ANALYSIS OF OIL SPILL RISK USING SPACE-TIME PATTERN OF INCIDENTS IN THE NIGER DELTA, NIGERIA

Abstract

Oil spills are hazardous due to their impact on the people, the environment and economic wellbeing of the society. The study examined the space-time trend and pattern of oil spills in the Niger Delta (2006-2016). Data were sourced from the National Oil Spill Detection and Response Agency (NOSDRA). Mann-Kendal trend and Getis-Ord Gi* statistics were used to identify the spatial and temporal trend. A total of 6,487 incidents were identified, about 31% were recorded on Tuesday and Wednesday while the highest number (10%) of incidents occurred in May. The average nearest distance among events indicates spatial clustering, as the observed mean distance (0.29km) is significantly (Z-score -121.69, P < 0.01) smaller than expected mean distance (1.39km). Of the 8,900 total locations (grid size of 2.8km, 132 months), 941 (10.57%) contain at least one incident for at least one-time step interval. These 941 locations comprise 124,212 space-time bins of which 4,262 (3.43%) have incident counts greater than zero. These resulted in trend statistics of 10.61 (P<0.05), indicating that there is a statistically significant increase in point counts over time (monthly). Statistically significant event grids (62) were found to be hot spots. These hot spot clusters are spread across Bayelsa (4), Delta (2) and Rivers (3) States. The results gave a clear indication of the areas where over time and space oil spills are prominent. Thereby, identifying statistically significant space-time pattern of incidents which could inform mitigation measures and facilitate the implementation of adequate risk management and contingency planning.

Keywords

• oil exploration
• oil theft
• emerging pattern
• spatio-temporal
• risk analysis