International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Volume XLII-3/W2
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W2, 157–162, 2017
https://doi.org/10.5194/isprs-archives-XLII-3-W2-157-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W2, 157–162, 2017
https://doi.org/10.5194/isprs-archives-XLII-3-W2-157-2017
© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 4.0 License.

  16 Nov 2017

16 Nov 2017

MODELLING CARRYING CAPACITY FOR THE THANDA PRIVATE GAME RESERVE, SOUTH AFRICA USING LANDSAT 8 MULTISPECTRAL DATA

Z. Oumar1, J. O. Botha2, E. Adam1, and C. Adjorlolo3 Z. Oumar et al.
  • 1University of Witwatersrand, School of Geography, Archaeology and Environmental Studies, Private Bag 3, Witwatersrand 2050, South Africa
  • 2Department of Agriculture and Rural Development, Private Bag X9059, Pietermaritzburg 3200, South Africa
  • 3South African National Space Agency (SANSA), SANSA Earth Observation, P.O. Box 484, Silverton 0127, South Africa

Keywords: Carrying capacity, Landsat 8, PLS regression, Broadband indices

Abstract. Rangelands which consist of grasslands, shrublands and savannahs are used by wildlife for habitat and are the main source of forage for livestock. The assessment and monitoring of rangeland condition is one of the most important factors for rangeland scientists in order to calculate the carrying capacity of livestock with consideration for coexisting wildlife. This study assessed the potential of Landsat 8 multispectral bands and broadband vegetation indices to model woody vegetation parameters such as tree equivalents (TE) and total leaf mass (LMASS) for the Thanda Private Game Reserve using partial least squares regression (PLSR). The PLSR model predicted TE with an R2 value of 0.76 and a root mean square error (RMSE) of 1411 TE/ha using an independent test dataset. LMASS was predicted with an R2 value of 0.67 and a RMSE of 853 kg/ha on an independent test dataset. The predictive models were then inverted to map TE and LMASS over the study area. The modelled TE and LMASS layers were integrated with conventional grazing and browse capacity models to map carrying capacity for the Game Reserve. The study indicates the potential of Landsat 8 multispectral data in carrying capacity modelling. The result is significant for rangeland monitoring in Southern Africa using remote sensing technologies.