A characterisation of the terrestrial phenology of Bushbuckridge, South Africa, using MISR-HR

Abstract

This study documents (i) the seasonal cycles of vegetation canopies in the Bushbuckridge local municipality (BBR) as well as (ii) their sensitivities to non-seasonal perturbations during the period (2000 to 2018). The Fraction of Absorbed Photosynthetically Active Radiation (FAPAR) product computed by the Multi-angle Imaging SpectroRadiometer-High Resolution (MISR-HR) processing system was used to characterise the phenology of vegetation. This product exploits the measurements collected by NASA's Multi-angle Imaging SpectroRadiometer (MISR) instrument, which has been operational since 24 February 2000. The BBR is predominantly characterised by savanna vegetation, which is critically important in terms of biodiversity and for the sustainable socio-economic development of the region. The performance of parametric double S-shaped models (Gaussian, hyperbolic tangent, logistic and sine) to characterise the vegetation canopy's seasonal cycle in spite of the presence of missing values and noise in the FAPAR data was explored. Simulated FAPAR and in situ Gross Primary Productivity (GPP) measurements derived from the Skukuza Eddy Covariance flux tower were integrated over vegetative seasons and compared. The phenology models were all capable of fitting the FAPAR record when the observations were dense enough during the growth and senescence phases of the vegetative season, but some models turned out to be more sensitive than others to the presence of outliers or long data gaps. It was also shown that integrated FAPAR adequately tracks GPP in this environment as long as GPP varies in the range from 0 to 5 gC m-2 d -1 , beyond which this relation tends to saturate. This study investigated the sensitivity of vegetation productivity and precipitation anomalies, as reported by the Climate Hazards Group InfraRed Precipitations with Stations (CHIRPS), to historical ENSO periods occurring within the last two decades. These teleconnections were found to differ between five sites within the BBR, characterised by different vegetation types, underlying geology and protection status. The sensitivities of vegetation phenology and precipitation were specifically related to varying intensities and sequences of El Niño – Southern Oscillation (ENSO) events (i.e., El Niño and La Niña) at the scale of ecosystems. This study found FAPAR to be better correlated with ENSO than the CHIRPS precipitations. The productivity of vegetation observed in these ecosystems shows that they may be quite sensitive to the particular sequence and relative strengths of the El Niño and La Niña events, but the severity of the impact is related to the physical properties of the site. Regional natural resource management should acknowledge significant differences in the response of vegetation to ENSO events. This, in turn, might inform risk mitigation and adaptation strategies at the level of the BBR municipality.