Early detection of crop water and heat stress for effective crop management requires continuous and accurate monitoring of cropland photosynthesis activity. Satellite measurements can complement the restrictive coverage afforded by in-situ measurements and have the potential to facilitate the monitoring of cropland photosynthesis over a large spatial scale in a cost-effective manner. Traditionally, space-based monitoring of cropland photosynthetic activity, especially Light-use efficiency (LUE), has relied on empirical relationships between satellite spectral reflectance and ground climate and vegetation conditions. Space-borne retrievals of sun-induced chlorophyll fluorescence (SIF), an independent measurement, has shown to provide a more direct estimation of photosynthetic activity than traditional methods, and may further allow the inference of LUE.
This study has empirically explored the possibility of remotely monitoring large-scale LUE by calculating the ratio of photosynthetically active radiation (PAR) normalized SIF to the Enhanced Vegetation Index (EVI). We applied this calculation to demonstrate the spatial patterns and seasonal dynamics of LUE and its related measurements in response to land surface temperature (LST) across Australian rainfed croplands from 2007 to 2016. LST was used to provide an integrated measure of vegetation water and heat stress at the canopy level.
Our results showed that LUE tends to be higher in the geographical middle zones than in either the warmer northern or the cooler southern regions. Temporally, we found that there was a seasonal asymmetry of LUE and its related measurements in response to LST change throughout the winter crop-growing season. Statistical tests revealed that the optimum LST range for satellite-based LUE was 16.6–17.6 °C during August. The more LST exceeded this optimum, the more sensitive LUE was found to be. Pixels in August with optimum LST across the ten-year sampling period (Augusts of 2007–2016) were distributed in the southern-middle to middle zones of the Australian rainfed croplands.
Our results provide new opportunities for large-scale cropland heat and drought stress detection under a future warmer and drier climate and can also support remote analyses of crop photosynthetic activity over large spatial scales.
