Evaluation of three new surface irrigation parameterizations in the WRF-ARW v3.8.1 model: The Po Valley (Italy) case study

Irrigation is a method of land management that can affect the local climate. Recent literature shows that it affects mostly the near-surface variables and it is associated with an irrigation cooling effect. However, there is no common parameterization that also accounts for a realistic water amount, and this factor could ascribe one cause to the different impacts found in previous studies. This work aims to introduce three new surface irrigation parameterizations within the WRF-ARW model (v3.8.1) that consider different evaporative processes. The parameterizations are tested on one of the regions where global studies disagree on the signal of irrigation: the Mediterranean area and in particular the Po Valley. Three sets of experiments are performed using the same irrigation water amount of 5.7 mm d-1, derived from Eurostat data. Two complementary validations are performed for July 2015: monthly mean, minimum, and maximum temperature with ground stations and potential evapotranspiration with the MODIS product. All tests show that for both mean and maximum temperature, as well as potential evapotranspiration simulated fields approximate observation-based values better when using the irrigation parameterizations. This study addresses the sensitivity of the results to human-decision assumptions of the parameterizations: start time, length, and frequency. The main impact of irrigation on surface variables such as soil moisture is due to the parameterization choice itself affecting evaporation, rather than the timing. Moreover, on average, the atmosphere and soil variables are not very sensitive to the parameterization assumptions for realistic timing and length.