Answers are grounded in this paper and cite their sources.
Agrivoltaic systems change the radiation, wind, soil temperature and rainfall that reach a crop. Because each field measurement is tied to one site, the resulting change in water use across designs, crops and climates is usually estimated with models. This review describes how those models calculate water. Scopus and IEEE Xplore were searched for publications from 2011 to the first half of 2026, with complementary searches in August 2026. The review includes 39 studies whose water values are calculated rather than measured, and each reported value was traced back to the equation that produced it. In 24 studies, evaporative demand is calculated from weather and scaled to the crop; in 10, evapotranspiration is obtained as one term of a balance of energy, vapor or soil water; and 5 report a water value without modelling a water process. Because these calculations return different quantities, a reported percentage change in water use may refer to atmospheric demand, crop demand or actual evapotranspiration. The photovoltaic (PV) array can enter the calculation through seven routes: shading, longwave radiation from the panels, wind and surface roughness, soil temperature, stomatal response, air temperature and humidity and rainfall interception. Shading is represented in 33 of the 34 studies that model a water process, and each of the other routes in only one to eight. Three further choices change the reported value: where shading is applied in the calculation, whether soil water stress is included and whether position beneath the array is resolved. Two calculated water values for two similar PV installations are not comparable unless the equation and these three choices match. This review sets out the equations, the points at which a PV array can act on them and the choices behind each reported value, so that a calculation can be chosen for the terms it carries rather than inherited with a parent model. The counts describe this corpus and are conditional on the search strategy. What remains open is how these calculations perform against measurements, above all in dry regions, where water rather than light limits production.
No takes yet. Share an insight, caveat, or question.
Jain et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: