Premium
The First Step Towards a Decision‐support System for Sugar‐beet Growing: Selection of a Basic Growth Model
Author(s) -
Smit A. B.,
Struik P. C.
Publication year - 1995
Publication title -
journal of agronomy and crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.095
H-Index - 74
eISSN - 1439-037X
pISSN - 0931-2250
DOI - 10.1111/j.1439-037x.1995.tb00215.x
Subject(s) - sugar beet , sugar , selection (genetic algorithm) , regression , agricultural engineering , regression analysis , agronomy , mathematics , computer science , statistics , chemistry , biology , machine learning , engineering , food science
Four simulation models were tested for their suitability as a basic growth model in a decision support system for sugar beet growing in The Netherlands. SUCROS and SUBEMO are complex, mechanistic models; LINTUL and PIEteR are relatively simple regression models including causal relationships at a higher level of integration. All four models are dynamic, i.e. they calculate daily development and production rates. The selected model had to be able to predict root and sugar yields accurately and to correct for suboptimal water contents of the soil. It should be possible to include location‐specific data and new modules, e.g. for nitrogen fertilization or plant density. Finally, the farmer should be able to collect the required input data easily and cheaply. The tests showed that PIEteR predicted root and sugar yields best, partly because it contained water‐balance corrections, based on location‐specific soil characteristics. PIEteR could not be applied universally because of its regression character at a high level of integration, but it met the requirements specified.