Paul C. Doraiswamy

Crop Yield Assessment from Remote Sensing

Monitoring crop condition and production estimates at the state and county level is of great interest to the U.S. Department of Agriculture. The National Agricultural Statistical Service (NASS) of the U.S. Department of Agriculture conducts field interviews with sampled farm operators and obtains crop cuttings to make crop yield estimates at regional and state levels. NASS needs supplemental spatial data that provides timely information on crop condition and potential yields. In this research, the crop model EPIC (Erosion Productivity Impact Calculator) was adapted for simulations at regional scales. Satellite remotely sensed data provide a real-time assessment of the magnitude and variation of crop condition parameters, and this study investigates the use of these parameters as an input to a crop growth model. This investigation was conducted in the semi-arid region of North Dakota in the southeastern part of the state. The primary objective was to evaluate a method of integrating parameters. retrieved from satellite imagery in a crop growth model to simulate spring wheat yields at the sub-county and county levels. The input parameters derived from remotely sensed data provided spatial integrity, as well as a real-time calibration of model simulated parameters during the season, to ensure that the modeled and observed conditions agree. A radiative transfer model, SAIL (Scattered by Arbitrary Inclined Leaves), provided the link between the satellite data and crop model. The model parameters were simulated in a geographic information system grid, which was the platform for aggregating yields at local and regional scales. A model calibration was performed to initialize the model parameters. This calibration was performed using Landsat data over three southeast counties in North Dakota. The model was then used to simulate crop yields for the state of North Dakota with inputs derived from NOAA AVHRR data. The calibration and the state level simulations are compared with spring wheat yields reported by NASS objective yield surveys.

An Improved ASTER Index for Remote Sensing of Crop Residue

Unlike traditional ground-based methodology, remote sensing allows for the rapid estimation of crop residue cover (fR). While the Cellulose Absorption Index (CAI) is ideal for fR estimation, a new index, the Shortwave Infrared Normalized Difference Residue Index (SINDRI), utilizing ASTER bands 6 and 7, is proposed for future multispectral sensors and would be less costly to implement. SINDRI performed almost as well as CAI and better than other indices at five locations in the USA on multiple dates. A minimal upgrade from one broad band to two narrow bands would provide fR data for carbon cycle modeling and tillage verification.

EPIC Modeling of Soil Organic Carbon Sequestration in Croplands of Iowa

Depending on management, soil organic carbon (SOC) is a potential source or sink for atmospheric CO(2). We used the EPIC model to study impacts of soil and crop management on SOC in corn (Zea mays L.) and soybean (Glycine max L. Merr.) croplands of Iowa. The National Agricultural Statistics Service crops classification maps were used to identify corn-soybean areas. Soil properties were obtained from a combination of SSURGO and STATSGO databases. Daily weather variables were obtained from first order meteorological stations in Iowa and neighboring states. Data on crop management, fertilizer application and tillage were obtained from publicly available databases maintained by the NRCS, USDA-Economic Research Service (ERS), and Conservation Technology Information Center. The EPIC model accurately simulated state averages of crop yields during 1970-2005 (R(2) = 0.87). Simulated SOC explained 75% of the variation in measured SOC. With current trends in conservation tillage adoption, total stock of SOC (0-20 cm) is predicted to reach 506 Tg by 2019, representing an increase of 28 Tg with respect to 1980. In contrast, when the whole soil profile was considered, EPIC estimated a decrease of SOC stocks with time, from 1835 Tg in 1980 to 1771 Tg in 2019. Hence, soil depth considered for calculations is an important factor that needs further investigation. Soil organic C sequestration rates (0-20 cm) were estimated at 0.50 to 0.63 Mg ha(-1) yr(-1) depending on climate and soil conditions. Overall, combining land use maps with EPIC proved valid for predicting impacts of management practices on SOC. However, more data on spatial and temporal variation in SOC are needed to improve model calibration and validation.

Spectral Reflectance of Wheat Residue during Decomposition and Remotely Sensed Estimates of Residue Cover

Remotely sensed estimates of crop residue cover (fR) are required to assess the extent of conservation tillage over large areas; the impact of decay processes on estimates of residue cover is unknown. Changes in wheat straw composition and spectral reflectance were measured during the decay process and their impact on estimates of fR were assessed. Proportions of cellulose and hemicellulose declined, while lignin increased. Spectral features associated with cellulose diminished during decomposition. Narrow-band spectral residue indices robustly estimated fR, while broad-band indices were inconsistent. Advanced multi-spectral sensors or hyperspectral sensors are required to assess fR reliably over diverse agricultural landscapes.

Crop classification in the U.S. Corn Belt using MODIS imagery

Landcover classification is essential in studies of landcover change, climate, hydrology, carbon sequestration, and yield prediction. The potential for using NASAs MODIS sensor at 250-meter resolution was investigated for USDAs operational programs. This research was conducted over Iowa and Illinois to classify corn and soybean crops. Multitemporal 8-day composite 250-meter-resolution surface reflectance product time series were used to generate the NDVI data, which were used to differential between corn and soybean crops in the U.S. Corn Belt. The results of the MODIS-based classification were compared with the Landsat-based classification for the 2-year period. The overall classification accuracy for Iowa was 82%, and for Illinois 75%. In conclusion, this method has been used successively during the 2002-2006 years to develop crop classifications and products for crop conditions and potential yield maps for Iowa and Illinois.

Carbon Sequestration and Land Degradation

Storing carbon (C) in soil as organic matter is not only a viable strategy to sequester CO2 from the atmosphere, but is vital for improving the quality of soil. This presentation describes (1) C sequestration concepts and rationale, (2) relevant management approaches to avoid land degradation and foster C sequestration, and (3) a summary of research quantifying soil C sequestration. The three primary greenhouse gases (CO2, CH4, and N2O) derived from agriculture have increased dramatically during the past century. Conservation management practices can be employed to sequester C in soil, counter land degradation, and contribute to economic livelihoods on farms. Trees can accumulate C in perennial biomass of above-ground and below-ground growth, as well as in the deposition of soil organic matter. Minimal disturbance of the soil surface with conservation tillage is critical in avoiding soil organic C loss from erosion and microbial decomposition. Animal manures contain 40–60% C, and therefore, application to land promotes soil organic C sequestration and provides readily-available, recycled nutrients to crops. Green manures can be used to build soil fertility, often with leguminous plant species having symbiotic root associations with nitrogen-fixing bacteria. Grasslands have great potential to sequester soil organic C when managed properly, but can also be degraded due to overgrazing, careless management, and drought leading to accelerated soil erosion and undesirable species composition. Opportunities exist to capture and retain greater quantity of C from crop and grazing systems when the two systems are integrated. Fertilization is needed to achieve production goals, but when applied excessively it can lead to environmental pollution, especially when considering the energy and C cost of manufacture and transport. Agricultural conservation management strategies to sequester CO2 from the atmosphere into soil organic matter will also likely restore degraded land and/or avoid further land degradation.

Changes of crop rotation in Iowa determined from the United States Department of Agriculture, National Agricultural Statistics Service cropland data layer product

Abstract. Crop rotation is one of the important decisions made independently by numerous farm managers, and is a critical variable in models of crop growth and soil carbon. In Iowa and much of the Midwestern United States (US), the typical management decision is to rotate corn and soybean crops for a single field; therefore, the land-cover changes each year even though the total area of agricultural land-use remains the same. The price for corn increased from 2001 to 2010, which increased corn production in Iowa. We tested the hypothesis that the production increase was the result of changes in crop rotation in Iowa using the annual remote sensing classification (the cropland data layer) produced by the United States Department of Agriculture, National Agricultural Statistics Service. It was found that the area planted in corn increased from 4.7 million hectares in 2001 to 5.7 million hectares in 2007, which was correlated with the market price for corn. At the county level, there were differences in how the increase in corn production was accomplished. Northern and central counties had little land to expand cultivation and generally increased corn production by converting to a corn–corn rotation from the standard corn–soybean rotation. Southern counties in Iowa increased corn production by expanding into land that was not under recent cultivation. These changes affect the amount of soil carbon sequestration.

Improved Techniques for Crop Classification using MODIS Imagery

Brazil has become a major player in world soybean markets, second only to the U.S. Brazil Crop area is about 10 million hectares and is now rapidly expanding into the Brazilian savannah (Cerrado) and the Amazonian region where forested area is being converted to cropland. There is a need for accurate updated information on the newly expanded agricultural areas in Brazil and the current total production. The objective of this research was to develop an operational method for assessing soybean crop area that would facilitate developing remote sensing based algorithms for assessing crop yields in major producing areas. The moderate resolution imaging spectroradiometer (MODIS) onboard the Terra satellite offers a good potential for assessing crop area as well as provide opportunity to retrieve crop condition parameters that can be used to assess crop yields. A three-year MODIS data set was acquired for the study and this research describes the methods used for processing the 8-day composite reflectance data from bands 1 and 2 and its use in developing the classification of soybean crop area in four major soybean producing areas in Brazil. The results suggest methods that can be used for operational application of MODIS 250m data for classification as well as potential use in crop yield assessment.

Classification techniques for mapping biophysical parameters in the US southern Great Plains

This research was part of a 1997 NASA-USDA collaborative research study for regional mapping of soil moisture in the US southern Great Plains. The study was conducted during a transition from winter to summer crops in June and July. Classification techniques were developed using Landsat TM data to separate the transitional vegetation from the natural vegetation. Areal photography and associated ground truth data for selected areas were used in developing the spectral signatures for each vegetation class. Classification of the entire region used these signatures. This research suggests techniques to be used in mapping vegetation classes in a very mixed land use area and suggests ways of using multi temporal satellite imagery for separating vegetation classes and minimizing cloud cover problems.

Improved modeling of soil organic carbon in a semiarid region of Central East Kazakhstan using EPIC

Inappropriate land use and soil mismanagement produced wide-scale soil and environmental degradation to the short-grass steppe ecosystem in the semiarid region of central east Kazakhstan. A limitation for determining the impacts of land use changes on soil organic carbon (SOC) is the dearth of information on SOC stocks under the predominant land uses in the region. Here we used the Environmental Policy Integrated Climate (EPIC) model to study long-term impacts of land use changes and soil management on SOC to a depth of 50 cm during 1955–2030, in degraded agricultural lands of central east Kazakhstan. Simulated land uses were: native rangeland vegetation, wheat (Triticum aestivum L.), wheatgrass (Agropyron cristatum L.), and abandoned croplands. The EPIC model was initialized with soil properties obtained from a soil map of the study area. Data on crop management, fertilizer application and tillage practices were gathered from local expert knowledge. Simulations were performed for each polygon on a land use classification map, resulting in 4661 simulations. Our results showed that simulated SOC explained 50% of the variation in measured SOC. Of the 1.38 million hectares in the study area, 78% were under native vegetation, 3% cultivated to wheat, 8% on wheatgrass, and 11% were abandoned croplands in 2005. If land use remained constant, total stock of SOC would decrease at an annual rate of 723 kg C ha−1. However, if best management practices are implemented, resulting in reallocation of land use according to the land capability with abandoned croplands being converted to reduced-tillage wheat or wheatgrass, total stock of SOC would increase to an equivalent of 4700 kg C ha−1 yr−1. Combining land use classification and soil maps with EPIC, proved valid for studying impacts of land use changes and management practices on SOC; an important aspect of this approach is the ability to scale up site-specific SOC to the region. With the available data, EPIC produced relatively accurate results but more data on spatial and temporal variation in SOC are needed to improve model calibration and validation.