Add waterDrainFrac to allow flooding.

[dlebauer]

Currently, SIPNET drains all soil water in excess of water holding capacity (WHC) instantly. To allow realistic simulation of saturated or flooded soil conditions, our chosen option is to add a parameter that controls drainage rate.

Add a parameter (e.g., waterDrainFrac ($f_{\text{drain}}$ in documentation) that controls the fraction of excess soil water drained per day:

Setting waterDrainFrac < 1 would enable prolonged saturated/flooded conditions. This will enable realistic representation of managed flooding (e.g., rice), as well as poorly drained / wetland ecosystems.

[Alomir]

In setupModel(), where we initialize the soil water pool, we will need to think about this behavior, esp. w.r.t restart capabilities

  if (envi.soilWater < 0) {
    envi.soilWater = 0;
  } else if (envi.soilWater > params.soilWHC) {
    envi.soilWater = params.soilWHC;
  }

Update: we are removing the else branch.

[Alomir]

Open question: in moisture(), where we calculate transpiration, should we cap available water at soil WHC? (It seems that excess water from flooding/pooling would not be not available for transpiration)

[Alomir]

Open question: what do we do about the evaporation calc:

    // TODO: consider replacing this with getClippedWaterFrac()
    double waterFrac = water / params.soilWHC;
    rsoil = exp(params.rSoilConst1 - params.rSoilConst2 * (waterFrac));
    rd = (params.rdConst) / (climate->wspd);  // aerodynamic resistance (sec/m)
    *evaporation = CONVERSION * climate->vpdSoil / (rd + rsoil);
    // by using vpd we assume that relative humidity of soil pore space is 1
    // (when this isn't true, there won't be much water evaporated anyway)

If there's water pooled on the surface, is calculating evaporation from the soil even a thing? Or, is this really modeling surface evaporation, and it's sufficient to just leave the waterFrac capped at 1?

[dlebauer]

Yes, I concur with both of your suggestions:

1. in setupModel yes remove upper clamp; this seems essential for valid restart
2. in moisture(), cap plant available water at WHC. This follows the physiological mechanism that water above saturation does not affect ET.
3. For evaporation, leave waterFrac capped at 1; SIPNET doesn't separately represent standing / ponded water. Perhaps add a comment that evaporation from standing water is approximated as saturated surface evaporation.