PlumX Metrics
Embed PlumX Metrics

Determining the onset of autumn grass senescence in subtropical sour-veld grasslands using remote sensing proxies and the breakpoint approach

Ecological Informatics, ISSN: 1574-9541, Vol: 69, Page: 101651
2022
  • 10
    Citations
  • 0
    Usage
  • 5
    Captures
  • 0
    Mentions
  • 14
    Social Media
Metric Options:   Counts1 Year3 Year

Metrics Details

  • Citations
    10
    • Citation Indexes
      10
  • Captures
    5
  • Social Media
    14
    • Shares, Likes & Comments
      14
      • Facebook
        14

Article Description

Information on the onset of autumn grass senescence in subtropical grasslands is essential for ascertaining the duration of poor forage quality. It is well-established that during senescence, grass leaves lose their nutrients to the rooting systems; which affects the quality and quantity of forage resources. The timing of the onset of autumn grass senescence is critical in determining the potential lifespan through which the provision of quality forage can be sustained in a grazing area. However, objective and robust methods for estimating the onset of autumn grass senescence at a rangeland scale are limited. Hence, this study sought to characterize the onset of autumn senescence in mesic subtropical sour-veld grasslands using remotely sensed data. Ten monthly vegetation indices were generated from the Sentinel 2 data and used as proxies to explain the onset of autumn grass senescence. The performance of the proxies was validated using the corresponding field-measured monthly grass chlorophylls. Results showed that the Chlorophyll Red-Edge (CHL-RED-EDGE) and the Normalized Difference Red Edge Index (NDVI 705 ) were the most important proxies for characterizing the autumn grass senescence. In addition, monthly (i.e., January to June) mean values of the two best proxies were fitted in a piecewise linear regression model with a breakpoint approach to determine the start of the autumn grass senescence. The first proxy (i.e., NDVI 705 ) predicted that the grass in the study area starts senescing on day number ± 98 of the year (R 2  = 0.97, RMSE = 0.024) while the second (i.e., CHL-RED-EDGE) suggested day number ± 106 of the year (R 2  = 0.96, RMSE = 0.052). Overall, this study demonstrated the value of remote sensing proxies in estimating the autumn grass senescence and in determining its onset. These results provide a basis for understanding the impact of autumn senescence on foraging resource provision in rangeland ecosystems.

Bibliographic Details

Provide Feedback

Have ideas for a new metric? Would you like to see something else here?Let us know