<p>With a changing climate, it is becoming increasingly critical to understand vegetation responses to limiting environmental factors. Here, we investigate the spatial and temporal patterns of light and water limitation on photosynthesis using an observational framework. Our study is unique in characterizing the nonlinear relationships between photosynthesis and water and light, acknowledging approximately two regime behavior (no limitation and varying degree of limitation). It is also unique in using an observational framework instead of using model-derived photosynthesis properties. We combine data from three different satellite sensors, i.e., solar-induced chlorophyll fluorescence (SIF) from TROPOMI, surface soil moisture from SMAP, and above-ground canopy density from MODIS. We find both single-regime and two-regime models describe SIF sensitivity to soil moisture and photosynthetically active radiation (PAR) across the globe. The distribution and strength of soil moisture limitation on SIF are mapped in the water-limited environments while distribution and strength of PAR limitations are mapped in the energy-limited environments. Two-regime behavior is detected in 73 % of the cases for water limitation on photosynthesis, while two-regime detection is much lower at 36 % for light limitation on photosynthesis. SIF sensitivity to PAR strongly increases along moisture gradients, reflecting mesic vegetation’s adaptation to making rapid usage of incoming light availability on the weekly timescales. The transition point detected between the two regimes is connected to soil type and mean annual precipitation for the SIF-soil moisture relationship and for the SIF-PAR relationship. These thresholds have therefore an explicit relation to properties of the landscape, although they may also be related to finer details of the vegetation and soil interactions not resolved by the spatial scales here. The simple functions and thresholds are emergent behaviors capturing the interaction of many processes. The observational thresholds and strength of coupling can be used as benchmark information for land surface and Earth system models, especially those that characterize gross primary production mechanisms and vegetation dynamics.</p>