Woody Cover Mapping in the Kruger National Park using Sentinel-1 time series and LiDAR data

This data repository presents a workflow to derive woody cover information for the Kruger National Park, South Africa, from freely available Sentinel-1 C-Band time series and LiDAR data (modified from Smit et al. 2016) using machine learning (MLR and Ranger in R). The methodology is described in following publication:

Urban, M., K. Heckel, C. Berger, P. Schratz, I.P.J. Smit, T. Strydom, J. Baade & C. Schmullius (2020): Woody Cover Mapping in the Savanna Ecosystem of the Kruger National Park Using Sentinel-1 C-Band Time Series Data. Koedoe.

In order to derive woody cover percentage information, download all files into one folder and run the R-Files consecutively from 01_ to 04_. Follow the instruction within each of the R-Files, which are written as comments in the programming code.

The data repository consist of the following files:

R-Files:

1. Script 1: 01_MLR_tune_spatial_final

2. Script 2: 02_MLR_cross_validation_spatial_final

3. Script 3: 03_MLR_RANGER_train_final

4. Script 4: 04_MLR_prediction_woody_cover_final

Training dataset - ENVI FILE (layerstack of Sentinel-1 VH and VV backscatter between 2016 and 2017 and the woody cover reference derived from the LiDAR data) :

1. S1_A_VH_VV_16_17_lidar

Data for prediction - ENVI FILES (3 example regions in the Kruger National Park):

1. S1_A_VH_VV_16_17_subset_example_Letaba_Rest_Camp

2. S1_A_VH_VV_16_17_subset_example_Lower_Sabie

3. S1_A_VH_VV_16_17_subset_example_Pafuri

Final woody cover maps of the Kruger National Park:

1. xx_woody_cover_map_final.rar (contains final maps in 10m, 30m, 50m and 100m spatial resolution as .tif and a QGIS project)

References:

Smit, I.P.J., Asner, G.P., Govender, N., Vaughn, N.R. & Wilgen, B.W. van, 2016, ‘An examination of the potential efficacy of high-intensity fires for reversing woody encroachment in savannas’, Journal of Applied Ecology, 53(5), 1623–1633.