Title

Impact assessment of agricultural watersheds using geospatial data mining techniques: A conceptual framework

Document Type

Conference Proceeding

School

School of Science

RAS ID

14751

Comments

Originally published as Nallan, S.A., and Armstrong L.J. (2012). Impact assessment of agricultural watersheds using geospatial data mining techniques: A conceptual framework. Asian Federation for Information Technology in Agriculture, (AFITA) 2012, 8th Asian Conference for Information Technology in Agriculture, 2012, Taipei, Taiwan. Sep 3-6, 2012. Abstract available here.

Abstract

Application of novel geospatial data mining techniques to real world datasets in the field of agricultur e will enhance the understanding of unknown patterns in these datasets. These techniques have been used in drought management, vegetable quality classification, pest management and several other areas in agriculture and allied areas. One such area which requires the application of geospatial data mining analysis is watersheds. This paper attempts to analyze different data mining techniques with watersheds data to understand its spread and impact of its development. Government of India started developing a gricultural watersheds over the last four decades as a priority program for increasing water availability in the rainfed agricultural lands to sustain the livelihoods of the people with a socio -economic progress. Agricultural researchers and watershed dev elopment officials are extremely interested to assess the impact and identify the patterns in the current development and looking for the areas for further development with a proper understanding of local needs and environmental requirements. Due to chang e in the soil types, the water harvesting in different soils behaves in different ways resulting in improper recharge in some areas and high evaporation in some other areas. Geospatial data mining techniques such as association, clustering, classification and trend analysis applied on watersheds data will give proper understanding of the spread of watersheds and helps in visualization of key factors in the process. Since the watersheds are mainly affected by its location and upstream, downstream abnormali ties, it is very appropriate to use such techniques to understand the spatial distribution and actual impact. The open source geospatial and data mining software such as WEKA, PostGIS, jGRASS will be used to apply the data mining algorithms. An integrate d framework will be developed and tested with the available data in the light of available literature and location specific requirements.

Share

 
COinS