Agricultural adaptation to a changing climate requires frequent, accessible, and reliable information on water availability at spatio-temporal scales relevant to end users. In response to this need, we have developed an approach that evaluates agricultural drought conditions while minimising data and processing costs. This is achieved through an approximation of the land surface temperature–vegetation index (LST-VI) population, which is used in the calculation of the vegetation temperature condition index (VTCI), and the implementation of non-linear edge finding, which more accurately defines the warm and cold boundaries of the population. This 'fast' approach to defining the LST-VI population equally applies to similar vegetation-temperature-based approaches to drought monitoring and is critical to enabling such approaches to be used with higher resolution thermal datasets at scale due to the reduction in run-time and memory requirements it enables. We tested the approach in Taranaki, New Zealand, using the full Moderate Resolution Imaging Spectroradiometer (MODIS) LST and surface reflectance record available through Google Earth Engine. We assessed three vegetation indices and evaluated fast-VTCI outputs against data from 10 ground-based soil moisture monitoring stations. Statistically significant correlations (Pearson’s and Spearman’s R=0.21–0.88, p<0.01) demonstrate that fast‑VTCI reliably captures spatial and temporal variability in field conditions, achieving performance equivalent to the traditional VTCI method. The strength of the observed correlations varies spatially and temporally for both indices. Overall, fast-VTCI substantially lowers computational costs, enabling efficient, operational drought monitoring across regional to national scales with spatial resolutions that support informed agricultural decision-making.