Integrating Environmental and in situ Hyperspectral Remote Sensing Variables for Grass Nitrogen Estimation in Savannah Ecosystems

A. Ramoelo, M.A. Cho, R. Mathieu, A.K. Skidmore, M. Schlerf, I.M.A. Heitkonig, H.H.T. Prins

Research output: Contribution to conferenceConference paperAcademic

2 Citations (Scopus)

Abstract

Information about the distribution of grass nitrogen (N) concentration is crucial in understanding rangeland vitality and facilitates effective management of wildlife and livestock. A challenge in estimating grass N concentration using remote sensing in savannah ecosystems is that these areas are characterised by heterogeneity in edaphic, topographic and climatic factors. The objective is to test the utility of integrating environmental variables and in situ hyperspectral remote sensing variables for predicting grass N concentration along a land use gradient in the greater Kruger National Park. Data used include i) environmental variables, ii) measured grass N concentration and iii) in situ measured hyperspectral spectra. Non-linear partial least square regression was used. Results showed that several environmental variables were important for N estimation. Integrating environmental variables with in situ hyperspectral variables increased grass N estimation accuracy. The study demonstrated the importance of integrated modelling for savannah ecosystem state assessment.
Original languageEnglish
Publication statusPublished - 2011
Event34th ISRSE on The GEOSS Era : Towards Operational Environmental Monitoring, Sydney, Australia -
Duration: 10 Apr 201115 Apr 2011

Conference/symposium

Conference/symposium34th ISRSE on The GEOSS Era : Towards Operational Environmental Monitoring, Sydney, Australia
Period10/04/1115/04/11

Keywords

  • Foliar nitrogen concentration
  • Hyperspectral
  • Resources
  • Savannah ecosystem

Fingerprint

Dive into the research topics of 'Integrating Environmental and in situ Hyperspectral Remote Sensing Variables for Grass Nitrogen Estimation in Savannah Ecosystems'. Together they form a unique fingerprint.

Cite this