SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Classification and technology development for area estimation of corn, soybeans, wheat, barley, and sunflowers are outlined. Supporting research for corn and soybean foreign commodity production forecasting is highlighted. Graphs profiling the greenness and brightness of the crops are presented.
Landsat Multispectral Scanner (MSS) data covering a three-county area in northern Illinois were classified using computer-aided techniques as corn, soybeans, or 'other.' Recognition of test fields was 80% accurate. County estimates of the area of corn and soybeans agreed closely with those made by the USDA. Results of the use of a priori information in classification, techniques to produce unbiased area estimates, and the use of temporal and spatial features for classification are discussed. The extendability, variability, and size of training sets, wavelength band selection, and spectral characteristics of crops were also investigated.
Red, photographic infrared, near infrared spectral data of corn, cotton, soybeans, sugar beets, sorghum, sunflowers and tobacco were collected throughout the entire growing season by using a three band handheld radiometer. Different radiance patterns were found among these crops based on their morphology, green biomass duration and leaf size. Results show near infrared radiance is a good indicator of water content in plant tissue under small scale experimental conditions.
The author has identified the following significant results. Like polarization was successful in discriminating corn and soybeans; however, pasture and woods were consistently confused as soybeans and corn, respectively. The probability of correct classification was about 65%. The cross polarization component (highest for woods and lowest for pasture) helped in separating the woods from corn, and pasture from soybeans, and when used with the like polarization component, the probability of correct classification increased to 74%.
Early harvest information helps guide agricultural commodity assessments in Argentina, providing valuable planning information to identify potentially food-insecure regions, anticipate transportation and storage demands, predict price fluctuations, and project commodity trends. However, crop yield estimates are currently subjective, based on interviews with qualified informants (i.e., farmers, agribusiness actors). In partnership with the Buenos Aires Grain Exchange, we leveraged Terra Moderate Resolution Imaging Spectroradiometer (MODIS), Soil Moisture Active Passive (SMAP), and Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (GPM IMERG) NASA Earth observations to develop a Google Earth Engine (GEE) toolset to monitor vegetation growth. The first component of the toolset produces spatial and temporal maps of temperature, precipitation, soil moisture, and the Normalized Difference Vegetation Index (NDVI), allowing users to visualize the influence of the region’s climate and weather. Next, we developed an autoregressive model to predict NDVI several months in advance. Lastly, we created a linear regression model of crop yield and NDVI for soybeans, corn, and wheat, and input the forecasted NDVI to generate a predicted crop yield output. The NDVI forecasting model produced accurate predictions at two, four, and six months when examining the most recent growing season. In the crop yield model, soybeans exhibited moderately strong correlation, wheat had consistent weak correlation, and corn varied from weak to strong correlation depending on zone. This information is vital for vegetation growth monitoring by identifying areas of high growth and allocating resources to areas of lower growth to efficiently maximize crop yields.
Two series of carefully controlled experiments were conducted. First, plots of important crops (corn, soybeans, and sorghum), prairie grasses (big bluestem, switchgrass, tal fescue, orchardgrass, bromegrass), and forage legumes (alfalfa, red clover, and crown vetch) were manipulated to produce wide ranges of phytomass, leaf area index, and canopy architecture. Second, coniferous forest canopies were simulated using small balsam fir trees grown in large pots of soil and arranged systematically on a large (5 m) platform. Rotating the platform produced many new canopies for frequency and spatial averaging of the backscatter signal. In both series of experiments, backscatter of 5.0 GHz (C-Band) was measured as a function of view angle and polarization. Biophysical measurements included leaf area index, fresh and dry phytomass, water content of canopy elements, canopy height, and soil roughness and moisture content. For a subset of the above plots, additional measurements were acquired to exercise microwave backscatter models. These measurements included size and shape of leaves, stems, and fruit and the probability density function of leaf and stem angles. The relationships of the backscattering coefficients and the biophysical properties of the canopies were evaluated using statistical correlations, analysis of variance, and regression analysis. Results from the corn density and balsam fir experiments are discussed and analyses of data from the other experiments are summarized.
The author has identified the following significant results. Preliminary acreage estimates for corn, soybeans, and other cover types made from classification of ERTS-1 data compared well with those made by the U.S. Department of Agriculture. Registration of multiple frames of ERTS-1 CCT data over Lynn County, Texas and DeKalb County, Illinois was achieved to a high degree of accuracy. Spectral/temporal computer pattern recognition analysis was carried out for the first time using satellite data.
The Statistical Reporting Service of the U.S. Department of Agriculture is evaluating ERTS-1 imagery as a potential tool for estimating crop acreage. A main data source for the estimates is obtained by enumerating small land parcels that have been randomly selected from the total U.S. land area. These small parcels are being used as ground observations in this investigation. The test sites are located in Missouri, Kansas, Idaho, and South Dakota. The major crops of interest are wheat, cotton, corn, soybeans, sugar beets, potatoes, oats, alfalfa, and grain sorghum. Some of the crops are unique to a given site while others are common in two or three states. This provides an opportunity to observe crops grown under different conditions. Results for the Missouri test site are presented. Results of temporal overlays, unequal prior probabilities, and sample classifiers are discussed. The amount of improvement that each technique contributes is shown in terms of overall performance. The results show that useful information for making crop acreage estimates can be obtained from ERTS-1 data.
The author has identified the following significant results. ERTS imagery was used as a tool in the identification and refinement of soil association areas; to classify land use patterns between crop and fallow fields; to identify corn, soybeans, and oats; and to identify broad generalized range ecosystems. Various data handling techniques were developed and applied to accomplish these tasks. A map outlining soil associations and relative land values was completed on a base mosaic of ERTS imagery and is included as an appendix to the report.
The results of classifications and experiments for the crop identification technology assessment for remote sensing are summarized. Using two analysis procedures, 15 data sets were classified. One procedure used class weights while the other assumed equal probabilities of occurrence for all classes. Additionally, 20 data sets were classified using training statistics from another segment or date. The classification and proportion estimation results of the local and nonlocal classifications are reported. Data also describe several other experiments to provide additional understanding of the results of the crop identification technology assessment for remote sensing. These experiments investigated alternative analysis procedures, training set selection and size, effects of multitemporal registration, spectral discriminability of corn, soybeans, and other, and analyses of aircraft multispectral data.
The purpose of this study was to determine the statistical separability of multispectral measurements from agricultural cover types: corn, soybeans, green forage (hay and pasture) and forest, in one to twelve spectral channels. Multispectral scanner data in twelve spectral channels in the wavelength range 0.4 to 11.7 microns, acquired for three flightlines were analysed by applying automatic pattern recognition techniques. The same analysis was performed for the data acquired a month later over the same three flightlines to investigate the effect of time on statistical separability of agricultural cover types. In the subsets of one to six spectral channels, the combination of wavelength regions (where V, N, M and T denote the visible, near infrared, middle infrared and thermal infrared wavelength regions, respectively): V, V M, V N M, V N M T, V V N M T, V V N M M T, respectively, were found to be the best choices for getting good overall statistical separability of the agricultural cover types for the data acquired.
A plan for quantifying the crop identification performances resulting from the remote identification of corn, soybeans, and wheat is described. Steps for the conversion of multispectral data tapes to classification results are specified. The crop identification performances resulting from the use of several basic types of automatic data processing techniques are compared and examined for significant differences. The techniques are evaluated also for changes in geographic location, time of the year, management practices, and other physical factors. The results of the Crop Identification Technology Assessment for Remote Sensing task will be applied extensively in the Large Area Crop Inventory Experiment.
The results of classifications and experiments performed for the Crop Identification Technology Assessment for Remote Sensing (CITARS) project are summarized. Fifteen data sets were classified using two analysis procedures. One procedure used class weights while the other assumed equal probabilities of occurrence for all classes. In addition, 20 data sets were classified using training statistics from another segment or date. The results of both the local and non-local classifications in terms of classification and proportion estimation are presented. Several additional experiments are described which were performed to provide additional understanding of the CITARS results. These experiments investigated alternative analysis procedures, training set selection and size, effects of multitemporal registration, the spectral discriminability of corn, soybeans, and other, and analysis of aircraft multispectral data.
Data related problems in the acquisition and use of satellite data necessary for operational forecasting of global crop production are considered for the purpose of establishing a measurable baseline. For data acquisition the world was divided into 37 crop regions in 22 countries. These regions represent approximately 95 percent of the total world production of the selected crops of interest, i.e., wheat, corn, soybeans, and rice. Targets were assigned to each region. Limited time periods during which data could be taken (windows) were assigned to each target. Each target was assigned to a cloud region. The DSDS was used to measure the success of obtaining data for each target during the specified windows for the regional cloud conditions and the specific alternatives being analyzed. The results of this study suggest several approaches for an operational system that will perform satisfactorily with two LANDSAT type satellites.
The author has identified the following significant results. Data sets for corn, soybeans, winter wheat, and spring wheat were used to evaluate the following schemes for crop identification: (1) per point Gaussian maximum classifier; (2) per point sum of normal densities classifiers; (3) per point linear classifier; (4) per point Gaussian maximum likelihood decision tree classifiers; and (5) texture sensitive per field Gaussian maximum likelihood classifier. Test site location and classifier both had significant effects on classification accuracy of small grains; classifiers did not differ significantly in overall accuracy, with the majority of the difference among classifiers being attributed to training method rather than to the classification algorithm applied. The complexity of use and computer costs for the classifiers varied significantly. A linear classification rule which assigns each pixel to the class whose mean is closest in Euclidean distance was the easiest for the analyst and cost the least per classification.
The radiometric measurements over bare field and fields covered with grass, soybean, corn, and alfalfa were made with 1.4 GHz and 5 GHz microwave radiometers during August - October 1978. The measured results are compared with radiative transfer theory treating the vegetated fields as a two layer random medium. It is found that the presence of a vegetation cover generally gives a higher brightness temperature T(B) than that expected from a bare soil. The amount of this T(B) excess increases in the vegetation biomass and in the frequency of the observed radiation. The results of radiative transfer calculations generally match well with the experimental data, however, a detailed analysis also strongly suggests the need of incorporating soil surface roughness effect into the radiative transfer theory in order to better interpret the experimental data.
Over 600 fields in the corn, soybean and wheat growing regions of the Argentine pampa were categorized by crop or cover type and ancillary data including crop calendars, historical crop production statistics and certain cropping practices were also gathered. A summary of the field work undertaken is included along with a country overview, a chronology of field trip planning and field work events, and the field work inventory of selected sample segments. LANDSAT images were annotated and used as the field work base and several hundred ground and aerial photographs were taken. These items along with segment descriptions are presented. Meetings were held with officials of the State Secretariat of Agriculture (SEAG) and the National Commission on Space Investigations (CNIE), and their support to the program are described.
The mobile agricultural radar sensor was used to collect data from 31 fields in the floodplain of the Kansas River east of Lawrence, Kansas during the summer of 1981. Corn, soybeans, and wheat crops were observed from May 1 to November 11. Radar backscattering measurements were acquired at 10.2 GHz for VV and VH polarizations at 50 deg incidence angles for all fields and at 30 deg, 40 deg, 50 deg, 60 deg, and 70 deg for nine of the 31 fields. Target parameters describing the vegatation and soil characteristics, such as plant moisture, plant height, soil moisture, etc., were also measured. The methodology, radar backscatter data and associated ground-truth data obtained during this experiment are documented.