable of within-field collection of soil samples at multiple depths. 2) Develop a miniaturized flow injection analysis (FIA) system using ion-selective field effect transistors (ISFETs) to sense nutrient levels in soil extracts. The overall FIA/ISFET system must survive field operation and be capable of self-calibration. 3) Integrate and evaluate the soil sampling and extraction system and the FIA/ISFET system to develop a prototype real-time soil analysis system for testing in a soil bin and in fields.

APPROACH: The time lag between obtaining a soil sample and the result must be minimized (< 10 s). The sampling period should be considerably less, to allow averaging of individual readings to account for extremely short range variability. The consumption of extracting solutions must be minimized to reduce costs and eliminate potential environmental or crop damage from extractants deposited in the soil. Objective 1: The extraction of a consistent percentage of the nutrient in the soil sample must be achieved for a reliable soil nutrient analysis system. The effect of soil type and texture, sample volume, sample preparation, and rate of extraction solution injection on extraction performance is being tested. The soil sampling system will use multiple sensors to analyze the nutrient concentration at specific depths to obtain multiple readings from which the average nutrient levels of the profile can be calculated. Different configurations of the soil sampling mechanism will be tested in a soil bin. The precision of the system will be evaluated in terms of the volume of soil collected, soil to extract ratio, and nitrate concentration of the filtrate obtained. The interaction between travel speed, soil type and texture, soil compaction, soil moisture, mixing and filtration time will be evaluated. Objective 2: Previous work on the ISFET/FIA analysis system used conventional peristaltic pumps and injection valves. A miniaturized system using micro-valves, which are now commercially available, will provide a significant improvement in performance over the previous system. The integrated FIA/ISFET analysis system will include a injection loop to control the sample volume to eliminates the effect of pressure pulsations and inconsistent flow rates of the sample stream. The integrated system will include the necessary configuration for automatic calibration, and will be designed to switch from analyzing unknown samples to standard solutions, which will be used to automatically determine the calibration constants. The deviations between the predicted nitrate concentration for individual ISFETs and the mean predicted concentration for all the sensors provide the opportunity for error checking. During analysis and self-calibration, a large deviation for any single sensor may indicate failure of that particular sensor or an outlier result. Objective 3: The complete system must operate without significant operator intervention, even during a calibration sequence. Control software will be developed to automatically run the system and record the results. The software will track and flag potentially erroneous results, warn the operator when failure of an individual ISFET is detected or if the calibration is not within specifications. The integrated system will be tested in a soil bin and in fields having a range of soil types and conditions. The system will be tested in a soil bin at various soil moisture contents and soil nitrate levels. The results of the tests will be analyzed to determine the precision and accuracy of the sensor and the recovery efficiency of the system.

NON-TECHNICAL SUMMARY: Over the last half century, agricultural production has increased dramatically due to the introduction of new hybrids, improved agronomic knowledge, mechanization of agriculture, pesticides and chemical fertilizers. Unfortunately, the widespread use of nitrate fertilizers has increased the potential for contamination of water resources leading to human health risks. The goal of this project is to develop a real-time soil nitrate sensor. A real-time nitrate sensor used in conjunction with established soil nitrate test recommendations, such as the pre-sidedress nitrate test (PSNT) could have a significant impact on fertilizer application. There is potential for reduction in fertilizer inputs with negligible reductions in yield, while reducing the potential for environmental degradation due to excess nitrate in the environment.

PROGRESS: 2002/01 TO 2002/12
A critical component to successful completion of this project is the ability to automatically sample a known mass (or volume) of soil for extraction and analysis. During 2001, laboratory tests were conducted to evaluate the effect of soil type (3 types), soil density (2 compaction) and soil moisture (3 Moisture Contents) on obtaining a known sample mass. The tests proved that the laboratory system was relatively successful in obtaining a constant mass of soil with coefficient of variation of less than 20% for all conditions. In the past year a prototype field sampling system has been designed, and limited field tests conducted in different conditions. The results are encouraging under drier soil conditions. However, as expected high moisture conditions cause operational problems. Further tests will be conducted in the upcoming spring. The soil mixing, extraction system and filtration system has been designed and the relevant software control is being developed. The extraction and filtration system will be tested in a soil bin during the winter, and the initial field testing is planned for the upcoming spring.

IMPACT: 2002/01 TO 2002/12
Precision is a management strategy that seeks to address within-field variability and to optimize inputs such as pesticides and fertilizers on a point-by-point basis within a field and not according to the field average. The full benefit of precision ag. will only be realized if the spatial variation across the field is accurately determined. This often requires data collection on a finer spatial resolution than is feasible with manual and/or laboratory methods, due to prohibitive cost. Sensors will allow the collection of data on a much finer spatial resolution, to more accurately characterize within-field variability. Sensor technology currently lags behind the other enabling technologies necessary for precision agriculture. The successful development of real time soil nutrient sensors would have a major impact on reducing the cost of implementing precision agriculture and significantly affect the adoption rate of the technology in commercial agriculture.

PUBLICATIONS: 2002/01 TO 2002/12
No publications reported this period

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 8 of 13

INVESTIGATOR: Horton, R.; Jaynes, D. J.

PERFORMING INSTITUTION:
AGRONOMY
IOWA STATE UNIVERSITY
AMES, IOWA 50011

WATER FLOW AND PREFERENTIAL SOLUTE TRANSPORT PROPERTIES OF FIELD SOILS

OBJECTIVES: Develop a new method that can rapidly and simultaneously measure both hydraulic properties: saturated hydraulic conductivity and the macroscopic capillary length, and chemical transport properties: especially mobile water content, mass transfer coefficient, and dispersion coefficient in fields. Use the new method to measure the spatial distributions of hydraulic and solute transport properties of agricultural fields. Evaluate the new method using both observed data and numerical simulations.

APPROACH: The new method combines a point source method with time domain reflectometry (TDR). Irrigation drippers and TDR probes installed beneath the irrigation drippers will be used to simultaneously determine both hydraulic and chemical transport properties at multiple field locations. The method is suited to characterize spatial distributions of surface hydraulic and chemical transport properties of field soils. Once soil hydraulic and chemical transport properties are measured, the properties will be used in existing solute transport models to predict water and chemical leaching through soil profiles. The predictions of water and solute transport through soil profiles will be compared with observed tracer leaching through soil profiles. Comparison of predicted and measured tracer distributions will determine how well the surface hydraulic and transport properties obtained from the new method can be used to predict chemical leaching.

NON-TECHNICAL SUMMARY: There are no practical field methods to measure both soil water and chemical transport properties. Therefore, we are not able to screen for how quickly chemicals leach through soil. The purpose of this study is to develop and evaluate a new method that can rapidly measure both water and chemical transport properties of field soil.

PROGRESS: 2002/01 TO 2002/12
We performed a field experiment to evaluate the effect of surface solute transport properties on subsurface leaching. The field plot was 12 m by 12 m with a tile drain crossing its center at a depth of 1.2 m. Water with relatively low electrical conductivity (EC) was applied to the plot by a portable sprinkler system until a steady state water condition was attained. After reaching the steady state condition, about 14 cm of water with calcium chloride of 0.1 M concentration and 8 cm of water with low EC were applied consecutively by the same sprinkler system. A time domain reflectometry (TDR) set up was utilized to determine the near-surface chemical transport properties by measuring the bulk electrical conductivity in the top 2 cm of soil. In order to ascertain the distribution of surface solute transport properties, a total of 45 TDR probes were installed at different locations covering the various field operational management practices such as crop row, traffic and non-traffic furrows. The electrical conductivity (EC) of the tile flow was measured continuously. The breakthrough curves measured by the TDR probes will be used to estimate the surface solute transport properties that will subsequently be used to predict the tile flow EC.

IMPACT: 2002/01 TO 2002/12
New data have been obtained for describing soil chemical transport properties. These new measurements will enable better characterization of soil properties. They also enable quantitative evaluation of soil management system effects on chemical leaching.

PUBLICATIONS: 2002/01 TO 2002/12
1. Al-Jabri SA, R Horton and DB Jaynes. 2002. A point-source method for rapid simultaneous estimation of soil hydraulic and chemical transport properties. Soil Sci. Soc. Am. J. 66:12-18.
2. Al-Jabri SA, R Horton, DB Jaynes and A Gaur. 2002. Field determination of soil hydraulic and chemical transport properties. Soil Sci. 167:353-368.
3. Lee J, R Horton and DB Jaynes. 2002. The feasibility of shallow time domain reflectometry probes to describe solute transport through undisturbed soil cores. Soil Sci. Soc. Am. J. 66:53-57.

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 9 of 13

INVESTIGATOR: Owen, M. D.; Liebman, M. Z.; Jurik, T. W.

PERFORMING INSTITUTION:
AGRONOMY
IOWA STATE UNIVERSITY
AMES, IOWA 50011

LIGHT, SEED WEIGHT, AND EMERGENCE TIMING EFFECTS ON WEED RESPONSE TO NITROGEN

OBJECTIVES: We hypothesize that light interception, weed seed weight, and weed emergence date partially explain the variable weed response to soil fertility in previous studies. We propose the investigation of these on three factors, with specific objectives to: 1) determine the effect of nitrogen application date and crop density on the competitiveness of selected weeds in field corn; 2) determine the effect of light availability and seed weight on interspecific weed response to nitrogen availability; and 3) determine the effect of weed seedling emergence time relative to corn seedling emergence on weed response to nitrogen availability.

APPROACH: Understanding how light availability, seed weight and emergence time influence weed response to soil fertility will allow producers to optimize fertilization and planting configuration strategies to reduce weed competitiveness for nutrients and light. The objectives will be met by a three-year field study of the effect of two nitrogen application dates and two corn population densities on the competitiveness of three weed species. The experiment will describe the treatment effects on early and mid-season vegetative growth, as well as the seed production of both crop and weed species. Furthermore, we will grow six weeds in sand culture. The weeds will be grown in environments with high and low nitrogen availability and light availability. Last, corn and Setaria faberi will be grown in competition and varying the weed emergence timing relative to the corn. These plants will be grown in sand culture environments with high or low nitrogen. Functional growth analysis will determine the rapidity and severity of each species' response to the treatment combinations.

NON-TECHNICAL SUMMARY: The role of soil fertility management in integrated weed management systems is not well understood. The research will investigate the response of six weeds to varying nitrogen and light availability to determine the impact of different corn populations and nitrogen application timing on weed growth.

PROGRESS: 2002/01 TO 2002/12
We studied the responses of eight crop and weed species, in light and shade, to daily fertilization with either 7.5 or 0.2 mM NH4NO3 in a growth chamber. Dry weights and leaf areas of all species at 18 days after emergence (DAE) were greater with high N than with low N. These favorable responses to high N were also greater in light than in shade in all species. Dry weights with high N were up to 100% greater in shade, and up to 700% greater in light, than with low N. These responses suggested that shade reduced the benefit of N to the plants. Dry weight and leaf area responses to N were positively correlated with their respective relative growth rates under high N conditions. The regression slopes of relative growth rates with high N on relative growth rates with low N were less than unity. These slopes demonstrated an ecological trade-off in which the fastest-growing species with high N suffered greater proportional decreases in growth with low N. Relative growth rates were also negatively correlated with mean seed weights among species. We compared the effect of pre-emergence (PRE N) and post-emergence (POST N) ammonium nitrate applications on the competitive abilities of giant foxtail, velvetleaf, and waterhemp in 'Pioneer 33V08' corn. The N timing effects were studied in corn densities of 5.4 and 7.9 plants per meter in order to understand the influence of other resources, including shading, on N response in weeds. By late June of each year, PRE N had increased shoot dry weight by over 50% for corn, 100% for velvetleaf, and at over 20% for giant foxtail, compared to POST N. Common waterhemp was unaffected by N timing. Corn density increased corn shoot dry weight by 21 to 32%, but had little effect on weeds. Corn density did not generally affect weed shoot dry weights when evaluated in late June. Corn grain and weed seed yields were significantly affected by interactions between N timing and weed species. Corn yield was decreased 13 to 20% by velvetleaf for PRE N compared to POST N. Conversely, corn yield was decreased 12 to 15% by POST N compared to PRE N. Similarly, PRE N increased velvetleaf seed weight by 13 to 195% compared with PRE N, but decreased giant foxtail seed weight by 55%. Velvetleaf and giant foxtail seed weights with the greater corn density were decreased by 23 to 56% and 30 to 62%, respectively. Neither N timing or corn density affected corn yield loss or weed seed production in treatments that included common waterhemp.

IMPACT: 2002/01 TO 2002/12
Relative growth rate and seed size may predict weed community responses to ecological weed management strategies that rely on shade and nitrogen manipulation. Economically important weeds are affected by nitrogen application timing, which may be incorporated into integrated weed management (IWM) strategies that reduce environmental impacts of crop production and crop yield losses. Nitrogen application timing also has an important effect on corn growth that may be used to increase the competitiveness of that crop thus improving the efficiency of herbicide use and further reduce crop yield losses.

PUBLICATIONS: 2002/01 TO 2002/12
Harbur MM and MDK Owen. 2002. Effect of Nitrogen application timing and corn density on the competitiveness of three Iowa weed species. Proc. North Cent. Weed Sci. Soc. 57:54

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 10 of 13

INVESTIGATOR: Isenhart, T. M.; Schultz, R. C.; Simpkins, W. W.

PERFORMING INSTITUTION:
NATURAL RESOURCE ECOLOGY & MANAGEMENT
IOWA STATE UNIVERSITY
AMES, IOWA 50011

ASSESSMENT AND PREDICTION OF THE FATE OF NITRATE IN RE-ESTABLISHED RIPARIAN BUFFERS

OBJECTIVES: The overall goal of this work is to determine the effectiveness of riparian buffers re-established on previously cropped land in regulating nitrate flux to surface and groundwaters. The central hypothesis is that the riparian buffers will reduce nitrate movement to surface and groundwaters, but that such benefit is dependent upon the hydrogeologic setting (i.e. surface topography, soils, stratigraphy, and hydraulic and geochemical properties of geological units) and the composition of the vegetation and age of the buffer ecosystems. Specific research objectives are to 1) Quantify nitrate flux within a chronosequence of re-established riparian buffers, 2) Evaluate geomorphic and hydrogeologic controls of nitrate transport within riparian zones, and 3) Evaluate the mechanistic linkages between the composition and age of riparian vegetation, soil quality, and nitrate loss processes.

APPROACH: The long-range goal of the Bear Creek Watershed Project is to develop locally-acceptable watershed management systems that increase the sustainability of agriculture in the Midwestern United States with respect to surface and ground water quality, while improving the integrity of the aquatic and terrestrial ecosystems. These systems include re-establishing a suite of perennial plant-based conservation buffers, constructing or restoring strategically placed wetlands, and implementing management practices that are meant to complement in-field best management practices. As part of the project, we have already established over 11 km of riparian buffer along an agricultural stream in North-Central Iowa, providing a chronosequence ranging from 0 to 11 years since establishment. We have also defined and collected data on comparison sites of riparian zones under crops, cool-season grasses, and forest. Many sites are already implemented with the networks of groundwater and vadose zone monitoring equipment crucial to assessing shallow groundwater chemistry along groundwater flow paths. Objectives of this research will be achieved through intensive sampling and experimental approaches focused on these existing re-established riparian buffers and comparison sites. We will take advantage of the intensity of monitoring at these locations to define the mechanisms of nitrate attenuation and to assess the spatial and temporal heterogeneity of nitrate removal within established riparian buffers. Alluvium depth, texture, and extent beneath re-established riparian buffers and other vegetation will be assessed using geophysical techniques and core sampling. An analytic element model will be used to simulate the groundwater flow system at the watershed scale. The quantity and quality of carbon by depth and other soil quality parameters will be assessed beneath the re-established riparian buffers of different ages and compared with riparian zones under other vegetation. These soil quality parameters will be mechanistically linked to the ability of riparian soils to denitrify or immobilize nitrate. This information will be used to improve the ecological performance of buffers and help realize the stewardship goals of the National Conservation Buffer Initiative.

NON-TECHNICAL SUMMARY: Surface waters within intensively agricultural watersheds of the Midwestern USA contain some of the highest nitrate concentrations in the nation. Such nitrate loads have potentially widespread impacts on both public health and ecosystem function. This research examines the effectiveness of re-established riparian buffers in regulating nitrate flux to surface and groundwaters.

PROGRESS: 2002/01 TO 2002/12
The central hypothesis of this research is that riparian buffers will reduce nitrate movement to surface and groundwaters, but that such benefit is dependent upon the hydrogeologic setting (i.e. surface topography, soils, stratigraphy, and hydraulic and geochemical properties of geological units) and the composition of the vegetation and age of buffer ecosystems. The research focus area consists of nearly 11 km of re-established riparian buffer ranging in age from 1 to 12 years since establishment within the Bear Creek Watershed in Central Iowa. Ongoing monitoring includes monthly sampling of water-table monitoring wells for nitrate-nitrogen, dissolved oxygen, temperature, specific conductance, and chloride and quarterly sampling for dissolved organic carbon, pH, and alkalinity. Nitrate concentrations in groundwater show well-defined decreases at two of the three intensively monitored sites. In order to assess the effectiveness of riparian buffers in regulating nutrient flux at the watershed scale, a regional, two-dimensional analytic element model (GFLOW 2000), which simulates groundwater and surface water conjunctively, was developed to represent the region. The constructed model represents approximately 2200 km2 of central Iowa (with a near-field of about 600 km2) and is centered on Bear Creek. Unconfined conditions were assumed for the model simulations. Model input parameters include hydraulic conductivity (K=1.7 x 10-5 m/s), aquifer thickness (137 m), areal recharge (81 mm/yr), porosity (0.2), pumping from municipal wells, and discharge from drainage tiles into Bear Creek. Inhomogeneities along the creeks and rivers, which consist of alluvial and outwash deposits, were assigned K values of 3.5 x 10-4 m/s. Model calibration utilized up to six years of hydraulic head data from 41 piezometers along Bear Creek and stream discharge data from 3 gaging stations. Preliminary simulations show a deviation of 5 to 11 ft between actual and modeled heads (mean squared difference=8.5), which suggests that K and recharge need to be further optimized in the model to more closely match observed heads. Once an optimized solution is obtained, the model results can assist in calculating potential nitrate flux and removal through buffered areas of the watershed. Ongoing process-based studies are investigating denitrification potential and dissolved organic carbon (DOC) at five depths in warm season and cool season grass buffers. Denitrification potential was found to be much greater in surface soils than in subsurface soils. Cool season grass surface soils had higher denitrification potential than warm season grass. DOC is generally higher in surface soils and in cool season grass plots. DOC is not a strong predictor of denitrification potential in these sites. Carbon additions increase DEA more than nitrate additions compared to the control group. This suggests that carbon is limiting to denitrification in these soils. Two graduate students are supported on this project.

IMPACT: 2002/01 TO 2002/12
This research demonstrates that riparian buffers re-established on previously cropped or pastured land have tremendous potential to remediate nonpoint source pollution in agricultural ecosystems. The research also provides insight into the geomorphic, hydrologic, and biologic factors controlling the effectiveness of streamside buffers. This information is being used to improve the performance of buffers in the field and help realize the promise of conservation buffer technology.

PUBLICATIONS: 2002/01 TO 2002/12
No publications reported this period

PROJECT CONTACT:

Name: Good, C.
Phone: 515-294-4544
Fax: 515-294-2909
Email: cgood@iastate.edu

Item No. 11 of 13

INVESTIGATOR: Jaynes, D. B.; Saleh, A.; Arnold, J. G.

PERFORMING INSTITUTION:
USDA/ARS
NATIONAL SOIL TILTH LABORATORY
2150 PAMMEL DR
AMES, IOWA 50014

MODIFICATION AND EVALUATION OF SWAT MODEL FOR ARTIFICIALLY DRAINED WATERSHED

OBJECTIVES: There are two main objectives to this proposed work. The first is to modify the computer code for the watershed-scale SWAT model and develop the model protocols for simulating the unique effects that distributed subsurface tile drainage systems have on hydrology and nutrient fate and transport within surface streams. The second objective is to evaluate the accuracy and robustness of the modified model against the extensive and lengthy data set collected for the tile-drained Walnut Creek watershed in central Iowa.

APPROACH: In much of the Midwest, subsurface drains were installed piecemeal over an extended period of time. Tile maps are either nonexistent, inaccurate, or not up-to-date. Many of these tile drainage systems are distributed throughout the watershed, connecting areas of poor drainage to surface streams or ditches. Thus, precise knowledge of number and location of tile drains is not possible and instead an effective drainage network must be used to simulate the impact of artificial subsurface drainage on hydrology and nutrient contamination of surface waters. We will generate a subbasin map for the Walnut Creek watershed from the a digital elevation map using the r.watershed command in the GRASS (Geographic Resources Analysis Support System) GIS program. The Walnut Creek watershed will be divided into 7 subbasins based on the elevation map and the major sampling sites within this watershed. A virtual basin approach will be used to divide the Walnut Creek watershed into multiple units (virtual subbasins) for SWAT simulation. Each virtual subbasin will be assumed to have homogeneous land use, management, tile drain system, and soil characteristics. This approach provides the opportunity to better describe the different conditions including tile-drain systems, management practices, and soil characteristics within the Walnut Creek watershed. Once simulation of each virtual subbasin is completed, the SWAT program, revised to explicitly account for tile drains, will route the computed water flow, sediment, and nutrient loading from each virtual subbasin to the outlet of the sub-watershed. Each of the 7 subwatersheds in Walnut Creek watershed will be divided into 10-30 virtual subbasins. This will divide the Walnut Creek watershed into a total 70 to 210 virtual basins. The area of each virtual basin will range from 24 to 73 ha. Predicted flow, sediment, and nutrient loading will be verified against the measured values at each sub-watershed. The first phase of calibration and testing of the modified SWAT model will use Walnut Creek stream and county drain data from 1992. Evaluation of model performance will include comparison at daily and monthly time scales of water discharge and nitrate concentration and loads. Multiple comparison criteria including maximum error, root mean square error, coefficient of determination, modeling efficiency and coefficient of residual mass will be used to evaluate SWAT model performance. After initial calibration and testing of the model with 1992 data, the model will be used to predict watershed response for the years 1991 and 1993-1998 to test the robustness of the model - calibrated parameter set to accurately predict watershed for a range of weather patterns. If satisfactory, the model will be used to test the impact of various N management practices on water quality within Walnut Creek. Initially, the model will be used to investigate the effect of switching from a fall to spring N-fertilizer application scenario and predictions compared to the results of a watershed project currently underway within Walnut Creek using this strategy.

NON-TECHNICAL SUMMARY: States will be required to identify surface waters impaired by excess nutrients and develop total maximum daily loads (TMDLs) for nutrients in these impaired watersheds. Proper TMDL calculations will require accounting for nutrients derived from all sources, including agriculture, and may lead to mandated reductions in fertilizer use. Because stream databases are poorly developed for most watersheds, computer models will be relied upon for the determination of both TMDLs and the identification of best management practices for pollutant reduction efforts. To be affective, these models must accurately predict water flow and nutrient loads within complex watersheds, which will require the accurate description of all major hydrologic processes operating within a watershed. Currently, there are no watershed-scale models that include the effect of subsurface drainage tiles. This is especially important for much of the Midwest corn belt where 50% of the crop land in some states is drained by tiles. In this research, we will modify the extensively used computer model SWAT, to specifically include the efects that tiles have on surface hydrology and nutrient fate and transport. Once modified, we will evaluate the model for its accuracy and reliability in predicting stream flows and nitrate concentrations. This testing will use the multi-year data set being collected on Walnut Creek in central Iowa - a stream contamination by nitrate due to agricultural activity in a watershed dominated by tile drains. This data set includes stream flow and nitrate concentration measurements covering

PROGRESS: 2001/10 TO 2002/09
A Specific Cooperative Agreement has been executed between ARS and Tarleton State University to transfer grant funds to Dr. Saleh as specified by the proposal. Data from Walnut Creek has been verified by Dr. Jaynes and collected into data sets that can be directly imported into SWAT. This data has been shared with all co-investigators on the project. A post-doctoral investigator has been hired at Tarleton State University to provide the majority of the programming and model simulation effort. Initial modification and testing of the SWAT model has been completed. The model is faithfully matching the monthly hydrographs measured within Walnut Creek over the 1992 - 2000 seasons. The nitrogen component is currently being modified and evaluated against measured data. Early results showed that the model underestimated nitrate concentrations during the summer and fall months. Model simulations are being overseen by Dr. Saleh and model modifications are being made by Dr Arnold.

IMPACT: 2001/10 TO 2002/09
This research will develop and test a computer model that will allow States and others to quantify the impact of specific farming practices on water quantity and quality at the watrshed scale. The model will be specifically modified to simulate watersheds in the Midwest corn belt where extensive artificial subsurface or "tile" drainage networks have been installed and serve as the primary source of nitrate contamination of surface waters.

PUBLICATIONS: 2001/10 TO 2002/09
Arnold, F.G., R. Srinivasan, M. DiLuzio, S.L. Neitsch, K.W. King. 2002. SWAT2000 - Capabilities and improvements in watershed modeling. Proc. ASAE Annual Meeting. Paper No. 02-022137, Chicago, Il. July 2002.

PROJECT CONTACT:

Name: JAYNES, D. B.
Phone: 515-294-8243
Fax: 515-294-8125
Email: jaynes@nsti.gov

Item No. 12 of 13

INVESTIGATOR: Cheng, C. L.

PERFORMING INSTITUTION:
BIOLOGICAL SCIENCES
UNIVERSITY OF IOWA
IOWA CITY, IOWA 52242

GENETIC ENGINEERING OF SELF-DESTRUCTIVE COVER CROPS.

OBJECTIVES: 9601839. The goal of the proposed research is to establish the feasibility of genetic engineering winter cover crops that will self-destruct in response to an environmental cue. The objectives are: Transgenic white clover plants carrying a cytotoxic gene (barnase) under the control of a heat shock promoter will be constructed and used as a model to test the survival of the plants at permissive temperatures and the killing effect at the heat shock temperature. To identify promoters that respond to the most reliable environmental signal, photoperiod. These promoters, once characterized, can be used to engineer winter cover crops that will self-destruct in the early summer.

APPROACH: To fulfill the first objective, the following approaches will be taken: Completepilot experiments with transgenic tobacco plants. The performance of T2 and T3 plants and their heat-induced destruction will be evaluated. White clover will be transformed with the heat shock promoter/barnase gene cassettes using a new Agrobacterium-mediated transformation method (Larkin et al., 1996). The heat-induced destruction of the transgenic white clover will be analyzed. To fulfill the second objective, genes that are induced in response to daylength will be identified: Differential display of mRNA method (Liang et al., 1993; Bauer et al., 1993) will be used to isolate cDNAs from N. Sylvestris that are induced under long-day conditions.

PROGRESS: 1999/10 TO 2000/09
Results obtained from Part I of the project indicate that our concept of creating a self-destructive cover crop The goal of the research is to establish the feasiility of genetic engineering winter cover crops that will self-destruct at a desired time. There are two parts of the proposal. Part I uses a heat shock promoter (Phs) to test the feasibility of the strategy. Self-destruction gene cassettes in which a plant Phs controls expression of barnase have been constructed. Because barnase is extremely toxic to the cell, and the Phs is leaky, a copy of the barstar gene driven by the CaMV35S promoter (P35s) was cloned adjacent to the barnase gene. Barstar inhibits barnase by combining with it in a one-to-one complex. We tested three versions of the barnase gene, wild-type, a translation-initiation attenuated mutant (atBN), and a missense mutation in transgenic tobacco. Among all the T1 plants exhibiting the desirable phenotype, only T2 plans from one of the T1 carrying the atBN construct, 105, exhibited heat shock-induced necrosis. Leaves were taken from T3 105 that exhibited the desirable phenotype and from control plants were subjected to heat shock treatments. RNA was extracted from the leaves and RNA blot analysis was conducted. Heat shock protein 70 RNA were induced at least 30 fold after 2 h heat shock. In contrast, the expression of a control gene RBCS was high in both 105 and the control plants before heat shock treatment and was nearly depleted in 105 but unchanged in control plants after heat shock. Because the expression levels of barnase and barstar were low, RTPCR was used to measure the mRNA levels. Consistent with the RBCS expression, the expression levels of barnase increased whereas those of barstar decreased slightly in response to heat shock treatment. In conclusion, we determine that atBN is the desirable version of the barnase to use in this constructs and we tested successfully the concept of creating self-destructive copy crop in response to an environmental factor. In Part II of the project, we used a similar strategy to create transgenic plants that will self-destruct in response to increasing daylength-the most predictable environmental signal. The promoter of the constance (CO) gene whose expression increases with daylength was fused to atBN and linked to barstar driven by the salicylic acid (SA)-inducible promoter of PR-1a gene. The expression of PR-1a is low under normal conditions, but is induced to high levels by salicylic acid. Thus, the low level of barstar expression will save-guard the basal levels of barnase expressed in short day (SD) condition. More importantly, the activity of barnase must be suppressed for seed setting. In control experiments we generated transgenic Arabidopsis plants carrying the CO-GUS or PR-barstar. The expression of the transgene were induced by their respective agents, long day (LD) and SA. The levels of the transgene expression support our hypothesis. However, by repeated attempts we were not able to obtain any transformant with the constructs carrying both barnase and barstar. This failure may be explained by two seniors.

IMPACT: 1999/10 TO 2000/09
Results obtained from Part I of the project indicate that our concept of creating a self-destructive cover crop in response to an environmental factor (temperature) is feasible. Part II of the project will identify genes whose transcription are activated to a high level in response to daylength. The promoter of such a gene can used to construct an ideal self-destructive crop in response to daylength-the most reliable environmental factor. This environmental friendly cover crop will promote the practice of cover crop planting thereby strengthening the sustainable agriculture.

PUBLICATIONS: 1999/10 TO 2000/09
1. Two publications will result from this research: (2000). A manuscript titled: "Genetically engineered self-destruction: an alternative to herbicides for cover crop systems" has been prepared and is ready to be submitted to Plant Physiology after internal review.
2. After further characterization of the long-day inducible genes that we isolated recently, we will prepare a manuscript for publcation. (2000).

Item No. 13 of 13

INVESTIGATOR: Jaynes, D. B.; Kaspar, T. C.; Parkin, T. B.; Moorman, T. B.

PERFORMING INSTITUTION:
NATIONAL SOIL TILTH LABORATORY
2150 PAMMEL DR
AMES, IOWA 50014

NITRATE REMOVAL FROM SUBSURFACE FIELD DRAINS

OBJECTIVES: 9800858. 1) Model the impact of deep-placement of field drain-tubes on water flow paths and denitrification rates. 2) Quantify the capacity of wood chips as a medium for supporting denitrification around field drain-tubes using laboraotry microcosums and 2-D bench scale drainage systems. 3) Verify in the field the rate and extent of denitrification caused by deep-placement of drain-tubes and surrounding drain-tubes with wood-chip/soil mixtures.

APPROACH: Hydraulics and denitrification dynamics of drain-tube submergence will be evaluated with analytical and numerical models of water flow. Influences of hydraulic conductivity and aeration changes with depth will be modeled. Mixing wood-chips into soil surrounding drain-tubes as a carbon source for in-situ denitrification will be studies in laboratory microcosms and bench-scale two-dimensional drain systems. Kinetics of denitrification/assimilation will be quantified. Field verification of drain-tube submergence and wood-chip mediated denitrification will be conducted using plot-scale field drain-tubes in a controlled, complete block study. Denitrification rate, nitrate concentration and loss, and microbial activity will be measured to ascertain effect of drain-tube treatment.

PROGRESS: 1998/12 TO 2001/11
In a laboratory study, we documented the ability of different carbon substrates to support removal of high nitrate (NO3) concentrations under saturated conditions. Several organic substrates were incubated for 180 d under saturated, anaerobic (<0.5 ppm O2) conditions. Substrates used included: corn stalks, wood chips, wood chips soaked in soybean oil, and waste cardboard. Nitrate was repeatedly added to the substrates to sustain a concentration of 100 mg/L NO3-N. The organic materials sustained NO3 removal from water in the following order: corn stalks > ground corn stalks > ground cardboard > cardboard > ground wood and oil > wood and oil > ground wood> wood chips. Corn stalks were able to sustain removal of 77 g N /kg substrate while wood chips were able to sustain removal of 11 g N/kg substrate in 180 d of incubation. Nitrogen immobilization accounted for <2 % of the nitrogen removed from water indicating denitrification as the primary nitrate removal mechanism. A second laboratory study determined the relationships between flow rate and NO3 removal. Columns were filled with a woodchip (<10cm long) and subsoil mixture. Water containing 50 mg/L of NO3-N tagged with 15N (10 atm %) was pumped into the bottom of each column at four flow rates. For 83 d. Results showed that the low flow rate (2.83 cm3/cm2/d) never yielded any nitrate in the effluent, while the effluent at the higher flow rates (6.67, 9.08, 14.29 cm3/cm2/d) contained increasingly higher NO3-N concentrations (18, 24, 34 mg/L). The low flow rate had a removal rate of at least 0.0132 (+/- 0.0029) mg N removed/g wood/d while the highest flow rate had a removal rate of 0.0169 (+/- 0.0044) mg N removed/g wood/d, a significant increase. Nitrogen immobilization accounted for < 6% of the nitrogen removed in all treatments indicating denitrification as the major removal mechanism in all flow rate treatments. Nitrous oxide production was low in all treatments (<0.05% of denitrification) indicating that biofilters may not be great producers of this greenhouse gas. A field site was completed in the fall of 1999 consisting of 12, 150 x 100 ft plots representing four repetitions of three treatments. Plot treatments consisted of a control (standard tile installation), a deep drain line (0.6 m deeper than the control but with an outlet at the same depth as the control), and a buried biofilter (a tile line with a trench filled with wood chips on each side). A tile from each plot was directed to a sump where flow rate and water quality samples were collected. The field plots were planted to corn in 2000 and soybean in 2001. Average nitrate-N concentrations for all samples were 26 mg/L for the control, 24.8 mg/L for the deep tiles, and 9.5 mg/L for the woodchip plots. Only the nitrate-N concentrations for the woodchip treatment were significantly different than the control (P=0.05). The field results confirm the laboratory results that considerable nitrate removal is possible with the buried biofilter design.

IMPACT: 1998/12 TO 2001/11
Tile drainage from fields growing corn and soybean consistently contains nitrate at concentrations exceeding the health standard for drinking water. This nitrate may also cause detrimental environmental effects in the Gulf of Mexico and other salt water estuaries. This study tested two approaches to reducing nitrate in tile drainage by modifying the design of the tiles themselves. We found that adding a carbon source such as wood chips around a tile increases denitrification (a natural microbial process where nitrate is reduced to harmless nitrogen gas) and significantly reduces nitrate in tile drainage. Wide spread adoption of this tile design could greatly reduce nitrate concentrations in surface waters.

PUBLICATIONS: 1998/12 TO 2001/11
1. Greenan, Moorman, Parkin, Jaynes, and Kaspar. 2001. Variable flow rate effects on efficiency of denitrifcation in biofilters. In Agron. Meeting Abstr. Oct 21-25, 2001. Charlotte, NC. Am. Soc. Agron. CD-ROM.
2. Greenan, Moorman, Parkin, Kaspar, and Jaynes. 2000. Enhancing denitrification in the subsurface environment: Efficiency and sustainability of carbon substrates. Agron. Abstr. 92:394.