Assessment of vegetation dynamics in Upper East Region of Ghana based on wavelet multi-resolution analysis

摘要

Vegetation variation offers significant information for environmental planning, management, sustainability and prompts caution of ecosystem degradation, particularly for the semiarid regions. Normalized difference vegetation index (NDVI) discloses the coverage growth situation, biomass and photosynthesis strength of vegetation and land-cover alterations. Wavelet was used to decompose NDVI time series into subseries at various timescales. This study used a multi-resolution analysis in association with Mann-Kendall and Sen's Slope at 95% confidence interval to determine the trends in vegetation dynamics at the Upper East Region (UER) of Ghana. GIMMS NDVI3g time series was used to evaluate the performance of the vegetation at seasonal, interannual and intraannual timescale from 1982 to 2015. The results showed that the variability in NDVI in the region is annually significant. At the seasonal level, the whole surface area showed negative vegetation trend. In terms of the intraannual changes, 11.76% of the surface area showed critical patterns. At the interannual scale, results revealed that 4.40% of the surface area demonstrated significant patterns, while 95.60% indicated nonsignificant pattern. Overall, there was negative performance in the vegetation growth from 1982 to 2015. The 16.6% decrease in vegetation dynamics can be attributed to anthropogenic activities. The results from this study would benefit and provide helpful assistance to water resources managers, agricultural and ecological development officers for sustainable planning of UER.