Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Research Project #431325

Research Project: Improving Agroecosystem Services by Measuring, Modeling, and Assessing Conservation Practices

Location: Hydrology and Remote Sensing Laboratory

2021 Annual Report

Objective 1: Assess the status and trends of the Lower Chesapeake Bay agroecosystem through measurements and modeling. Subobjective 1.1 Establish long-term data streams for the LCB-LTAR project to assess agroecosystem status and trends. Sub-objective 1.2 Assess data streams as a function of spatial differences in land use. Objective 2: Develop and test remote sensing methods to assess crop conditions, conservation practices, and nutrient use efficiency. Subobjective 2.1: Improve remote sensing methods for assessing crop conditions using plant phenology at field to watershed scales. Subobjective 2.2: Develop remote sensing methods to assess crop residue cover and soil tillage intensity at field to watershed scales. Subobjective 2.3: Develop and test methods using high-spatial-resolution remote sensing from small unmanned aircraft systems for precision agriculture. Subobjective 2.4: Retrieve leaf optical properties by remote sensing foliar water content to improve estimation of plant nitrogen status. Subobjective 2.5: Use LiDAR, Synthetic Aperture Radar, and Landsat to map and characterize wetlands and riparian buffers. Objective 3: Quantify environmental processes within agricultural landscapes to evaluate ecosystem services and best management practices. Subobjective 3.1: Improve measurement and modeling approaches to describe agrochemical emissions and transport from agricultural operations. Subobjective 3.2: Characterize the influence of canopy structure on the deposition of agrochemicals to riparian buffers. Subobjective 3.3: Quantify the spatial and temporal variability and assess the fate of atmospheric ammonia on the Delmarva Peninsula. Subobjective 3.4: Assess the effects of wetland hydroperiod on carbon storage. Subobjective 3.5: Quantifying impacts of watershed characteristics and crop rotations on winter cover crop nitrate uptake capacity within agricultural watersheds using the SWAT model.

Much of the research will be conducted within the LCB-LTAR study area (Appendix 2) in support of the LTAR network goals. Two types of studies will be performed as part of the network, monitoring for long-term trends and conducting experiments to identify, quantify, and understand the underlying agroecosystem processes causing the trends. Thus, measurements of soil, water, and air quality are a priority. Within the LCB-LTAR, the Choptank River Watershed on the Delmarva Peninsula (Figure 3) has been a research site since 2004 for the USDA-NRCS Conservation Effects Assessment Program (CEAP) (Hively et al. 2011; Maresch et al. 2008; McCarty et al. 2008; Niño de Guzmán et al. 2012; Richardson et al. 2008; USDA-NRCS 2011; Tomer and Locke 2011; Tomer et al. 2014, Whithall et al. 2010). The approaches include remote sensing, in-situ monitoring, long term sampling scenarios, and modeling efforts. The Optimizing Production Inputs for Economic and Environmental Enhancement (OPE3) experimental site consists of a 22-ha production field and adjacent riparian area that has been studied by this team since 1998. OPE3 is an outdoor laboratory at the USDA-ARS Beltsville Agricultural Research Center (BARC) to explore energy, water, nutrient, and agrochemical processes.

Progress Report
The Fiscal Year 2021 was the fifth year of this project in National Program 212, Soil and Air. ARS scientists from Beltsville, Maryland, made progress conducting interdisciplinary research to address all objectives of the project plan. The first objective of this project is to assess the Lower Chesapeake Bay agroecosystem via measurements and modeling and the establishment of the Lower Chesapeake Bay Long-Term Agroecosystem Research (LCB-LTAR) site. The main locations within the LCB-LTAR project are the Optimizing Production Inputs for Economic and Environmental Enhancement (OPE3) experimental watershed located near the Beltsville Agricultural Research Center (BARC) in Beltsville, Maryland, and the Choptank River watershed (CRW) located on the Delmarva Peninsula. During the current fiscal year, ongoing measurements of meteorological conditions, surface fluxes, crop phenology, and other environmental data were collected. Beyond the LTAR Network, these datasets were submitted to the Ameriflux and PhenoCam networks. Also, the micrometeorological tower located in CWR was updated to improve data collection and processing. Equipment was purchased for two in-situ phosphorus monitoring systems to be installed at gage stations in CWR. Data were collected at multiple LTAR Network locations as a part of the Watershed Lag Time Project (WLTP). A post-doctoral researcher was hired to analyze the effects of land use, hydrology, and other environmental characteristics on lag time. Also, a new project was initiated to evaluate the use of the plant Miscanthus to mitigate phosphorus pollution in the waterways of LCB-LTAR. Two modules describing the sediment dynamics of particulates and dissolved organic carbon in aquatic environments were evaluated in the USDA Soil Water Assessment Tool model (SWAT) model using a four-year observational dataset. The results showed good agreement between the model output and observations, suggesting the model may be a valuable tool for assessing ecosystem service within watersheds. The second objective is to enhance the utility of remote sensing using airborne and satellite platforms to measure biophysical variables related to agricultural production and environmental assessment. The within-season termination (WIST) algorithm, which maps cover crop termination dates in the near real-time, was assessed over Maryland and Delaware using a Harmonized Landsat and Sentinel-2 (HLS) dataset that had been calibrated against cover crop biomass measurements collected in the field. Maps of the cover crop termination date from WIST have been provided to the Maryland Department of Agriculture winter cover crop program in 2020 and 2021. Similarly, the within-season emergence (WISE) algorithm for mapping crop emergence date was used with HLS data to create 30-m resolution maps over four Corn Belt states (Iowa, Illionois, Indiana, Minensota) from 2017 to 2020. These maps were compared to planting dates and National Agricultural Statistics Service weekly crop progress reports; the results showed WISE captured the spatial and temporal variability of crop emergence over the study region. Remote sensing tools were used to map crop residue cover and soil tillage intensity over several states (South Dakota, North Dakota, Minnesota) using 10 years of Landsat data. These maps were validated against field-level survey data. A crop emergence map over Iowa derived from the HLS dataset was produced to guide the selection of satellite images for mapping crop residue and assess remote sensing indices to characterize tillage intensity classes in Iowa. Since the diffuse reflectance of a leaf is determined by its chemical composition, information about how these properties are related is fundamental to using hyperspectral remote sensing to monitor vegetation. To understand leaf optical properties, the diffuse reflectance of an infinitely thick leaf was computed and compared to the canopy reflectance from a radiative transfer model. Also, imagery from WorldView-3, Landsat, Sentinel, VENµS, and PlanetScope was acquired over BARC; it will be used to assess the influence of spatial and temporal resolutions on these relationships. Building on improvements to SWAT that enhance its ability to simulate wetland ecosystem function, an improved method for appraising wetland carbon storage was developed. This work supports the implementation of environmental regulations related to wetlands and riparian buffers within CRW. The third objective focuses on characterizing environmental processes within agricultural landscapes to evaluate ecosystem services and best management practices. Although field and laboratory activities were limited due to restrictions related to COVID, progress was made toward achieving the goals associated with this objective. The research was conducted to examine chemical and environmental processes affecting dicamba drift and volatilization. Maps of poultry house density on the Delmarva Peninsula were generated, and a new ammonia sampling regime was developed. A collaborative study with the U.S. Forest Service established the relationship between wetland hydroperiod and termite feeding. This information facilitates the use of the InVEST model to assess the impact of wetlands on carbon storage and climate mitigation. A preliminary assessment of carbon storage in the wetlands in the Mid-Atlantic region was conducted; refinements to this approach are ongoing. A synthesis of studies conducted in CRW during the last decade as a part of the USDA Conservation Effects Assessment Project was completed. These studies used remote sensing and modeling approaches to assess the capacity of winter cover crops to take up residual nitrogen, decrease nitrogen losses to groundwater, and reduce soil erosion. The results of the study showed that nitrogen reduction efficiency decreases by nearly 35% as the planting date of the cover crop shifts from October 1 to November 1. Project Summary: This project combined field measurements, remote sensing using both airborne and satellite platforms, and multiple modeling schemes to better understand environmental processes and improve methods for evaluating crop health, monitoring conservation practices, and assessing soil, air, and water quality. As part of this project, ARS scientists in Beltsville, Maryland, and their collaborators developed new approaches to monitoring crop phenology that combines remote sensing imagery from multiple platforms. The work provided improve methods for creating high-resolution maps of the vegetations density and robust methods for predicting the start and termination dates of crop growth in near real-time. Enhanced methods for monitoring crop residue and tillage practices at sub-field to regional scales via remote sensing were also developed. These techniques provide tools that allow growers and local conservation districts to determine the effect of conservation practices on soil health and identify when additional conservation efforts are needed. Remote sensing methods for evaluating the performance of winter cover crops in limiting erosion and nutrient runoff from agricultural land were developed. This project has also yielded new guidelines and methods for using imagery from satellites and unmanned aircraft for monitoring agricultural systems. For example, novel approaches for mapping crop health and monitoring crop damage due to insects were developed. These methods will be helpful for the precision application of irrigation, fertilizers, and other agrochemicals. Several tools to map landscape characteristics, such as the texture, denitrification potential, and carbon distribution of soils, using topographic information were developed. These mapping tools use Light Detection and Ranging (LiDAR) and digital elevation model data to produce high-resolution maps with a broad range of applications, such as improving the ability of watershed hydrology models to predict soil drainage and implementing technologies to minimize the environmental impact of crop production. As a part of this project, ARS scientists at Beltsville, Maryland enhanced our understanding of flow pathways and their influence on the effectiveness of riparian buffers in maintaining water quality, the value of winter cover crops for mitigating nitrate loading in aquatic systems, and the impact of soil carbon content and tillage practices on the rate of soil erosion. These activities have yielded new measurement techniques and contributed to several improvements to SWAT. For example, new modules were incorporated into the model to improve its ability to describe sediment dynamics and the water flow from depressional wetlands to streams. As a part of this project, the factors influencing the volatilization, atmospheric transport, and fate of agrochemicals and other agricultural pollutants were investigated. The results yielded improve methods for measuring agrochemical fluxes in the field and the development of a model to predict the atmospheric transport of particulates, ammonia, volatile organic compounds, and other agrochemicals. The model can be used to assess the effectiveness of vegetative environmental buffers (VEB) for capturing these pollutants. This information has been incorporated into NRCS guidelines for the use of VEBs as a remediation strategy.

1. Development of a new approach to detect cover crop termination dates for agroecosystem services. Cover crops are planted to reduce soil erosion, increase soil fertility, and improve watershed management. Cost-sharing programs require that winter cover crops be planted and terminated within the specified time window. Usually, cover crop planting and termination dates are obtained through field surveys, which is labor-intensive. Detection of cover crop termination using remote sensing has had limited success due to the lack of high spatial and temporal resolution observations and proper methods. ARS scientists proposed a new within-season termination (WIST) algorithm to map cover crop termination dates using high spatial and temporal resolution remote sensing imagery. Termination dates from remote sensing data were compared to the field operation records in the Beltsville Agricultural Research Center (BARC) experimental fields in 2019 and 2020. Results show that cover crop termination dates can be reliably detected and may become routine in the future. Detecting cover crop termination within the season using remote sensing provides a quick and economical way to support agroecosystem services.

2. Reconstruct daily 30-m Normalized Difference Vegetation Index (NDVI) for crop monitoring. High spatial and temporal resolution remote sensing data are required for monitoring crop progress and conditions at the sub-field scale. Data fusion approaches have been developed to fuse remote sensing imagery from different sensors to generate frequent observations at a high spatial resolution. However, these approaches have been challenging to apply in highly heterogeneous areas, especially in complex agricultural landscapes. ARS scientists have developed a novel method to reconstruct daily 30 m Normalized Difference Vegetation Index (NDVI) using a crop reference curve (CRC) extracted from 500 m pure MODIS (Moderate Resolution Imaging Spectroradiometer) pixels. The CRC-based method was applied over a complex agricultural landscape in the Choptank River watershed on the eastern shore of Maryland. Results show that the CRC method outperforms the image pair-based data fusion algorithm when clear Landsat images are scarce. The approaches resulting 30 m NDVI time-series data produced by this approach would enable accurate crop monitoring at the sub-field scale.

3. Estimates of conservation tillage practices using satellite remote sensing. The USDA Environmental Quality Incentives Program (EQIP) provides technical and financial assistance to encourage producers to adopt conservation practices. Historically, one of the most common practices is conservation tillage, which protects the soil with more crop residue cover than intensive tillage. This research identified agricultural fields with crop residue cover in the 58,000 square mile study area in South Dakota, North Dakota, and Minnesota using 10 years of Landsat Thematic Mapper data. The results were validated against field-level survey data. This study demonstrated that researchers could implement retrospective estimates of conservation tillage with sufficient accuracy using the Landsat Archive, which is available at no cost.

4. Spectral discrimination using infinite leaf reflectance and simulated canopy reflectance. Biodiversity is an important indicator of ecosystem health, so an important objective of remote sensing is to identify the dominant plant species living in an area. Different plant species have foliage with similar chemical compositions, differing mostly in their relative amounts. However, plants of the same species have differences in leaf reflectance caused by environmental factors, such as light exposure and soil nutrient availability. The question is how to enhance the remotely-sensed signal from foliar chemical composition and at the same time how to suppress the contribution of extraneous environmental factors. Based on physical models for the optics of glass plates, one solution may be the theoretical reflectance for an infinitely thick stack of leaves related to chemical concentration. Infinite leaf reflectance provides an estimate of plant canopy reflectance at a very high leaf area index. A simulation model created nine groups of leaf and canopy reflectance spectra representing different leaf morphologies and chemical composition. Five different algorithms quantified similarity between infinite leaf reflectance and canopy model simulations. While infinite leaf reflectance may not be used to estimate canopy reflectance, these results indicate that infinite leaf reflectance may be used to compare chemical composition for monitoring biodiversity.

5. Mapping forested wetland inundation using Deep Convolutional Neural Networks. Forested wetlands are most common along the Atlantic Coastal Plain, including the Delmarva Peninsula. The inundation status of those wetlands provides a key indicator of climate variability and shifts in hydrological (e.g., floodwater storage), biogeochemical (e.g., carbon sequestration), and biological (e.g., habitat) functions. However, many of these forested wetlands occur in small, shallow depressions and possess an overstory of forests, and many wetlands are only inundated for a short period throughout the year, usually in early spring. These characteristics are a challenge to mapping wetland inundation using commonly available optical imagery. In this study, we used artificial intelligence (AI) models based on deep neural networks to map wetland inundation using optical imagery and demonstrate that AI has an advantage over other statistical approaches for mapping inundation. This use of AI will greatly improve the accuracy of wetland maps and improve estimates of ecosystem services provision by wetlands in agricultural landscapes.

6. Use of digital soil disaggregation maps in a low-relief landscape to support wetland restoration. The field of digital soil mapping has developed in response to the growing need for soils data and the enormous advances in remote sensing and information technology that permit rapid generation of soil property and classification maps. The goal of this study was to explore the potential of digital soil mapping techniques to improve the identification of wetland soils and map soil properties on a low relief depressional wetland landscape. Separate models were constructed to predict natural soil drainage and texture class on forest and cropland using soil profile data collected by local soil surveyors and other sources of soil expert knowledge. The models produced maps with greater than 70% accuracy in predicting natural soil drainage and texture class for forested depressions. These maps have the potential to improve watershed models and inform our understanding of wetland hydrology in agricultural landscapes.

7. Using satellites to remotely monitor and evaluate the performance of winter cover crops. Winter cover crops have been shown to limit erosion and nutrient runoff from agricultural land. To promote their usage, the Maryland Department of Agriculture (MDA) subsidizes farmers who plant cover crops. The effectiveness of cover crops depends on management practices and agronomic factors such as planting date, method, and crop species. In partnership with the MDA, NASA's DEVELOP program utilized imagery from Landsat 5, Landsat 8, and the European Space Agency’s Sentinel-2 to create a decision support tool (DST) for satellite-based monitoring of cover crop performance throughout Maryland. A series of DEVELOP teams created the DST based on an interactive graphical user interface in Google Earth Engine which analyzes satellite imagery to calculate an index for measuring cover crop growth. With this DST, the MDA can analyze the effectiveness of cover crops with reduced need to manually spot-check enrolled production fields can identify variables influencing overall cover crop performance to optimize implementation of their winter cover crop program via adaptive management approaches. Citation (372430): Peredo, J., Wayman, C., Whong, B., Thieme, A., Kline, L.R., Yadav, S., Eder, B., Lenske, V., Portillo, D., McCartney, S., Fitz, J., Oddo, P., Keppler, Hively, D., Bolten, J., McCarty, G.W., Lyon, A. 2020. Utilizing Landsat and Sentinel-2 to remotely monitor and evaluate the performance of winter cover crops throughout Maryland. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. 42, 125-130.

8. Use of topographic models for mapping soil properties. A study of landscape topography is an assessment of the current terrain features and a representation of the landform. Because topography reflects elevation changes within detailed landform features over a region, it can significantly impact soil processes. Rapid advancements in aerial, space and geographic technologies have led to large-scale availability of digital elevation models (DEMs), providing detailed topographic information for topographic model generation. We compared the performances of two types of topography-based models using advanced statistical methods. We found that the combined use of these models allows for local calibration of a general regional topographic model. This approach leads to the generation of accurate soil maps with a very limited number of new samples from a location under study.

Review Publications
Goldman, M., Needleman, B., Rabenhorst, M., Lang, M., McCarty, G.W. 2020. Digital soil disaggregation in a low-relief landscape to support wetland restoration decisions. Geoderma. 373:114420.
Meyers, E., Kerekes, J., Daughtry, C.S., Russ, A.L. 2019. Assessing the impact of satellite revisit rate on estimation of phenological transition timing.. Remote Sensing. 11(21):2558.
Hannun, R., Wofle, G., Kawa, S., Haniscol, T., Newman, P., Alfieri, J.G., Barrick, J., Clark, K., Digangi, K., Diskin, G., King, J., Kustas, W.P., Mitra, B., Noormets, A., Nowak, J., Thornhill, K., Vargas, R. 2020. Spatial heterogeneity in CO2 and CH4 fluxes: insights from airborne eddy covariance measurements over the Mid-Atlantic region. Ecological Research. 15/035008.
Sun, L., Gao, F.N., Xie, D., Anderson, M.C., Chen, R., Yang, Y., Yang, Y., Chen, Z. 2020. Reconstructing daily 30 m vegetation index over complex agricultural landscapes using crop reference curves approach. Remote Sensing of Environment. 253:112156.
Du, L., McCarty, G.W., Zhang, X., Lang, M., Vanderhoff,M.K., Li, X., Huang, C., Lee, S. 2020. Mapping forested wetland inundation in the Delmarva peninsula, USA using a deep convolutional neural network. Remote Sensing. 12:644.
Peredo, J., Wayman, C., Whong, B., Thieme, A., Kline, L., Yadav, S., Eder, B., Lenske, V., Portillo, D., McCartney, S., Fitz, J., Oddo, P., Keppler, J., Bolten, J., McCarty, G.W., Iyon, A. 2020. Utilizing Landsat and Sentinel-2 to remotely monitor and evaluate the performance of winter cover crops throughout Maryland. Photogrammetry and Remote Sensing International Archives. 42:125-130.
Qi, J., Zhang, X., Lee, S., Wu, Y., Moglen, G.E., McCarty, G.W. 2020. Modeling sediment diagenesis processes on riverbed to better quantify aquatic carbon fluxes and stocks in a small watershed of the mid-Atlantic region. Carbon Balance and Management. 15:13.
Li, X., McCarty, G.W., Du, L., Lee, S. 2020. Use of topographic models for mapping soil properties and processes. Soil Systems. 4:32.
Beeson, P.C., Daughtry, C.S., Wallander, S.A. 2020. Estimates of conservation tillage practices using Landsat archive. Remote Sensing. 12(16):2665.
Gao, F.N., Anderson, M.C., Hively, W.D. 2020. Detecting cover crop end-of-season using VENS and Sentinel-2 satellite imagery. Remote Sensing. 12(21):3524.
Hunt Jr, E.R. 2021. Spectral discrimination using infinite leaf reflectance and simulated canopy reflectance. . International Journal of Remote Sensing. 42(8):3039-3055.
Nino De Guzman, G., Hapeman, C.J., Millner, P.D., Torrents, A., Jackson, D., Kjellerup, B. 2018. Presence of organohalide-respiring bacteria in and around a permeable reactive barrier at a trichloroethylene-contaminated Superfund site. Environmental Pollution. 243(2018):766-776.