Papers on Precision Farming

The Design of a Remote Sensing Data Acquisition Campaign for Precision Agriculture and Some Early Results
D. Rickman1, J. C. Luvall1, JM Wersinger2, P. Mask2, D. E. Kissel3
1GHCC/MSFC/NASA Huntsville, AL, 2Auburn University, 3University of Georgia,

Abstract
In the 1970s NASA and the Department of Agriculture attempted to use the new Landsat MSS system for agricultural purposes. The program had relatively little success. With the advent of differential GPS, yield monitors on harvest equipment and higher spatial resolution remote sensing systems it seemed likely the situation should be reexamined. Therefore, a campaign of data acquisition involving remote sensing and other modalities with dependent research was assembled and funded by the Space Grant Consortia in Alabama and Georgia.

The design of the remote sensing data acquisition was driven by the biology and physics of the crop system and limited by the available sensor platforms. Major parameters included crop stage, spatial resolution, seasonal and daily weather conditions, and which portion of the EM spectrum would actually capture the most discriminating information. Joint visible and Near IR with Thermal IR would permit use of existing indices, such as greenness, as well as phenomena driven by the plant's evapotranspiration. Spatial resolution in the 2-5 meter range was chosen, avoiding many complexities caused by aliasing crop row spacing at higher resolutions yet finer than the harvester's tightest recording rate. This dictates use of an airborne system. Use of an airborne system also makes scheduling around weather much simpler than use of satellite data. Active video calibration was recognized as essential if quantitative measures were ever to be obtained or reproduced. The system would also have to have onboard geometry recorded during data acquisition. The only system which is currently available that meets all of these criteria is the Atlas scanner, flown out of NASA's Stennis Space Center.

Based on these elements 3 data acquisitions have been flown. Seven flight lines were flown twice in 1998 and 16 lines flown in 1999. Total raw data is several GBytes. All of the data has now been geometrically corrected and some preliminary analysis accomplished. The thermal bands have an extremely high correlation with yield. For one test case with corn, correlation in excess of 0.86 was obtained from a data acquisition two months prior to harvest! Soil images show significant within field variation in clay, soil brightness and emissivity. Light wind has been found to effect the reflectance and temperature of broad leaf crops, including soybeans, cotton and peanuts. Clearly, this work has already demonstrated some very important results. With continued development of the remote sensing technology there is good reason to believe this research will soon be able to help the individual farmer.


Determining Surface Soil Clay Concentration at a Field Scale for Precision Agriculture
F. Chen1, D. E. Kissel1, R. Clark1, L.T. West1, D. Rickman2, J. Luvall2, and W. Adkin1
1University of Georgia, 2GHCC-NASA at Huntsville

Abstract
The soil's clay and organic matter are the chemical reactive soil components that affect soil processes important to optimum nutrition of crops. Knowing their concentrations in soil may be valuable for decision making about crop production inputs. Previous studies have shown that remote sensing data can be successfully applied to directly estimate the organic carbon concentration of surface soil. However, the information from remote sensing data for estimating soil clay concentration is untested in agricultural applications. Some data transformation techniques will be essential to derive the needed information. The soil electronic conductivity may also be used to estimate soil clay concentration. The objective of this study was to determine whether the clay concentration of surface soil could be predicted from remotely sensed imagery. The soil electronic conductivity was measured at shallow depth as an alternative method for comparison. A 115 hectare field located in Crisp County, Georgia, was selected for the study. A total of 59 samples of surface soil, with their sub-meter GPS locations, were obtained from the field. The soil clay concentrations of the soil samples were determined using standard laboratory procedures. Two approaches were examined for this study. In the first approach, a scanned image of a color aerial photograph slide of the bare surface soil was used, and the image data was transformed by principal components analysis. The principal component 3 was then selected for further analysis since from a visual check, it appeared that this principal component may represent the clay concentration over the field. A linear relationship between the principal component 3 and the clay content of surface soil was developed based on 28 samples taken initially from the field, and the clay content distribution of surface soil was determined over the field based on this relationship.

For the alternative method, a total of 21,653 soil electrical conductivity measurements for the 0-0.25 meter depth were taken over the field with their sub-meter GPS locations using the Veris model 3100 sensor of soil conductivity. Based on the measured data, the distribution of the electrical conductivity of surface soil was then derived with spatial statistical analysis. A raster file was created to store the electrical conductivity of surface soil over the field with a 2 by 2 meter cell size. Based on the locations of 40 surface soil samples, the surface electrical conductivity values at these 40 locations were determined by averaging the nearest 9 cells within a 3 by 3 window centered at each location. Then, a linear relationship between electrical conductivity and clay concentration of surface soil at the 40 locations was developed. Finally, the soil clay concentration map was developed based on this relationship.

For both approaches, five levels of clay concentration of surface soil were classified for the field. The results indicated that the clay concentration of surface soil could be estimated from an aerial color photograph slide by the use of principal components analysis. However, the effect of organic carbon confounded the results in depressional areas of high organic carbon. Therefore, the prediction of the clay concentration of surface soil in the depressional areas was less accurate. The study also showed the feasibility of estimating surface soil clay content from soil electrical conductivity data. The effect of organic carbon on the prediction for the depression areas from electrical conductivity was not significant.


Remote Sensing of Urban Thermal Landscape Characteristics and Their Affects on Local and Regional Meteorology and Air Quality: An Overview of NASA EOS-IDS Project ATLANTA
Dale A. Quattrochi, Jeffrey C. Luvall, and Maurice G. Estes, Jr.
NASA
Earth Science Department
SD60
Marshall Space Flight Center
Global Hydrology and Climate Center
Huntsville, Alabama 35812

Abstract
As an entity, the city is a manifestation of human "management" of the land. The act of city-building, however, drastically alters the biophysical environment, which ultimately, impacts local and regional land-atmosphere energy exchange processes. Because of the complexity of both the urban landscape and the attendant energy fluxes that result from urbanization, remote sensing offers the only real way to synoptically quantify these processes. One of the more important land-atmosphere fluxes that occurs over cities relates to the way that thermal energy is partitioned across the heterogeneous urban landscape. The individual land cover and surface material types that comprise the city, such as pavements and buildings, each have their own thermal energy regimes. As the collective urban landscape, the individual thermal energy responses from specific surfaces come together to form the urban heat island phenomena, which prevails as a dome of elevated air temperatures over cities. Although the urban heat island has been known to exist for well over 150 years, it is not understood how differences in thermal energy responses for land covers across the city interact to produce this phenomenon, or how the variability in thermal energy responses from different surface types drive its development. Additionally, it can be hypothesized that as cities grow in size through time, so do their urban heat islands. The interrelationships between urban sprawl and the respective growth of the urban heat island, however, have not been investigated. Moreover, little is known of the consequential effects of urban growth, land cover change, and the urban heat island as they impact local and regional meteorology and air quality.

Research is currently ongoing on how remote sensing data obtained from aircraft and satellite platforms, can be used to measure land cover changes associated with urbanization, and to quantify thermal energy responses across the city landscape. This study, known as Project ATLANTA (ATlanta Land use ANalysis: Temperature and Air quality), seeks to observe, measure, model, and analyze how the rapid growth of the Atlanta, Georgia, USA metropolitan area since the early 1970's has impacted the meteorology and air quality of the region. The objectives of this research effort are: 1) To investigate and model the relationship between Atlanta's urban growth, land cover change, and the development of the urban heat island phenomenon; 2) To investigate and model the relationship between Atlanta's growth on air quality; and 3) To model the overall effects of urban development on surface energy budget characteristics across the Atlanta metropolitan area. Our key goal is to derive a better scientific understanding of how land cover changes associated with urbanization in the Atlanta area, principally in transforming forest lands to urban land covers through time, has, and will, affect local and regional meteorology, surface energy flux, and air quality characteristics. Allied with this goal is the prospect that the results from this research can be applied by urban planners, environmental managers, and other decision-makers, for determining how urban sprawl has impacted the overall environment of the Atlanta area.


Thermal remote sensing: A powerful tool in the characterization of landscapes on a functional basis.
Luvall, Jeffrey C.1, James Kay2, and Roydon Fraser3.
1NASA, Marshall Space Flight Center Huntsville, AL 35812 (256)922-5886
2Dept. of Environment and Resource Studies, University of Waterloo,Waterloo, Ontario, Canada N2L 3G1,
3Dept. of Mechanical University of Waterloo,Waterloo, Ontario, Canada N2L 3G1.

Abstract
Thermal remote sensing instruments can function as environmental measuring tools, with capabilities leading toward new directions in functional landscape ecology. Theoretical deduction and phenomenological observation leads us to believe that the second law of thermodynamics requires that all dynamically systems develop in a manner which dissipates gradients as rapidly as possible within the constraints of the system at hand. The ramification of this requirement is that dynamical systems will evolve dissipative structures which grow and complexify over time. This perspective has allowed us to develop a framework for discussing ecosystem development and integrity. In the context of this framework we have developed measures of development and integrity for ecosystems. One set of these measures is based on destruction of the energy content of incoming solar energy. More developed ecosystems will be more effective at dissipating the solar gradient (destroying its exergy content). This can be measured by the effective surface temperature of the ecosystem on a landscape scale. These surface temperatures are measured using airborne thermal scanners such as the Thermal Infrared Multispectral Scanner (TIMS) and the Airborne Thermal/Visible Land Application Sensor(ATLAS) sensors. An analysis of agriculture and forest ecosystems will be used to illustrate the concept of ecological thermodynamics and the development of ecosystems.


Responsible Official: Dr. Steven J. Goodman (steven.goodman@nasa.gov)
Technical Contact: Dr. Doug Rickman (doug.rickman@msfc.nasa.gov)
Page Curator: Diane Samuelson (diane.samuelson@msfc.nasa.gov)


Last Updated: October 4, 1999