Derek and others:

CROPGRO, CERES and all other models that are part of DSSAT suite use a
fixed depth increment for each soil layer.  This is defined internally as 5
cm for the first layer, 10 cm for the second layer, 15 cm for the third,
fourth and fifth layers, and 30 cm for the remaining layers.  The main
reason behind this approach was to a create stable system for the soil
water balance, including both soil evaporation from the soil surface, water
movement between the various layers, and drainage from the bottom
layer.  Because we use a daily time step to calculate the soil water
balance, this was critical.  Internally in the model we recalculate the
water holding and other soil characteristics, based on the relative
contribution of each original soil layer as defined by the user. The soil
inputs should be measured for each horizon or partial horizon that has
uniform characteristics.and are independent of the model.

Several model users, including some within our own model development team,
have complained about our "fixed layer" approach.  We are planning to add
flexibility in the next version so that the user will be able to select
which approach to use.

The graphics program has an option to recalculate the model outputs based
on your measured depths for soil water content and soil nitrogen.  In
volume 4, chapter 1, a detailed description is presented on how to define
your measured data in fileT.

Thank you,

GH

At 12:52 PM 7/16/2002 -0600, you wrote:
>Hello,
>I am trying to do a sensitivity analysis for CERES-Wheat of, among other
>things, soil parameters and I have noticed that the model outputs soil
>related information (water content, root length density, etc...) for
>different layers than I input into the model.  I have also noticed that
>this occurs for the example FILEX's provided with DSSAT.  For example, the
>example fileX Ksas8101.whx, provided with DSSAT3.5, lists it soil file as
>IBWH980018.  This soil file lists depth increments of (0-15, 15-30, 30-60,
>60-90, 90-120, 120-150, 150-180).  After running the model for this FILEX
>the model outputs soil information for layers (0-5, 5-15, 15-30, 30-45,
>45-60, 60-90, 90-120, 120-150, 150-180).  The model seems to be
>interpolating soil data between the input soil layers to derive new soil
>input information for layer depths it decides.  Is there a way to get
>CERES-Wheat to output information per layer specified?  If not then how
>does CERES-Wheat decide for what layers it will output information?  If
>there is some set soil layers for which soil information needs to be
>specified what are these set layers?  Thank you very much for any help.
>
>Derek McNamara
>
>Derek McNamara
>Graduate Research Assistant
>Mountain Research Center, Montana State University
>P.O. Box 173490, 106 AJM Johnson Hall
>Bozeman, MT 59717-3490, USA
>406-994-5073
>[log in to unmask]