Spatial biodiversity modeling using high-performance computing cluster: A case study to access biological richness in Indian landscape. Issue 18 (8th November 2021)

• Record Type: Journal Article
• Authors: Singh, Hariom; Garg, R. D.; Karnatak, Harish C.; Roy, Arijit
• Is Part Of: Geocarto international. Volume 36:Issue 18(2021)
• Page Start: 2023
• Page End: 2043
• Publication Date: 2021-11-08
• Subjects: Distributed GIService -- Message passing interface -- Cluster computing -- Biodiversity modeling -- Participatory GIS; Remote sensing -- Periodicals; Geographic information systems -- Periodicals; Geology -- Periodicals; Cartography -- Periodicals; 621.3678
• DOI: 10.1080/10106049.2019.1678679
• Languages: English
• ISSNs: 1010-6049

Abstract

Abstract: The parallel processing and distributed GIServices provide an efficient approach to address the geocomputation challenges in biodiversity modeling. Using the widely applied Spatial Biodiversity Model (SBM) as an illustration, this study demonstrates parallelization of the spatial landscape algorithms based on Message Passing Interface (MPI) in cluster computing. The geocomputation based on MPI is performed to characterize the spatial distribution of Biological Richness (BR) for Indian landscape using developed high-performance cluster computing-based model named as SBM-HPC. In performance analysis, the execution time is reduced by 56.42%–81.41% (or the speedups of 2.29–5.38) using the parallel and cluster computing environment. Also, the spatial landscape algorithms of the model are extended to integrate large-scale geodata from online map services archives using distributed GIServices. To validate BR map, the phytosociological data is collected using participatory GIS approach. Furthermore, regression analysis between derived BR map and Shannon-Wiener index (Hˈ) represents high correlation coefficient R 2 values. Highlights: Development of spatial biodiversity model using parallel computing on the cluster. Geocomputation of spatial landscape indices using large-scale geospatial datasets. Distributed GIService integration in model to compute distributed data archives. Prediction of biological richness pattern and validation using participatory GIS. Characterize correlations between biological richness and bioclimatic patterns.