Long Paddock

GRASP

Introduction

Simplified model of the grazing system

Simplified model of the grazing system

GRASP (Littleboy and McKeon, 1996) is a model of the climate-soil-plant-animal-management of perennial grasses of Northern Australia.  It is a daily timestep model which uses daily climate data including rainfall, temperature, evaporation, radiation and vapour pressure.  Other model inputs include soil data (e.g. field capacities and wilting points), plant growth, cover, temperature responses, nitrogen, senescence, litter breakdown and animal intake.  It is a one dimensional model simulating a point on the landscape but is used spatially in AussieGRASS and a number of other products where output is averaged across spatial zones (e.g. paddocks, land types, 5 klm pixels).  There is a management component including management records which are management decisions like stocking rate and burning.  Within the management records observations of features like total standing pasture and pasture can be input to calibrate the model.

History

GRASP is based on the WATSUP simulation model by Rickert and McKeon (1982) and was improved upon into what has since been known as GRASP (McKeon et al 1982).  There have been a number of divergent models including PROBE (McKeon et al 1986), HERDECON GRASP and Perfect (Littleboy 1992). Greg McKeon's primary version of GRASP has been known as "Spaghetti" GRASP (primarily because of the appearance of the code).   In 1999 there was a new version of the command line version of GRASP, that was completely rewritten by Neil Flood with input by Greg McKeon and others, called "Cedar" GRASP (named after the initial location of it's creation, Cedar Creek Conference Centre).  Limited testing has shown that Cedar GRASP gives almost the same results as the older Spaghetti GRASP, but some modules are more tested than others.

Caveats

GRASP can simulate many land types through Australia reasonably well, however some land types and management types may need extensive re-parameterisation.  As stated above it was primarily written to model perennial grasses of Northern Australia.  How well a land type is simulated in GRASP depends primarily on the correct calibration of the parameter set.  Modified land types (e.g. irrigated, fertilized or improved pastures) may not be simulated correctly without extensive re-parameterisation and model enhancement.  If you are after a full list of the limitations please get in touch with the contact below.  The users of GRASP are expected to exercise caution when interpreting results and inform the developers (through the contact below) of problems they discover.

Further documentation

The main reference for GRASP is:

Rickert, K.G., Stuth, J.W. and McKeon, G.M. (2000). Modelling pasture and animalproduction. In ‘Field and Laboratory Methods for Grassland and animal Production Research’. (Eds. L. ‘t Mannetje and R.M. Jones). pp. 29-66 (CABI publishing: New York)

For further references see: GRASP references (PDF, 16K, last updated 10:46AM, 29 June 2010)*

Below are five documents: The GRASP report (Evaluating The Risk of Pasture and Land Degradation in Native Pastures in Queensland) was the original report in 1997 to RIRDC in which the GRASP manual was an appendix to. The GRASP in 17 OHTs (PDF, 538K, last updated 10:50AM, 29 June 2010)* (actually 19 OHTs) is a pdf version of a talk Greg McKeon does to introduce GRASP. The remaining two are the Swiftsynd/Minisynd manuals to aid it setting up and collecting observational data for GRASP.

Phone (07) 3896 9702 for further information.

NOTE: Many of the products show in these web pages are still under development and while all care has been taken to find problems there will still be some unfound bugs.
Please see our disclaimer before using the products on these pages.

* Requires Adobe Reader

Last updated: 04 December 2008

GRASP