Title

Geospatial data mining to explore watershed development in rainfed regions

Document Type

Conference Proceeding

School

School of Science

RAS ID

16700

Comments

Originally published as Nallan, S.A., Armstrong, L.J., Neuhaus, A., Croke, B., & Dunstan, N. (2013). Geospatial data mining to explore watershed development in rainfed regions. Paper presented at the EFITA-WCCA-CIGR Conference “Sustainable Agriculture through ICT Innovation”, Jun 23-27, 2013, Turin, Italy. Original article available here.

Abstract

Advances in the Information and Communication Technology (ICT) and availability of fine resolution remote sensing spatial data provides an opportunity to identify previously unknown and potentially useful patterns from the huge datasets. Application of geospatial data mining techniques has been used in various agricultural contexts such as drought management, vegetable quality classification, pest management and several other agriculture related areas. The recent development and implementation of watershed programs in rainfed regions of India has highlighted strategies that can be used to conserve water for irrigation of crops. Due to aberrant rainfall conditions in the rainfed regions and enormous changes in the landuse, the impact of the watershed development has greatly influenced the local hydrology. Any understanding of the impact of watershed development cannot be made through an examination of a individual watershed, but through the examination of all surrounding watersheds. This can be achieved using geospatial datasets and available novel data mining algorithms. For example, these techniques could provide an assessment of the effectiveness of watershed development using different land use patterns, cropping intensity, water availability, aquifer re-charge and positioning of different watersheds structures in a catchment. This paper attempts to explore the application of geospatial data mining techniques to watershed data sets. An evaluation and quantification of hydrological impacts of watershed development under varying climate and management scenarios using advanced techniques of geo-spatial data mining could be one means to improve the understanding of these impacts. This paper reports on the development of a robust data matrix of various parameters that affect the watershed and the application of various data mining algorithms. It is concluded that there is no single technique that can be used to assess the impact of watershed development.

Access Rights

free_to_read

Share

 
COinS