Remote sensing applications for monitoring the spatial distribution of Bracken fern (Pteridium) in the Cathedral Peak : Drakensberg, South Africa.

Bracken fern (Pteridium) is an invasive plant that presents serious environmental, ecological and economic problems world over.A comprehensive analysis of the current spatial distribution of bracken fern weeds is therefore essential for providing applicable management approaches at both local and regional scales.The main aim of the study was to assess remote sensing applications for monitoring the spatial distribution of bracken fern weeds in Drakensberg, South Africa.The first objective of the study focused on reviewing the progress and challenges in the remote sensing of bracken fern.Traditional mapping methods have been reported inefficient in discriminating the temporal and spatial distribution of bracken fern at a regional scale.The use of commercial satellite data with high resolution have demonstrated potential in providing fine spectral and spatial resolution capabilities that are essential to offer precise and reliable data on the spatial distribution of invasive species.Challenges encountered in remote sensing of bracken fern include problem of similarity in spectral signatures of bracken and other vegetation species leading to low classification accuracy.The second objective of the study assessed the effectiveness of the freely available Landsat 8 OLI sensor in detecting and mapping bracken fern.In evaluating the performance of Landsat 8 OLI in discriminating bracken fern, Landsat 8 OLI results were compared to that of Worldview-2.The study tested the utility of (i) spectral bands; (ii) derived vegetation indices as well as (iii) the combination of spectral bands and vegetation indices based on discriminant analysis classification algorithm.After resampling the training and testing data and reclassifying several times (n= 100) based on the combined data sets, the overall accuracies for both Landsat 8 and WorldView-2 were tested for significant differences based on Mann-Whitney U test.Findings of this study revealed that the combination of the spectral bands and vegetation indices yielded the best overall classification accuracy of 80.08% and 87.80% for Landsat 8 OLI and WorldView-2, respectively.There were significant differences {U (100) = 569.5,z = -10.8242,p < 0.01} between the classification accuracies derived based on Landsat OLI 8 and those derived using WorldView-2.Although there were significant differences between Landsat and WorldView-2 accuracies, the magnitude of variation (9%) between the two sensors was within an acceptable range.In this regard, considering the acquisition costs and limited spatial coverage for WorldView-2 sensor, it is recommendable to shift towards the use of freely and readily available new generation broadband sensors, such as Landsat 8 OLI.