Assessment of the Spatial Distribution of Wild and Cultivated Ensete ventricosum in Ethiopia Using Geospatial Tools
No Thumbnail Available
Date
2017-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Ensete ventricosum varieties are important for agricultural and economic developments of Ethiopia. However, little information is documented on the existing characteristics of ensete and distribution of different ensete verities across different agro-ecological zones. Hence, analyzing the spatial distribution of these species at spatial and temporal scale is of great importance for resource management and conservation planning. This study is attempted to identify, map and model the distribution of wild and cultivated Ensete ventricosum (ensete) with respective to land cover, climate, natural vegetation cover and agroecological zones. For this purpose, geospatial data analysis and MaxEnt modeling techniques were used to map the probabilistic distribution of both wild and cultivated verities. About 26 environmental factors were utilized as variables for distribution and modeling these includes 19 bioclimatic variables, DEM, LU/LC, vegetation, agro-ecology and soil type, and 192 cultivated and 20 wild ensete species records. Besides, Pearson correlation analyses were undertaken for 26 environmental variables to reduce highly correlated variables. The MaxEnt modeling has proven to be very effective at determining habitat use and species distributions for a variety of species and localities. The average test Area under curve (AUC) for the replicate runs was 0.842, and the standard deviation was, 0.046 and AUC 0.760 with standard deviation of 0.101 for cultivated ensete and wild ensete, respectively. Out of the determinants, the annual rainfall (15.7%) and LU/LC (53.7%) were the most important environmental variables that highly affected the distribution of cultivated and wild ensete, respectively. In addition, about 1.55% of the study area was covered by primarily hotpot areas of cultivated and 3.6% is for wild ensete. The probabilistic distribution of cultivated ensete is higher in the southern region, some part of Oromia region, and little areas of southern and eastern parts of Amhara region, whereas wild ensete is spatially highly distributed in Tigray and Benshangul Gumuz regions in addition to areas that cultivated ensete is distributed. In terms of agroecological zones, ensete is dominant to tropical sub-humid areas and tropical cool humid areas, typically at higher elevations (highlands), and areas having high rainfall and low temperature. Keywords: MaxEnt modeling, AUC, Ensete ventricosum, Geospatial data analysis
Description
Keywords
MaxEnt modeling, AUC, Ensete ventricosum, Geospatial data analysis