Progress 10/01/03 to 09/30/06
Outputs
In collaboration with Allegany County and Ontario/Yates Counties, New York, soil surveys, research focused on using proximal, diffuse hyperspectral reflectance spectroscopy to characterize the relationship between lab- and in situ soil spectral reflectance and selected soil properties, sampled by horizon, that are important to soil survey. This includes participation in field soil survey activities during 2004 and 2005 to sample several pedons in Allegany County and in Ontario/Yates Counties resulting in 62 samples. Sub-samples were obtained, from the bulk NRCS samples for characterization, for both hyperspectral and a limited set of laboratory analyses at Cornell University. This sample and the more detailed soil chemical and physical properties were analyzed by the national soil survey laboratory to correlate hyperspectral reflectance with selected soil properties important to soil survey. During the final year, we conducted statistical analyses of the spectral and
laboratory data to develop predictive models of selected soil properties which will be published in one M.S. degree thesis during spring 2007.
Impacts
Proximal, diffuse reflectance spectroscopy of soil samples provides a low-cost alternative to direct measurements of chemical and physical properties of soils where large numbers of observations are required to accurately characterize an area or where the cost of field survey and laboratory analysis is high.
Publications
- No publications reported this period
|
Progress 01/01/05 to 12/31/05
Outputs
In collaboration with Allegany County and Ontario/Yates Counties, New York, soil surveys, research continues to focus on using hyperspectral reflectance spectroscopy to characterize the relationship between lab- and in situ soil spectral reflectance and selected soil properties, sampled by horizon, that are important to soil survey. This includes participation in field soil survey activities during 2005 to sample six pedons in Allegany County and seven pedons in Ontario/Yates Counties resulting in 62 samples. Sub-samples were obtained, from the bulk NRCS samples for characterization, for both hyperspectral and a limited set of laboratory analyses at Cornell University. This sample and the more detailed soil chemical and physical properties to be analyzed by the national soil survey laboratory will be used to correlate hyperspectral reflectance with selected soil properties important to soil survey. In the coming year, we plan to conduct statistical analyses of the
spectral and laboratory data to develop predictive models of selected soil properties.
Impacts
Hyperspectral sensing of soil samples will provide a low-cost alternative to direct measurements of chemical and physical properties of soils where large numbers of observations are required to accurately characterize an area or where the cost of field survey and laboratory analysis is high.
Publications
- Hively, W.D., E.J. Neafsey, M. Havens, and S.D. DeGloria. 2005. Hyperspectral sensing of soil pedons for soil classification and survey. Proc. ASA-CSSA-SSSA Ann. Mtg. Am. Soc. Agronomy. Madison, WI (Abstract).
|
Progress 01/01/04 to 12/31/04
Outputs
During this reporting period, research focused on: (1) designing a spatial model to map foodsheds in New York State. Potential foodsheds will be mapped based on the physical and economic suitability of land in New York State for use in different classes of agricultural production and on the food demand of individual population centers. State Soil Geographic (STATSGO) data will need to be used. The source of climatic data has not yet been identified. The NYS method for calculating agriculture use-value for preferential tax assessment will serve as a method for estimating land use-value for different classes of agriculture. Total food need and total food demand for individual population centers will then be extrapolated based on population estimates from the Bureau of the Census. The suitability estimates, expected yields, use-value estimates, and food consumption estimates will be integrated in a GIS model to calculate the distance within which the food needs of
population centers could be supplied. (2) Anthropogenic soil sample collection was initiated to expand the soils represented in the hyperspectral soil analysis methodology assessment at Cornell University. Previously archived urban soil samples where inventoried to determine prior sample preparation methods for inclusion in assessment. Samples were also identified based on analytical methods and results of prior research use. This was to determine what archived samples had existing elemental content results. These soil samples are of particular interest to hyperspectral soil studies as they may provide a comparison of analytical methodologies. There are other potential archived samples with elemental data, but availability has not been secured.
Impacts
Increased accuracy and precision of predicted soil properties is required for effective use in farm- and watershed-scale management plans, biophysical process and land use suitability models, and decision support systems in rural and urban environments. Without knowing the spatial distribution of such properties, the environmental impacts of land use decisions cannot be assessed nor quantified.
Publications
- No publications reported this period
|
Progress 10/01/03 to 12/31/03
Outputs
(1) Soil survey updating, certification, and digital map finishing continued for several counties and selected watersheds in selected locations of New York State; (2) Familiarization with and laboratory testing of a field-portable spectroradiometer were conducted for anticipated acquisition of hyperspectral data of soil samples collected for National Cooperative Soil Survey investigations.
Impacts
Increased accuracy and precision of predicted soil properties is required for effective use in farm- and watershed-scale management plans, biophysical process and land use suitability models, and decision support systems in rural and urban environments. Without knowing the spatial distribution of such properties, the environmental impacts of land use decisions cannot be assessed nor quantified.
Publications
- No publications reported this period
|
|