Sensitivity of EVI-based harmonic regression to temporal resolution in the lower Okavango Delta

Reducing the temporal resolution of satellite data degrades the ability to detect both short-term and long-term ecological cycles, with the impact varying by vegetation type and disturbance regime. The study found that while woodland-dominated areas, riparian vegetation, and frequently flooded zones retained better cycle characterization even at reduced temporal depths, grasslands, mixed savannas, and frequently burned areas were more sensitive to data gaps. Additionally, random sampling of time-series vectors increased the risk of misinterpreting system variability as change, with randomly selected vectors producing larger artifacts than systematically chosen ones. These findings underscore the importance of carefully selecting satellite time series not only for temporal depth but also for temporal spacing to avoid misleading interpretations in dynamic environments like the Okavango Delta.

Mishra, N. B., Crews, K. A., & Neuenschwander, A. L. (2012). Sensitivity of EVI-based harmonic regression to temporal resolution in the lower Okavango Delta. International Journal of Remote Sensing, 33(24), 7703–7726. https://doi.org/10.1080/01431161.2012.701348