System and process for integrative computational soil mapping
Inventors
Fuentes Ponce, Bryan Andre • Owens, Phillip Ray • Dorantes, Minerva Justine
Assignees
US Department of Agriculture USDA • University of Arkansas at Little Rock • United States Department of the Army
Publication Number
US-11676375-B2
Publication Date
2023-06-13
Expiration Date
2041-08-30
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
An integrative computational soil mapping system and process that reduces the required number of soil property measurements without jeopardizing the statistical precision of the resulting digital soil maps. The integrative computational soil mapping system and process saves monetary resources and time by reducing the number of soil property measurements required to produce digital soil maps and by offering soil sample locations which capture the maximum amount of representativeness of the soil characteristics in a determined area. In addition, the inventive system and process are integrative computational soil mapping that utilize algorithms based on state-of-the-art computational statistics and machine learning methods for the production of digital soil property maps and also provides soil sampling locations to collect new soil property measurements. These soil property measurements can be used to update and potentially improve previous versions of digital soil property maps, produced by the computational process.
Core Innovation
The invention provides an integrative computational soil mapping system and process that produces digital maps of soil properties using a minimum required number of soil sample measurements without compromising the statistical precision of the results. The inventive system reduces the number of soil property measurements needed and offers soil sample locations that maximize the representativeness of soil characteristics in a given area. It incorporates computational statistics and machine learning methods to produce digital soil property maps and to guide the collection of new soil property measurements used to update and improve previous digital maps.
The problem being addressed is that conventional soil surveys are limited by coarse resolution, discontinuous soil property information, and expensive fieldwork and laboratory analysis, which constrain the production of detailed soil maps. Although digital soil mapping methods exist, they often require a vast number of soil samples to achieve precise results, which is not always feasible due to budget, time, and human effort limitations. The invention provides a cost- and time-effective solution that meets the demand for precise, fine-resolution soil property information in data-limited scenarios, supporting precision agriculture and environmental resource management.
Claims Coverage
The patent includes two independent claims directed to an integrative computational soil mapping system and a corresponding process. These claims encompass inventive features related to data handling, environmental layer analysis, clustering, soil property assignment, spatial modeling, and digital soil map generation.
System for integrative computational soil mapping
A computer system with software that receives soil data for an area of interest; automatically or semi-automatically generates environmental layers related to cluster categories; selects environmental layers capturing variability; performs data dimensionality reduction using Self Organizing Map technique; performs cluster analysis for each cluster category; determines an optimum number of clustering groups; generates generic soil-landscape classes from spatial interactions of clustering groups; assigns soil property measurements to each class; generates digital soil maps of the soil property measurements; outputs the maps; and performs spatial modeling using weighted averaging based on likelihoods.
Cluster analysis and environmental data processing
Grouping environmental layers according to cluster categories such as climate, vegetation, topography, or parent material; clustering environmental data to find naturally occurring groups; calculating statistical distributions of layers per cluster using kernel density estimates and empirical distribution functions; predicting these distributions across the area using locally weighted polynomial regression; and generating group-likelihood maps representing the likelihood of occurrence of each clustering group spatially.
Soil sample location assignment and class-likelihood mapping
Automatically generating a generic soil-landscape class map by intersecting clusters across categories; averaging group-likelihood maps to create class-likelihood maps; assigning soil sample locations corresponding to the highest likelihood of occurrence for each generic soil-landscape class; and associating soil property measurements from these locations to the corresponding classes.
Spatial modeling of soil property measurements
Organizing class-likelihood maps into matrices; identifying for each spatial element the two elements with the highest likelihood occurrences; retrieving corresponding soil property measurements; and calculating weighted averages of soil properties using likelihood values as weights to generate the final digital soil map.
The claims collectively cover a computer-implemented system and process that integrate environmental data analysis, clustering, soil sample location optimization, and statistical spatial modeling to produce precise digital soil property maps with reduced soil sampling.
Stated Advantages
Reduces the required number of soil property measurements needed to produce digital soil maps without compromising statistical precision.
Saves monetary resources and time by optimizing soil sampling locations to capture maximum representativeness of soil characteristics.
Produces precise, fine-resolution, continuous predictions of soil properties even in data-limited scenarios.
Supports precision agriculture by integrating digital soil maps into equipment guidance, variable rate fertilizer applications, and irrigation planning.
Supports integrative environmental resource management and sustainable ecosystem management by facilitating identification of soil amendment and soil health focal points.
Cost- and time-effective solution that leverages existing soil measurements and widely available environmental data.
Documented Applications
Use in precision agriculture routines such as equipment guidance, delineation of management zones, variable rate fertilizer application, carbon storage potential estimation, and irrigation planning.
Support for integrative environmental resource management and sustainable ecosystem management by providing spatially precise soil property estimates to identify focal points for soil amendments and soil health concerns.
Interested in licensing this patent?