ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Volume IV-5/W2
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-5/W2, 103–110, 2019
https://doi.org/10.5194/isprs-annals-IV-5-W2-103-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., IV-5/W2, 103–110, 2019
https://doi.org/10.5194/isprs-annals-IV-5-W2-103-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  05 Dec 2019

05 Dec 2019

POTENTIAL HABITAT MODELING OF WATER HYACINTH IN LAKES OF NEPAL USING MAXENT ALGORITHM

S. S. Thakuri1, P. Shrestha1, M. Deuba1, P. Shah1, O. P. Bhandari1, and S. Shrestha2 S. S. Thakuri et al.
  • 1Department of Geomatics Engineering, Kathmandu University, Dhulikhel, Nepal
  • 2Photogrammetry and Remote Sensing Unit, Land Management Training Centre, Government of Nepal

Keywords: Maxent Algorithm, Representative Concentration Pathway, Invasive Alien Plant Species, Water Hyacinth

Abstract. Invasive Alien Plant Species are spreading outside of their natural geographic range. Water hyacinth (Eichhornia crassipes) is one of the most widely and rapidly spreading invasive species throughout the tropical and subtropical regions of Nepal. In the last decade, water hyacinth has become a chronic problem in many major lakes of Nepal which have affected the habitat aquatic plants and animals. Our study focuses on potential habitat modeling of Water hyacinth over the major lakes of Nepal using Maxent algorithm. Primary data used for modeling were 19 bioclimatic variables and Shuttle Radar Topography Model (SRTM) Digital Elevation Model (DEM). After preparation of the species distribution model, major lakes of Nepal were overlaid over the model to prepare potential invasive map. The performance and accuracy of potential habitat distribution model was evaluated using parameter Area under the Receiver Operating Characteristic Curve (AUC) which was within the range of 0.9–1. Validation of the model was done for the year 2015 with precision and recall, overall accuracy and F-measure and its values are 93% and 85%, 87% and 89% respectively. The model prepared for 2030 and 2050 shows the most suitable habitat for water hyacinth is in province 2 of Nepal and the moderately suitable habitat for this species is plain area of Province 4, 7 and 5. Similarly, the area of potential habitat has been increasing from current scenario to 2030 and 2050. From the potential invasion map, it can be observed that lakes in the Terai and Churey regions have the high risk of invasion of water hyacinth.