Implementation of flood irrigation and paddy irrigation as well as the irrigation method dataset

[YiYaoVUB]

• [1] Currently, drip and sprinkler irrigation methods have been implemented in CTSM. However, the most widely used method, flood irrigation, has not yet been developed. The difference between drip and sprinkler irrigation is where water is applied (under canopy for drip and over canopy for sprinkler). We think for flood irrigation, the water should also be applied under the canopy, but the amount of water should be different. Considering that the soil normally will be saturated in a short period, we propose that we change the amount of water to the deficit between saturated soil water and current soil water.

• [2] To simulate irrigation-induced impacts more realistically, a surface dataset with the spatial distribution of the irrigation method will be needed. In #565 @billsacks has posted the enhancement required. Here we propose to use the data from https://hess.copernicus.org/articles/19/3073/2015/, in which they generated a global map of irrigation method methods at 0.5x0.5 resolution. For each CFT in every pixel, they provided the fraction of three irrigation methods (flood, sprinkler, and drip). In mksurfdata tool, we propose to remap the fraction of irrigation methods, then choose the dominant one to limit the computation cost.

• [3] Rice is a CFT that requires special irrigation. During the growth season, the surface water needs to be held in the field for some periods. We propose to change the surface water module of rice paddies, so water could be held at a given point (at the saturated point or a given depth).

[wwieder]

Thanks for creating this issue @YiYaoVUB. Next steps are to open a pull request that addresses each of the topics above. Let's start with the with flood irrigation work and move on from there.

Thanks again for these contributions!