Termite mounds morphometry in predicting groundwater potentiality using geospatial technology

Termite mounds are conspicuous long lasting landscape features in many tropical to sub-tropical regions of the world. They provide several ecological and economic benefits to mankind amongst which is the classical believe of them to be good indicators of groundwater. This hypothesis is however, yet to be scientifically substantiated. The aim of this study therefore, is to develop a groundwater potential model in Geographical Information System (GIS) and evaluate the prospect of termite mounds in predicting suitable zones for groundwater development. To achieve this, a ground-based survey, remotely sensed data and GIS techniques were employed. From field survey, termite mounds were mapped and their structural characteristics recorded along 68 road transects covering 156 km² of the study area. The effects of five (5) environmental factors (elevation, land use/land cover, geology, drainage and static water level) on the distribution of termite mounds as well as their structure, mortality rate and species diversity formed a knowledge-base for multi-criteria evaluation of suitable sites for termite nesting. Thereafter, twelve (12) groundwater conditioning factors (GCFs) (geology, drainage density, lineament density, lineament intersection density, land use/land cover, topographic wetness index (TWI), normalized difference vegetation index (NDVI), slope, elevation, plan curvature, static water level and groundwater level fluctuation) were passed through a feature selection filter (Correlation-based feature selection using the best first algorithm) to select the optimum groundwater control factors (GCFs) for groundwater prediction in the study area. To assess the productivity potentials of aquifers beneath termite mounds, forty (40) electrical resistivity soundings using the Vertical Electrical Sounding (VES) method were conducted and an additional twenty eight (28) VES conducted in areas adjacent to termite mounds for comparison of potentials. The result produced two (2) models; termite mounds site suitability model (TMSM) and groundwater potential prediction model (GPPM) with validation accuracy using the area under ROC of 74.2% and 86.5% respectively. For termite mounds site suitability, the result revealed that moderate to low elevation, rock cover types that are more susceptible to weathering, cultivated areas and shallow water table are factor classes that influence the distribution of termite mounds. Frequency Ratio (FR) and Spearman’s rank correlation applied to find relationships between termite mounds and the optimum GCFs revealed a consistent agreement that tall termite mounds (≥1.8m) and cathedral designed termite mounds are useful in locating groundwater prospective zones. The mean weights derived from electrical resistivity soundings also revealed that tall termite mounds (>2m), cathedral designed mounds and in addition, mounds built by the genus Nasutitermes showed greater aquifer productivity potential than other types of termite mounds. This study provides an exposition to the scientific rational for using termite mounds as bio- indicators of groundwater and the specific mounds types to adopt as guide