Tuning an improved irrigation scheme inside ORCHIDEE land surface model and assessing its sensitivity over land surface hydrology and energy budget

Water used to irrigate accounts for 70% of total water withdrawal and 90% of water consumption worldwide. As a result, irrigation has a strong impact on water and energy budgets, on the associated biogeochemical cycles, and on local and regional climate. Furthermore, water volume for irrigation is projected to increase, due to population growth and climate change. This has encouraged the inclusion of irrigation in an increasing number of land surface models (LSM), which represent the continental branch of the water cycle in Earth System Models. To this end, three key aspects of irrigation must be described: when to irrigate (timing), how to irrigate (irrigation method), and how much to irrigate (water amount).

We present a new irrigation scheme for the ORCHIDEE land surface model, developed to account for flood and drip irrigation techniques. In grid cells with irrigated areas, the water demand is deduced from the soil moisture deficit in the crop and grass soil column, i.e. is partially controlled by soil parameters. The soil column contains both irrigated and rainfed crops, but the fraction equipped for irrigation limits the water demand. The deficit is the difference between the actual soil moisture in the root zone and a soil moisture target. Both the root zone depth and the soil moisture target are user-defined. The volume of water utilized for irrigation is constrained by water availability from rivers and the unconfined groundwater reservoirs, while guaranteeing an environmental flow (i.e. irrigation cannot deplete completely the reservoirs). Additionally, priority in abstraction source (surface vs groundwater) is imposed based on the maps of Siebert et al., (2010). This means that a grid cell without infrastructure for groundwater irrigation, for example, will take all the water from the river, and vice versa. This adds an additional constraint to water availability. The water volume is put in the surface of the crop and grass soil column for infiltration, regardless of water source.

Using offline simulations at global scale, we will evaluate the sensitivity of four key factors:  definition of the root zone, setting of the soil moisture target, water availability and the decay of soil hydraulic conductivity with depth. We will then tune the irrigation scheme to match the irrigation volumes reported at country level by the AQUASTAT dataset, and evaluate the effect of irrigation on soil surface hydrology and energy balance. The perspective of this work is to explore the effects of irrigation over present and future climates, using coupled land surface – atmosphere simulations with the IPSL-CM6 climate model. 

S. Siebert et al., “Groundwater use for irrigation - A global inventory,” Hydrol. Earth Syst. Sci., vol. 14, no. 10, pp. 1863–1880, 2010.