Note: This page is no longer being maintained and is kept for archival purposes only.
For current information see our main page.
GWI Kurtz-Fernhout Software
Developers of custom software and educational simulations.
Home ... News ... Products ... Download ... Order ... Support ... Consulting ... Company
Garden with Insight
Product area
Help System
Quick start

Garden with Insight v1.0 Help: Soil patch next day functions: allocate water to plants

After each of the plants in the soil patch has carried out some of its next-day functions the soil patch presides over the allocation of water to each of the plants. First the soil patch goes through the list of plants asking each plant for its water demand.

Water demand for each plant is calculated using the potential-actual method in several steps. 1) The first estimate of water demand is the potential transpiration for the plant, which was calculated several functions before this. 2) That estimate is reduced by an exponential distribution which represents the decreasing spread of the plant roots as you move down the soil profile. 3) Then another adjustment is made for the plant's ability to compensate for a water deficit in one layer by pulling more water from another layer. The plant's ability to do this deficit compensation depends on the root growth constraint for the layer as an index of root health in the layer. 4) The next estimate is generated by reducing water demand if the amount of plant-available water (above the wilting point) is less than 25% of the possible plant-available soil water (field capacity minus wilting point). This limit represents physical limits on how much water the roots can quickly take up from the soil. 5) And finally the plant water demand is reduced by the root growth constraint in each soil layer. You would expect the root growth constraint to affect the addition of new root biomass directly, but instead the simulation uses the root growth constraint to reduce water demand by the roots in each layer, then uses the water use in each layer to calculate root growth. It comes out to the same thing.

Now the soil patch has a water demand for each of its plants. The total amount of plant-available water in the soil patch is the soil water content minus the wilting point. The reason the soil still has water at its wilting point is because plant roots cannot reach into very tiny soil pores, called micropores. The water in those tiny pores can never be absorbed by plants, and so they wilt in soil that still contains some water. The amount of water to be distributed among the plants in the patch, then, is the soil water content minus the wilting point, but we take an extra 10% of that amount off for good measure. The soil patch partitions the available water among its plants in exact relation to their demands. There is no weighting of individual plants by leaf area, or biomass, or even root biomass. This is because each plant takes its own root biomass distribution and root growth constraints into account when it figures its demand.

Actually, the water partitioning part of the simulation was our addition to EPIC. EPIC simulates an entire field of plants as one mass of living tissue and has no competition among plants. Another model based on EPIC, ALMANAC, simulates competition among several species of plants in a field (usually crops and weeds). We wanted a flexible approach that would allow any number of plants of any number of types, so we chose this simple method. We simply changed EPIC's "water use" method to read "water demand" and added partitioning between plants. This method is somewhat like the method in SPUR, a model of rangeland grasses. There are problems with this partitioning scheme: in reality, some species of plants outcompete other species by various methods; this model does not take into account the differing root shapes of different species (tap roots, surface root mats, etc); and some plants even actively discourage the growth of other plant roots. These are all areas for improvement in the future.

Also notice that the water partitioning model has no spatial component. All the plants in a soil patch share equally in the soil patch's water, even though one plant may be squarely in the soil patch and another may be at the edge. Two plants at the furthest reaches of a large soil patch inhibit each other's water uptake just as much as two plants right next to each other.

After the plant-available water is distributed, the plants record their water use and the total amount of water used in each soil layer is subtracted from the water in the layer.

calculation of potential transpiration, root growth constraint
EPIC Plant Water Use
Model contents

Home ... News ... Products ... Download ... Order ... Support ... Consulting ... Company
Updated: May 4, 1998. Questions/comments on site to
Copyright © 1998 Paul D. Fernhout & Cynthia F. Kurtz.