The Role of Irrigation Expansion on Historical Climate Change: Insights From CMIP6

Abstract

To produce food for a growing world population, irrigated areas have increased from approximately 0.63 million km2 of land in 1900 to 3.1 million km2 of land in 2005. Despite this massive expansion, irrigation is still overlooked in most state‐of‐the‐art Earth system models (ESMs) involved in the Coupled Model Intercomparison Project phase 6 (CMIP6). To our knowledge, only three CMIP6 models represent irrigation activities: CESM2, GISS‐E2‐1‐G, and NorESM2‐LM. Here, we investigate the role of irrigation on historical climate at global and regional scales by analyzing trends of key surface climate variables in CMIP6 simulations during 1900–2014. The three models including irrigation show distinct behavior from the 15 models without irrigation over intensively irrigated areas (i.e., >50% of grid cell area is equipped by irrigation): both annual (months that correspond to monthly air temperature higher than 274 K) mean latent heat flux (LHF) and soil moisture increase over time, in contrast to the models without irrigation that show no trend or even a negative trend. The positive LHF trend over intensively irrigated areas in the irrigation‐on models is consistent with three satellite‐based LHF products. While annual (considering the warmest month in a year) warming trends are found in these regions for most of the no‐irrigation models, the increase in LHF induces a cooling trend for the models with irrigation, which was not confirmed by observational air temperature data sets. These findings, supported by satellite data, indicate the importance of improved representation of land management in the next generation of ESM.

The biogeophysical effects of idealized land cover and land management changes in Earth System Models

“…Furthermore, EC-EARTH is currently not a viable model for a study of the biogeophysical effects of irrigation, as water fluxes from land are not communicated to the atmosphere. This limitation is worth addressing as the implementation of irrigation in ESMs has been shown to make them more realistic over regions of intense irrigation (Al-Yaari et al, 2022). Regarding land cover change, all ESMs still struggle to replicate observed patterns in energy fluxes (Fig.…”

“…Among land management change options, irrigation has a clear regional cooling effect, especially during warm episodes (Hirsch et al, 2017;Thiery et al, , 2020Chen and Dirmeyer, 2019;Gormley-Gallagher et al, 2022;Mishra et al, 2020). Despite its recognized imprint on local climate, only a few ESMs simulate irrigation explicitly, with only three ESMs including irrigation in the CMIP6 simulations (Al-Yaari et al, 2022). Wood harvesting has mostly been studied for its biogeochemical effects, while the analysis of the biogeophysical effects is still lacking in studies using ESMs.…”

Heatwave Characteristics in the Recent Climate and at Different Global Warming Levels: A Multimodel Analysis at the Global Scale

“…The summertime warm biases have been explored, especially over the middle latitudes, in numerous analyses. Previous studies attributed these warm biases to deficiencies in cloud representation and subsequent overestimation of downwelling shortwave radiation, underestimated evaporation and precipitation, and misrepresented surface layer SM and large‐scale atmospheric circulation (Al‐Yaari et al., 2019, 2022; Cheruy et al., 2014; Lin et al., 2017; Mueller & Seneviratne, 2012). As shown in Figure 2, the models tend to overestimate HWM over most of the considered regions.…”

Unsustainable groundwater-fed irrigation drives deltaic water scarcity and compound flooding

“…The vast majority of models used in the Coupled Model Intercomparison Project phase 6 (CMIP6) neglect irrigation activities. Such negligence is due to difficulties in properly representing the amount, timing, techniques and efficiency of irrigation practices, which could lead to uncertain future climate projections (39). Neglecting GWI in climate models results in overestimated runoff rates, which can be exacerbated by increased regional precipitation and CO2induced stomata closure (40).…”