Pasture data reader SQLite

CLEM.FileSQLitepasture

Reads external grass production database for pasture production

Description

The Data reader components are used to read external data for use by CLEM. It is recommended that you place this component with the activities using them or below the CLEM component.

This Data reader is used to read a pasture production data cube in SQLite format. This is the most efficient means of supplying pasture production data in terms of simulation time. The data cube contains the results of pasture production simulations for a given historic period across a range of land conditions, grass basal areas and stocking rates.

This is an advanced feature for modelling pasture (e.g. properties in northern Australia). Expertise is required to produce the lookup database with output from a pasture model. Contact the CLEM team to discuss using other models to provide pasture production.

Optimising your database for performance

As the pasture production data-cube contains pasture growth for all the required permutations of land condition, soil type, grass basal area and stocking rate, for at least the duration of your simulation, this is often a large file. The model needs to be query this database for pasture growth for each change in the ecological indicators associated with the pasture which accounts for a large proportion of the CPU requirement of your simulation. For this reason we highly recommend you add an Index to your database to significantly reduce the time taken to perform your simulation, especially for extended simulations and multiple runs using APSIM Experiments. Adding the Index can be performed in SQLite management software (such as DB Browser). The SQL statement to run on your database is:

CREATE INDEX CLEM_next_growth ON Native_Inputs (Region, Soil, GrassBA, LandCon, StkRate, Year, Month);

This index must be named CLEM_next_growth and should include the table and column names appropriate to your database.

CLEM will produce a warning if your database has not been optimised with an appropriate index.

The model will also automatically generate a number of other small tables in your database to store unique values of the ecological indicators to also improve performance.