An Innovative Approach for Predicting Farmers’ Adaptive Behaviors at the Large Watershed Scale: Implications for Water Quality

by  Adriana Valcu-Lisman,¹ Philip W. Gassman, J. Gordon Arbuckle, and Yiannis Panagopoulos

Predicted changes in climatic patterns (higher temperatures, changes in extreme precipitation events, and higher levels of humidity) will affect agricultural activity. The concept of adaptation to evolving climatic and/or economic conditions is an important aspect of the agricultural decision-making process (Wright Morton et al. 2014). Adopting cover crops and no-tillage (no-till) farming, extending subsurface tile drainage systems, and adjusting crop management are only a few examples of adaptive actions (Arbuckle and Laws 2014). These actions can be implemented provided they have private benefits, including increased profits and reduced risk. However, each adaptive action has unique economic and environmental impacts. Cover crops and no-till farming typically produce positive impacts on water quality, but tile drainage often results in reduced water quality (Gassman et al. 2023; Rittenburg et al. 2015). At the same time, cover crops can have a negative impact on crop yields while an expansion of tile drainage could boost crop yields and the acreage available to row crops. 

Understanding farmers’ adaptations to climate shifts is critical for understanding future US Corn Belt agricultural productivity and environmental consequences of land use change. Previous efforts have predicted future change by looking at historical change. While informative, this approach is limited because it relies on past experience and does not include farmers’ perspectives on adaptation. This research addresses this gap by providing a proof-of-concept regarding the importance of bridging disciplines to address complex questions related to future climate adaptation and water quality.

The goal of this study is to determine changes in water quality in response to prospective adaptive measures adopted by farmers. To answer this research question, we employ a multi-step approach: (a) estimate the likelihood that these actions will occur; (b) identify the agricultural areas where these actions are most likely to be implemented, thus creating plausible scenarios; and, (c) simulate the water quality impacts associated with each of these scenarios. We apply our modeling efforts to the Upper-Mississippi River basin (UMRB) and the Ohio-Tennessee River basin (OTRB), which comprise most of the Corn Belt region and are critical source areas relative to achieving reduction of the seasonal hypoxic zone in the northern Gulf of Mexico (Kling et al. 2014; Panagopoulos et al. 2015; Gassman et al. 2023). 

In our analysis, we integrate statistical estimation methods with an ecohydrological simulation model. First, we employ a logit model to estimate farmers’ probability to adopt either cover crop or tile drainage using data from a survey described below. Second, we use county-level data to extrapolate these results to the two large watersheds considered in our analysis. By extrapolating these results, we estimate the percentage of cropland dedicated to either conservation practice at the county level. Third, we estimate the additional area that needs to be treated with cover crops and/or tile drainage.

To determine the impact of adopting these conservation practices on water quality, we identify the spatial allocation of the existent conservation practices (NASS 2012) as well as the spatial allocation of the additional estimated treated acreages. Using additional analysis, we construct a baseline scenario (adoption level of conservation practices as of 2012) and several scenarios representing the additional adoption of cover crops, tile drainage, and the combination of the two practices. 

The likelihood of each adaptive agricultural action is estimated using data from a survey conducted in 2012. The survey used a large, representative sample of 4,778 farmers in the Corn Belt (figure 1) to elicit farmers’ perspectives on projected future climate and beliefs about the need for mitigation strategies (Arbuckle et al. 2013, 2014; Loy et al. 2013; Roesch-McNally et al. 2017). The survey also provided insights about farmer’s behavioral intentions regarding three of the most important agricultural adaptation strategies: no-till, cover crops, and tile drainage (Roesch-McNally et al. 2017). These data were used to study the relationships between intention to adapt, farmer characteristics, farm characteristics, and weather characteristics, and to predict the probability of adoption for each action. These estimated probabilities were then used to create different scenarios for the UMRB and ORTB. Finally, the impact of these scenarios on water quality was performed using the Soil and Water Assessment Tool (SWAT) ecohydrological model (Gassman et al. 2007; Williams et al. 2008; Arnold et al. 2012; SWAT 2025), which was calibrated/validated for the UMRB and OTRB (Panagopoulos et al. 2015). 

The survey process measured farmer’s potential adaptation strategies in response to projected climate change (Arbuckle et al. 2014; Roesch-McNally et al. 2017). We focus here on cover crops and tile drainage, which provide contrasting adaptation strategies in response to current climate versus future projected climate conditions. Roesch-McNally et al. (2017) describe an analysis of the survey data that evaluated farmer’s adaptive decision making in the context of Corn Belt agricultural production and climatic influences. This analysis builds on their previous adaptation assessment with further assessments of the probability that farmers in the region will adopt tile drainage and/or cover crops as strategies to offset future projected climate impacts. Although the survey occurred over a decade ago, it remains one of the largest and most comprehensive data sets regarding potential farmer adaptation trends in the Corn Belt. 

Cover crops are primarily implemented in winter fallow periods in the context of Corn Belt row crop production, using cereal rye and other vegetative cover (Christianson et al. 2021). Well-established cover crops can provide considerable environmental benefits within corn-soybean and other cropping systems, including potentially large reductions in nitrate leaching, phosphorus losses, soil erosion, and increased soil organic carbon (Tonitto et al. 2006; Fageria et al. 2011; Christianson et al. 2021). Cover crops can result in some negative externalities: (a) increased dissolved phosphorus losses under certain conditions (Liu et al. 2019; Christianson et al. 2021); and, (b) suppressed yields of subsequent grain crop yields due to allelopathic effects (Koehler-Cole et al. 2020). 

Installation of tile drains in the Corn Belt was initiated in the late 1800s and has expanded considerably since that time (McCorvie and Lant 1993; Blann 2009; Chakravorty et al. 2024). An extensive spatial tile drainage network of over 46 million acres now exists in the six most drained states in the region (Valayamkunnath et al. 2020). Tile drains facilitate row crop production by reducing water stress and increasing crop productivity, enabling earlier planting and reducing risk, and improving economic returns to crop producers (Randall and Goss 2008; Blann et al. 2009; Chakravorty et al. 2024). Tile drainage can provide some specific environmental benefits including reduced nitrous oxide emissions and soil erosion and increased water infiltration (Blann et al. 2009; King et al. 2015; Chakravorty et al. 2024). However, the Midwest tile drainage network has exacerbated nitrate and soluble phosphorus losses to the UMRB and OTRB stream systems, and ultimately to the Gulf of Mexico (Randall and Goss 2008; Blann et al. 2009; King et al. 2015; Gassman et al. 2023; Chakravorty et al. 2024).

One of the survey goals was specifically to determine if farmers are likely to implement different agricultural actions in response to climate change. Additionally, the survey also asked for farmers’ demographic information, acreage farmed, and land tenure information. The responses related to the use of different farming practices as adaption strategies were elicited through a question asking farmers whether they would increase/stay the same/decrease adoption of different practices. The surveyed regions (6-digit watersheds in figure 1), and additional UMRB and OTRB areas, were subdivided into more refined 12-digit Hydrologic Unit Code (HUC12) watersheds or subbasins (Jones et al. 2022), which typically range in size from 10,000 to 40,000 acres and are referred to as subbasins here. These 12-digit watersheds represent the basic unit analysis in SWAT. The baseline distribution of cropland versus non-cropland areas is shown in figure 2, based on the dominant land use determined at the subbasin level. Across the two major watersheds, a total of 4,166 subbasins (12-digit watersheds) were identified as cropland production subareas versus 7,913 non-cropland subbasins.

A statistical (logit) model was used to analyze the econometric strategies of the farmers. The probability of adopting a given agricultural option at the farm level, given the type of tenure (owner versus renter), is modeled as a function of a series of farmers’ demographic measures, soil quality, and a series of weather-related variables (See online appendix; Valcu-Lisman et al. 2018). The farmer demographics include age and farming experience, measured in years. The Soil Survey Geographic Database (SSURGO; USDA 2025) was used to define soil quality as the county-level percent of cropland under different land capabilities classes, where land capability classes 1 to 4 are generally suitable for cultivation versus classes 5 to 8 which are less suitable for crops. Farm characteristics include land tenure arrangement measured as the percent of cropland that is owned or rented. Additionally, data from the NASS 2012 census (USDA-NASS 2019b) was used to include the percent county level of cropland that was managed with tile drainage or cover crops. These county level cropland data (and non-cropland data) were converted to the 12-digit subbasin level using an area-based weighting process. The weather characteristics include statistical measures for temperature and precipitation measured as 2007–2011 seasonal or monthly averages. The Parameter-elevation Relationships on Independent Slopes Model (PRISM) data set (Daly et al. 2008; NACSE 2025) was used to construct different measurements for the weather characteristics (i.e., average vs. extreme weather measurements). 

Based on farmers’ survey responses and their tenure status (owner versus renter), the probability to adopt (increase) tile drainage and/or cover crops was predicted using the logit models. The final probability of adopting cover crops and/or tile drainage was computed as the weighted sum of the two probabilities. This probability was interpreted as a percentage of the cropland to be allocated to one or both of the management practices. The statistical results of the predicted probabilities of increased adoption of cover crops and tile drainage, based on the logit models, are shown in table 1.

The logit models (See appendix) and NASS 2012 county-level land use data (USDA-NASS 2019b) were then used to predict the additional cropland areas that would be managed with either cover crops and/or tile drainage, beyond the cropland already managed with those practices that were identified in the survey. Tabulated totals of baseline, and predicted additional and total subbasins, that were treated with cover crops and tile drainage are listed in table 2 for the OTRB, UMRB, and combined OTRB and UMRB regions. The distribution of baseline and predicted additional subbasins treated with cover crops, and remaining cropland and non-cropland subbasins, are shown in figure 3 for the UMRB and OTRB. Analogous distributions for subbasins treated with tile drains are shown in figure 4. Figure 5 further shows that 1,537 subbasins treated with cover crops and tile drainage, versus 443 subbasins treated only with cover crops and 1,616 subbasin treated only with tile drains (figure 5).

The SWAT-predicted water quality impacts for the expanded adoption of cover crops and tile drainage are presented in table 3. Increased use of cover crops resulted in reduced sediment, total nitrogen (N), and total phosphorus (P) losses, ranging from -8.2% to -13.1% reductions relative to the baseline. Implementation of expanded tile drainage was predicted to increase total N losses for the OTRB (12.5%) and UMRB (14.3%), which were dominated by nitrate losses via the tile drains and other flow pathways. Small increases of approximately 2% in sediment loss were predicted across the region in response to the increased tile drain adoption (table 3), which in turn led to slight increases of total P for the OTRB (2.4%) and UMRB (3.0%), and also contributed to the increase of total N (due to sediment-bound P and N). The increased sediment loss is somewhat counterintuitive regarding tile drainage providing a potential soil erosion benefit as previously discussed. However, the simulated sediment losses reflect cropland water balance dynamics that resulted in small surface runoff increases across the study region.  This may be due to the inability of the original SWAT empirical tile drainage routines (used in this study) to adequately replace water table fluctuations. The use of a more physically-based tile drainage option resulted in improved water balance and streamflow estimates in a more recent SWAT study for the Des Moines River basin (Brighenti et al. 2024). The use of this tile drain option may also result in more accurate sediment loss estimates.

Farmers are increasingly adopting agricultural conservation practices not only for their perceived economic and ecological benefits but also in response to changing weather patterns. It is equally important to understand not only what drivers are behind adoption but also their potential impact on the environment. Furthermore, while the adoption of a conservation practice is field specific, the environmental impacts, such as water quality, are further reaching. This study uses a proof-of-concept analysis that links the findings of a survey designed to understand farmers’ likelihood of adopting various agricultural practices as adaptive measures to climate shifts with large-scale watershed ecohydrological modeling to quantify water quality impacts. A statistical analysis using only the survey data is used to determine the farmers’ likelihood to adopt cover crops and/or tile drainage. Further analysis is employed to extrapolate the surveyed-based findings to a larger area and construct large scale scenarios to simulate the changes at water quality level. The results suggest that cover crops are predicted to increase at a lower rate than tile drainage. The SWAT-based analysis shows that the positive impacts on water quality associated with cover crops are offset by negative outcomes associated with tile drainage.

Footnotes

1. Author Valcu-Lisman was employed at Iowa State University when the research was conducted. The findings and conclusions in this article are those of the authors and should not be construed to represent any official USDA or US Government determination or policy.

References

Arbuckle Jr., J.G., J. Hobbs, A. Loy, L.W. Morton, L.S. Prokopy, and J. Tyndall. 2014. “Understanding Corn Belt Farmer Perspectives on Climate Change to Inform Engagement Strategies for Adaptation and Mitigation. Journal of Soil and Water Conservation. 69(6):505-516. https://doi.org/10.2489/jswc.69.6.505

Arbuckle, J., and L. Laws. 2014. “Farmer Perspectives on Climate Change." Resilient Agriculture, Iowa State University Extension and Outreach pp. 20-23. https://store.extension.iastate.edu/product/Farmer-Perspectives-on-Climate-Change   

Arbuckle Jr., J.G., L.S. Prokopy, T. Haigh, J. Hobbs, T. Knoot, C. Knutson, A. Loy, A.S. Mase, J. McGuire, L.W. Morton, J. Tyndall, and M. Widhalm. 2013. “Climate Change Beliefs, Concerns, and Attitudes Toward Adaptation and Mitigation Among Farmers in the Midwestern United States.” Climatic Change 117:943–950. https://doi.org/10.1007/s10584-013-0707-6

Arnold, J.G., D.N. Moriasi, P.W. Gassman, K.C. Abbaspour, M.J. White, R. Srinivasan, C. Santhi, R.D. Harmel, A. van Griensven, M.W. Van Liew, N. Kannan, and M.K. Jha. 2012. “SWAT: Model Use, Calibration, and Validation.” Transactions of the ASABE 55(4):1491-1508. https://doi.org/10.13031/2013.42256

Blann, K.L., J.L. Anderson, G.R. Sands, and B. Vondracek. 2009. “Effects of Agricultural Drainage on Aquatic Ecosystems: A Review.” Critical Reviews in Environmental Science and Technology 39(11):909-1001. https://doi.org/10.1080/10643380801977966

Brighenti, T.M., P.W. Gassman, J. Arnold, and J. Thompson. 2023. “Comparison of Two Tile-drain Methods in SWAT via Temporal and Spatial Testing for an Iowa Watershed.” Journal of the ASABE 66(6):1555-1569. https://doi.org/10.13031/ja.15534

Chakravorty, R., H. Feng, D.A. Hennessy and M.J. Castellano. 2024. “Agricultural Tile Drainage in the US Corn Belt: Past, Present, and Future.” Agricultural Policy Review Fall 2024. Center for Agricultural and Rural Development, Iowa State University. https://agpolicyreview.card.iastate.edu/fall-2024/agricultural-tile-drainage-us-corn-belt-past-present-and-future

Christianson, R., J. Fox, N. Law, and C. Wong. 2021. “Effectiveness of Cover Crops for Water Pollutant Reduction from Agricultural Areas.” Transactions of the ASABE 64(3). https://doi.org/10.13031/trans.14028

Daly, C., M. Halbleib, J.I. Smith, W.P. Gibson, M.K. Doggett, G.H. Taylor, J. Curtis, and P.P. Pasteris. 2008. “Physiographically Sensitive Mapping of Climatological Temperature and Precipitation Across the Conterminous United States.” International Journal of Climatology 28(15):2031-2064. https://doi.org/10.1002/joc.1688

Fageria, N.K., V.C. Baligar, and B.A. Bailey. 2005. “Role of Cover Crops in Improving Soil and Row Crop Productivity.” Communications in Soil Science and Plant Analysis 36(19-20):2733-2757.https://www.tandfonline.com/doi/abs/10.1080/00103620500303939

Gassman, P.W., J. Mount and K. Schilling. 2023. “A Review of Ecohydrological and Hydraulic/Flood Models for Evaluating the Water Quality and Flood Mitigation Potential of Natural Infrastructure (NI) Practices in the Mississippi-Atchafalaya River Basin.” Prepared for Environmental Defense Fund, Washington, DC. 23-SR 123. Center for Rural and Agricultural Development, Iowa State University. https://www.card.iastate.edu/products/publications/synopsis/?p=1375

Gassman, P.W., M. Reyes, C.H. Green, and J.G. Arnold. 2007. The Soil and Water Assessment Tool: Historical development, applications, and future directions. Transactions of the ASABE (Invited Paper Series) 50(4):1211-1250. https://doi.org/10.13031/2013.23637

King, K.W., M.R. Williams, M.L. Macrae, N.R. Fausey, J. Frankenberger, D.R. Smith, P.J.A. Kleinman, and L.C. Brown. 2015. “Phosphorus Transport in Agricultural Subsurface Drainage: A Review.” Journal of Environmental Quality 44(2): 467-485. https://doi.org/10.2134/jeq2014.04.0163

Kling, C.L., Y. Panagopoulos, A. Valcu, P.W. Gassman, S. Rabotyagov, T. Campbell, M. White, J.G. Arnold, R. Srinivasan, M.K. Jha, J. Richardson, L.M. Moskal, G. Turner, and N. Rabalais. 2014. “Land Use Model Integrating Agriculture and the Environment (LUMINATE): Linkages between Agricultural Land Use, Local Water Quality and Hypoxic Concerns in the Gulf of Mexico Basin.” European Journal of Agricultural Economics 41(3): 431-459. https://doi.org/10.1093/erae/jbu009 

Koehler-Cole, K., S.E. Everhart, Y. Gu, C.A. Proctor, M. Marroquin-Guzman, D.D. Redfearn, R.W. Elmore. 2020. “Is Allelopathy from Winter Cover Crops Affecting Row Crops?” Agricultural & Environmental Letters 5(1):e20015. https://doi.org/10.1002/ael2.20015 

Liu, J., M.L. Macrae, J.A. Elliott, H.M. Baulch, H.F. Wilson, and P.J.A. Kleinman. 2019. “Impacts of Cover Crops and Crop Residues on Phosphorus Losses in Cold Climates: A Review.” Journal of Environmental Quality 48(4):850-868. https://doi.org/10.2134/jeq2019.03.0119 

Loy, A., J. Hobbs, J.G. Arbuckle Jr., L.W. Morton, L.S. Prokopy, T. Haigh, T. Knoot, C. Knutson, A.S. Mase, J. McGuire, J. Tyndall, and M. Widhalm. 2013. “Farmer Perspectives on Agriculture and Weather Variability in the Corn Belt: A statistical Atlas.” https://changingclimate.osu.edu/webinars/ppt/moore-statistical-atlas-2014.pdf

CSCAP 0153-2013. Cropping Systems Coordinated Agricultural Project (CAP): Climate Change, Mitigation, and Adaptation in Corn-based Cropping Systems, Iowa State University. https://shop.iastate.edu/extension/farm-environment/natural-resources-and-environment/water-quality/cscap1532013.html

McCorvie, M.R., and C.L. Lant. 1993. “Drainage District Formation and the Loss of Midwestern Wetlands, 1850-1930.” Agricultural History 67(4):13-39. https://www.jstor.org/stable/3744552 

NACSE. 2025. “PRISM Climate Data.” PRISM Climate Group, Northwest Alliance for Computational Science & Engineering, Oregon State University. https://prism.oregonstate.edu/

Panagopoulos, Y., P.W. Gassman, M.K. Jha, C.L. Kling, T. Campbell, R. Srinivasan, M. White, and J.G. Arnold. 2015. “Towards the Development of an Integrated Modeling System for the Corn Belt: A Refined Regional Modeling Approach.” Journal of Hydrology 524:348-366. https://doi.org/10.1016/j.jhydrol.2015.02.039

Pollak, R.A., and T.J. Wales. 1991. “The Likelihood Dominance Criterion: A New Approach to Model Selection.” Journal of Econometrics 47(2–3):227-242. https://doi.org/10.1016/0304-4076(91)90100-R

Rittenburg, R.A., A.L. Squires, J. Boll, E.S. Brooks, Z.M. Easton, and T.S. Steenhuis. 2015. “Agricultural BMP Effectiveness and Dominant Hydrological Flow Paths: Concepts and a Review.” Journal of the American Water Resources Association 51(2):305-329. https://doi.org/10.1111/1752-1688.12293

Roesch-McNally, G.E., J.G. Arbuckle, and J.C. Tyn. 2017. “What Would Farmers Do? Adaptation Intentions under a Corn Belt Climate Change Scenario.” Agriculture and Human Values 34(2):333-346. https://doi.org/10.1007/s10460-016-9719-y

Randall G.W., and M.J. Goss. 2008. “Nitrate Losses to Surface Water Through Subsurface, Tile Drainage.” In: D.R. Keeney a and J.L. Hatfield (eds.) Nitrogen in the Environment: Sources, Problems, and Management (ch. 6) 2nd Ed. Elsevier Inc., Amsterdam, Netherlands. https://doi.org/10.1016/B978-0-12-374347-3.00006-8 

Schmidt, M. 2020. “Modernizing Agricultural Drainage Law in Iowa.” Iowa Environmental Council. https://www.iaenvironment.org/newsroom/water-and-land-news/iec-releases-report-on-modernizing-drainage-law-in-iowa

SWAT. 2025. Soil & Water Assessment Tool. Texas A&M AgriLife, Texas A&M University, Blacklands Research & Extension Center and U.S. Department of Agriculture, Agricultural Research, Grassland Soil and Water Research Laboratory. https://swat.tamu.edu/

Tonitto, C., M.B. David, and L.E. Drinkwater. 2006. “Replacing Bare Fallows with Cover Crops in Fertilizer-intensive Cropping Systems: A Meta-analysis of Crop Yield and N Dynamics.” Agriculture, Ecosystems & Environment 112(1):58-72. https://doi.org/10.1016/j.agee.2005.07.003

US Department of Agriculture (USDA). 2025. “Soil Survey Geographic Database (SSURGO).” US Department of Agriculture Natural Resources Conservation Service, Washington, D.C. https://www.nrcs.usda.gov/resources/data-and-reports/soil-survey-geographic-database-ssurgo

US Department of Agriculture National Agricultural Statistics Service (USDA-NASS). 2019a. “USDA Census of Agriculture Historical Archive: 2007.” US Department of Agriculture National Agricultural Statistics Service, Washington, D.C. https://agcensus.library.cornell.edu/archive/census_year/2007-census/index.html

US Department of Agriculture National Agricultural Statistics Service (USDA-NASS). 2019b. "USDA Census of Agriculture Historical Archive: 2012." U.S. Department of Agriculture National Agricultural Statistics Service, Washington, D.C. https://agcensus.library.cornell.edu/census_year/2012-census/

Jones, K.A., L.S. Niknami, S.G. Buto, and D. Decker. 2022. “Federal Standards and Procedures for the National Watershed Boundary Dataset (WBD). Techniques and Methods.” Federal Standards Book 11, Collection and Delineation of Spatial Data (fifth ed.), 11-A3, Chapter 3 of Section A. US Department of the Interior, US Geological Survey, and US Department of Agriculture Natural Resources Conservation Service, Washington, D.C. https://pubs.usgs.gov/tm/11/a3/

Valayamkunnath, P., M. Barlage, F. Chen, D.J. Gochis and K.J. Franz. 2020. “Mapping of 30-meter Resolution Tile-drained Croplands Using a Geospatial Modeling Approach.” Scientific Data 7:257. https://doi.org/10.1038/s41597-020-00596-x

Valcu-Lisman, A.M., Y. Ji, C.L. Kling, and J.G. Arbuckle. 2018. “Evidence from a Corn Belt Farmers’ Survey: A Multi-Layered Analysis of Understanding Farmers’ Adaption Strategies to Climate Change.” Agricultural and Applied Economics Association Conference, 2018 Annual Meeting, August 5-7, 2018, Washington, D.C. https://doi.org/10.22004/ag.econ.274414

Williams, J.R., J.G. Arnold, J.R. Kiniry, P.W. Gassman, and C.H. Green. 2008. “History of Model Development at Temple, Texas.” Hydrological Sciences Journal 53(5):948-960. https://doi.org/10.1623/hysj.53.5.948

Wright Morton, L., J. Hobbs, J.G. Arbuckle Jr., and A. Loy. 2014. “Corn Belt Farmers' Adaptation to Increased Precipitation.” Resilient Agriculture, Iowa State University Extension and Outreach, pp. 24-25. https://store.extension.iastate.edu/product/Corn-Belt-Farmers-Adaptation-to-Increased-Precipitation

Appendix A: Logit Models Used in the Analysis

Summary survey question: (figure A.1) with a focus on cover crops and tile drainage

• Suppose the following scenario were to happen in the near future:
  • Violent storms/extreme rain events will become more frequent, particularly in the spring.
  • More extreme rain events will increase the likelihood of flooding and saturated soils.
  • Periods between rains will become longer, increasing likelihood of drought.
  • Changes in weather patterns will increase crop insect, weed, and disease problems.
• If you knew with certainty that the above conditions would occur, would use of the following practices and strategies on the cropland you own and rent decrease, increase, or stay the same? (Please select one answer each for owned and rented land, if applicable.)
  • Cover crops
  • Tile drainage
  • Decrease, Stay the same, Increase, I don’t know

General method for processing the survey data 

• Eliminate states that are not of interest: Kansas and North Dakota
• Four possible initial answers: decrease (very few), stay the same, increase, I don’t know
• Drop the “I don’t Know” type of answers
• “Decrease” answers are coded as “Stay the same”, coded as “0” in the logit
• “Increase” are coded as “1”
• Run separate logit for “Owned” and “Rented”

The basic logit model specification

P_i (Increase = 1|LandOwned)
= f(farmer characteristics, farm characteristics, climate or weather)

P_i (Increase = 1|LandRented)
= f(farmer characteristics, farm characteristics, climate or weather)

Farmer characteristics:

• Age
• Experience

Climate/ Weather characteristics (season averages or monthly averages)

• Total Season (April –Sept) growing degree days above 50F, average 2007–2011
• Monthly average (April –Sept) growing degree days above 50F, average 2007–2011
• Total Season Precipitations (April –Sept), average 2007–2011
• Monthly average precipitations average 2007–2011
• Extreme precipitations
  • Heavy precipitation events are counted as any days when the daily precipitation exceeded the 99th percentile of daily precipitation for a given month. The 99th percentile was defined separately for each station and each month. As an example, the 99th percentile for May precipitation is found by assembling all daily precipitation in May from 1971–2011 for a particular station. Then the 99th percentile of this empirical distribution of about 31x41=1271 values is found. We consider the proportion of days with precipitation exceeding the 99th percentile for the five-year period 2007–2011. Note that one would expect this to be about 0.01 by chance.
• Aridity Index
  • The aridity index is a composite weather index that has been linked with crop yield. The index combines standardized precipitation and maximum temperature anomalies. Thus, a hot and dry month will have a positive index, while a humid and cool day will have a negative index.
• Heat stress degree days
  • Characterizes the cumulative impact of hot weather. The sum of maximum temperature over some threshold (86F for corn). Standardize SDD for each county and each year. Take the average over 2007–2011
  • Tmax, maximum temperature, April-September, average 2007–2011

Climate/ Weather characteristics

• Historical weather: 2007–2011 (initial analysis) 

Land quality

• Dominant land capability class (LCC), the LCC class with the highest percentage area, county level

Suggested citation

Valcu-Lisman, A., P.W. Gassman, J.G. Arbuckle, and Y. Panagopoulos. 2025. " An Innovative Approach for Predicting Farmers’ Adaptive Behaviors at the Large Watershed Scale: Implications for Water Quality." Agricultural Policy Review Fall 2025. Center for Agricultural and Rural Development, Iowa State University. https://agpolicyreview.card.iastate.edu/fall-2025/innovative-approach-predicting-farmers-adaptive-behaviors-large-watershed-scale