A new approach for using time-series remote-sensing images to detect changes in vegetation cover and composition in drylands: a case study of eastern Kenya

Vegetation cover and composition are important aspects of the dryland environment because they provide livelihood to humans and also protect soil resources against erosion. Currently, scientists are advancing various techniques for detecting vegetation degradation in the drylands and the possibilities for its control. This study contributed through the testing of time-series mixed-effects modelling of the normalized difference vegetation index (NDVI) and rainfall relationship to trace the footprints of vegetation dynamics in the drylands. The approach aimed at providing guidelines for quick diagnosis of the changes in vegetation cover and composition to trigger necessary action. The mixed-effects technique used in this study is a novel regression approach for simultaneous modelling of the NDVI–rainfall relationship in different dominant vegetation types. Its time-series application with Advanced Very High Resolution Radiometer (AVHRR) and Moderate Resolution Imaging Spectroradiometer (MODIS) NDVI images between 1982 and 2008 was tested in eastern Kenya. The results show how the original dominant vegetation types had been converted to cereal croplands, open grasslands, or reduced to bare ground in a span of 27 years. In some places, it shows how the changes in vegetation composition resulted in the overall loss of vegetation cover. Field validation positively confirmed these observations; thus, indicating that the method was a promising tool for tracing vegetation dynamics in the drylands. In spite of its success, the method was found to be only useful in detecting changes in large areas with dominant vegetation types. The technique can therefore be recommended for regional analysis, and can be used as a first approximation to guide more detailed subsequent analysis.

[1]  Francis N. Gichuki,et al.  More People, Less Erosion: Environmental Recovery in Kenya , 1994 .

[2]  L. M. Haines,et al.  Influence of precipitation and species composition on phytomass of a semi‐arid African grassland , 2001 .

[3]  Edwin W. Pak,et al.  An extended AVHRR 8‐km NDVI dataset compatible with MODIS and SPOT vegetation NDVI data , 2005 .

[4]  S. Nicholson,et al.  A comparison of the vegetation response to rainfall in the Sahel and East Africa, using normalized difference vegetation index from NOAA AVHRR , 1990 .

[5]  R. Jaetzold,et al.  Natural conditions and farm management information , 1989 .

[6]  B. Minasny,et al.  Nonlinear mixed effect modelling for improved estimation of water retention and infiltration parameters , 2006 .

[7]  D. P. Groeneveld,et al.  Broadband vegetation index performance evaluated for a low‐cover environment , 2006 .

[8]  Gerard B. M. Heuvelink,et al.  About regression-kriging: From equations to case studies , 2007, Comput. Geosci..

[9]  M. Schaepman,et al.  Proxy global assessment of land degradation , 2008 .

[10]  C. R. Black,et al.  Water use in a Grevillea robusta–maize overstorey agroforestry system in semi-arid Kenya , 2003 .

[11]  B. Jama,et al.  Comparative growth performance of some multipurpose trees and shrubs grown at Machakos, Kenya , 1989, Agroforestry Systems.

[12]  J. Ware,et al.  Random-effects models for longitudinal data. , 1982, Biometrics.

[13]  W. D. Plessis,et al.  Linear regression relationships between NDVI, vegetation and rainfall in Etosha National Park, Namibia , 1999 .

[14]  Hanoch Lavee,et al.  Vegetation response to possible scenarios of rainfall variations along a Mediterranean-extreme arid climatic transect. , 2000 .

[15]  C. Omuto,et al.  Combining pedometrics, remote sensing and field observations for assessing soil loss in challenging drylands: A case study of Northwestern Somalia , 2009 .

[16]  Tal Svoray,et al.  Assessing land-cover change and degradation in the Central Asian deserts using satellite image processing and geostatistical methods , 2008 .

[17]  D. Tongway,et al.  Monitoring ecological indicators of rangeland functional integrity and their relation to biodiversity at local to regional scales , 2004 .

[18]  P. Carmanchahi,et al.  Past, present and potential uses of native flora and wildlife of the Monte Desert , 2009 .

[19]  Ranga B. Myneni,et al.  Effect of orbital drift and sensor changes on the time series of AVHRR vegetation index data , 2000, IEEE Trans. Geosci. Remote. Sens..

[20]  Francis N. Gichuki,et al.  More People, Less Erosion: Environmental Recovery in Kenya , 1995 .

[21]  J. Hogg.  Quantitative remote sensing of land surfaces , 2004 .

[22]  C. Tucker,et al.  Desertification in the Sahel: a reinterpretation of a reinterpretation , 2007 .

[23]  J. Imbernon.  A comparison of the driving forces behind deforestation in the Peruvian and the Brazilian Amazon. , 1999 .

[24]  Jennifer Small,et al.  Can human-induced land degradation be distinguished from the effects of rainfall variability? A case study in South Africa , 2007 .

[25]  Yichun Xie,et al.  Remote sensing imagery in vegetation mapping: a review , 2008 .

[26]  J. Evans,et al.  Discrimination between climate and human-induced dryland degradation. , 2004 .

[27]  Herman Eerens,et al.  Empirical regression models using NDVI, rainfall and temperature data for the early prediction of wheat grain yields in Morocco , 2008, Int. J. Appl. Earth Obs. Geoinformation.

[28]  Margriet Nakken,et al.  Wavelet analysis of rainfall-runoff variability isolating climatic from anthropogenic patterns , 1999, Environ. Model. Softw..

[29]  F. Ojany,et al.  Kenya, a Study in Physical and Human Geography , 1975 .

[30]  H. Dregne.  Land Degradation in the Drylands , 2002 .

[31]  C. Omuto,et al.  Assessment of soil physical degradation in Eastern Kenya by use of a sequential soil testing protocol , 2008 .

[32]  V. Carey,et al.  Mixed-Effects Models in S and S-Plus , 2001 .