For simulating crop growth in smallholder systems it is important to realize
that low-input systems get almost all their nutrients from
soil-organic-matter or residue turnover. The DSSAT models, however, were
developed for systems that were not at all low-input oriented, so their
organic-matter module is quite weak. Also, this module is based on the
PAPRAN model, which was developed for annual pastures and small-grain crops
in a semiarid environment, and its authors explicitly state that the use of
PAPRAN should be limited to those systems. Now as part of DSSAT it is
applied to any system. For the cassava model this may not be of much
importance, but for the other DSSAT models it surely is.

We are working with DSSAT in low-input systems in Central America, where
fertilizer applications are quite limited. For such systems one has to
realize the limitations of the PAPRAN-based SOM module of DSSAT: [1] it only
distinguishes one type of SOM (i.e. 'humus') and
recently-added-but-not-yet-humidified residue; [2] it does not distinguish a
residue layer on top of the soil; [3] newly formed humus is given a fixed
C:N ratio of 10; [4] though three litter C pools are distinguished, there is
only one litter N pool. [5] For residues with C:N ratio < 25, the three
litter C pools decompose at a rate that is independent of the residue's N
concentration; [6] SOM and residue flows are independent of soil texture.

So we decided to link the SOM module from the CENTURY model to DSSAT and
added a new residue layer on top of the soil; this addresses all the points
mentioned. Modifications were also made to the senescence module of the
CROPGRO submodel, so that senesced material is now added daily to the soil
instead of collecting it till the end of the season as is done presently.
This latter is particularly of importance for green-manure systems with a
high input of senesced plant parts. This new SOM/residue module should make
DSSAT much more suitable for low-input systems. It will be part of the next
DSSAT version.


Arjan Gijsman
Centro Internacional de Agricultura Tropical (CIAT)
Cali
Colombia

-----Original Message-----
From: Roy Fawcett [mailto:[log in to unmask]]
Sent: 28 October, 1999 10:27 AM
To: [log in to unmask]
Subject: Re: Use of the DSSAT cassava model ...


Dssat cassava model is a very important contribution to the suite of
models which form the basis of domestic subsitence throughout the
tropics. Development of farming systems models can not take place
without this valuable contribution.

IERM Edinburgh is interested in the use of DSSAT models in small
holder farming. and will integrate the use of this in on going systems
work.
Roy H Fawcett