Skip to main content
ARS Home » Research » Publications at this Location » Publication #285906
Title: Mapping and spatiotemporal analysis tool for hydrological data: Spellmap

Author
item Guzman Jaimes, Jorge
item Moriasi, Daniel
item Starks, Patrick
item GOWDA, PRASANNA - Conservation Research Center
item CHU, MA - St Louis University
item Steiner, Jean

Submitted to: Environmental Modelling & Software
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/15/2013
Publication Date: 7/24/2013
Citation: Guzman Jaimes, J.A., Moriasi, D.N., Starks, P.J., Gowda, P.H., Chu, M.L., Steiner, J.L. 2013. Mapping and spatiotemporal analysis tool for hydrological data: Spellmap. Environmental Modelling & Software. 48:163-170. DOI:10.1016/j.envsoft.2013.06.014.

Interpretive Summary: Lack of data management and analyses tools is one of the major limitations to effectively evaluate and use large datasets of high-resolution atmospheric, surface, and subsurface observations. High spatial and temporal resolution datasets better represent the spatiotemporal variability of hydrological processes, and the associated fate and transport of sediments and solutes. Improper representation of the spatial patterns of the atmospheric or terrestrial processes can contribute to errors in models inputs. This is especially important in distributed models, which are used as cost-effective tools to assess the impact of climate and land use changes, land management, and conservation practices. In this study, a data management and analysis tool, SPELLmap, was developed by the USDA Agricultural Research Service to rapidly process, manipulate, analyze, and visualize large static and dynamic datasets taking into account spatial and temporal scales. The Fort Cobb Reservoir experimental watershed climate dataset (USDA-ARS MICRONET) was used to illustrate SPELLmap's capabilities including the effects of sampling data frequency and density, and data summation and separation in the spatial representation of climatic data. SPELLmap can also integrate grids and time series analysis for data interpolation, visualization, gridding, and pattern recognition. The ultimate goal of SPELLmap will be to assist scientists and engineers data preparation and analysis of spatially distributed water resources-related problems using models.

Technical Abstract: Lack of data management and analyses tools is one of the major limitations to effectively evaluate and use large datasets of high-resolution atmospheric, surface, and subsurface observations. High spatial and temporal resolution datasets better represent the spatiotemporal variability of hydrological processes, and the associated fate and transport phenomena. Data sampling frequency and density, data quality, and data aggregation and segregation drive the spatial pattern representation of the phenomena. Misrepresentation of the spatial patterns of the phenomena can contribute to uncertainty in models inputs. This is especially important in distributed models when assessing transport phenomena. A data management and analysis tool, SPELLmap, was developed by the USDA Agricultural Research Service using Delphi programming language to rapidly process, manipulate, analyze, and visualize large geo-referenced static and dynamic datasets while integrating the spatial and temporal domains. The Fort Cobb Reservoir experimental watershed climate dataset (USDA-ARS MICRONET) was used to illustrate SPELLmap's capabilities including the effects of sampling data frequency and density, and data aggregation and segregation in the spatial representation of climatic data. SPELLmap can be used to assess the trade-offs between computing time and resolution/network density. SPELLmap can also integrate raster and time series analysis for data interpolation, visualization, gridding, and pattern recognition. SPELLmap is an application that can be used to handle the ubiquitous nature of today’s hydrologic data and modeling. Future SPELLmap developments will integrate embedded and distributed database capabilities and parallel processing as well as incorporate multivariable statistic and pattern recognition functionalities.