Rates of worldwide irrigation have increased 1-2% annually from around 1960 to feed the increasing global population. This extra consumption totalling 3000-4000 km³ is estimated to be 3-4% of precipitation on land. Our regional model predicts evapotranspiration of 60% of this additional water, dispersed by advection (>x2) but with double the residence time (x2) acting on dry environments where its greenhouse effect is maximised at lower humidity (x2). Given water causes more than 80% of the natural atmospheric global warming of 33 ^oC, this water vapour charges air with heat by (i) direct absorption of shortwave solar radiation (a_sw), (ii) from latent heat of evapotranspiration (e_t) and (iii) greenhouse interception of surface longwave radiation resisting heat flow to space (g). An estimated increase of 1% in atmospheric water vapour from this extra irrigation since 1960 could correspond to a global increase in temperature up to 0.25 ^oC, comparable to the warming estimated from increasing CO₂ since 1960. By comparison, the El Nino-La Nina cycles result in a global temperature variation approaching 1^oC, with atmospheric water vapour varying 4%. In global climate models (GCMs), water is assigned an indirect role, as a positive feedback to other heating. This model of forcing of regional warming (h=a_sw+e+g) with local heat export by outgoing longwave radiation (olr) and advection by latitude (ad) is being tested in Google Earth Engine from MODIS satellite data. The MOD16 algorithm for evapotranspiration (e_t) is generated from the Penman-Monteith equation for Australian irrigated areas confirming water vapour generation.This is supported by our data from a mobile Vaisala HMP110 T and RH probe mounted on a DJI Mavic 2 drone obtained in the Ord and Namoi River basins. If confirmed as a significant contribution to warming, specific credits would be justified for improved water use efficiency in climate management.