Mines

Permanent URI for this collection

http://41.63.25.24:4000/handle/123456789/82

Browse

Browsing Mines by Subject "Baroste Wetlands--Environmental conditions."

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Earth Observatory-based Assessment of Spatial and Temporal trends in Inundation Extent in the Barotse Wetland,Western Zambia
    (University of Zambia, 2017) Zimba, Musonda Henry
    Climate change, construction of the Mongu-Kalabo Road coupled with increased mining activities in the upper catchment of the Barotse Wetland could alter the inundation extent dynamics and impact the flood dependent ecosystem. There is very little information available in public domain on inundation extent dynamics of the wetland. This study investigated trends in inundation extent in the Barotse Wetland in Western Zambia. The study methods included statistical analysis of inundation extent time series generated with the Desert Flood Index from MODIS satellite imagery, in-situ hydrological and climatological time series, and land cover change assessment with Landsat imagery. Variations in inundation extent were analysed with the one-way Repeated Measure Analysis of Variance test. The Mann-Kendall trends test was used to characterise trends in inundation extent, hydrological and climatological time series. The Pearson r and the Paired t test were used to analyse correlations between hydrological and climatological time series. The Pettit test was used to test for homogeneity in the time series. Land cover change statistics were derived from the Maximum Likelihood algorithm classified maps which were validated with ground truthing data. The results show that across the period 2003 to 2013 there were significant variations in inundation extent with the F-Ratios (27.21 and 15.73) greater than the F-critical values (2.35 and 2.17) and p-values <0.05 in the ascending/peak and descending periods respectively. In the same period inundation extent showed a significant (p value <0.05) rising trend with a Mann-Kendall Z statistic of 1.87. A very strong correlation, with coefficient of determination (R2) of 0.91, between inundation extent and discharge was observed. From 1984 to 2015 the change in forest cover was observed at 9.78 percent with an annual rate of decrease of 0.32 percent. Overall, the Mann-Kendall trend test showed significant declining trends in mean annual discharge for the period from 1952 to 2003 at Senanga and 1952 to 2004 for Lukulu (Z statistics of -3.38 and -2.88, p-values <0.05 in both scenarios, at alpha level 0.05). For the period 1974 to 2012, the rainfall time series at Kabompo Station, upstream of the wetland catchment, showed a generally insignificant declining trend with Z statistic of -1.14 and p-value of 0.12 at alpha level 0.05. Rainfall and discharge showed a significant (p-value < 0.05) positive correlation with a coefficient of 0.36. Maximum temperature correlated negatively with discharge and inundation extent with correlation coefficients of -0.44 and -0.16 respectively. The study shows that trends in inundation extent follow trends in discharge. By inference, between 1952 and 2004 inundation extent in the Barotse Wetland was in downward trend as was the trend in discharge. Between 2003 to 2013 inundation extent trend was on the rise as was the trend in water level (from 2000 to 2011). Overall, the downward trend in discharge and reduction in forest cover could negatively affect inundation extent patterns and the Barotse wetland’s ecosystem as a whole. In view of the results of the study the Government of Zambia must ensure that deforestation in the wetland catchment is addressed and that no extensive alterations are made to the current wetland landscape as such actions have potential to alter the flow regimes and inundation extent patterns with consequences on the wetland ecosystem.