Evaluation of a rice crop growth model for selected local Malaysian rice varieties at IADA Ketara, Terengganu
Rice is a staple food for more than half of the world’s population. Current recommended rates of fertilizers are determined through field, pot culture or greenhouse trial experiments. These approaches are costly, labour extensive and time consuming. As an alternative, simulations of rice crop growth...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | English |
Published: |
2020
|
Subjects: | |
Online Access: | http://psasir.upm.edu.my/id/eprint/104677/1/MUHAMAD%20FAIZ%20BIN%20CHE%20HASHIM%20-%20IR.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | Rice is a staple food for more than half of the world’s population. Current recommended rates of fertilizers are determined through field, pot culture or greenhouse trial experiments. These approaches are costly, labour extensive and time consuming. As an alternative, simulations of rice crop growth models are valuable tools to perform complex scenario studies including studies of nitrogen (N) fertilization. However, before a model can be used locally, the model’s performance must be evaluated. The main objective of this thesis is to evaluate the performance of a rice crop growth model, ORYZA(v3), in simulating six rice crop physiological traits for five Malaysian rice varieties namely MR219, MR220CL2, MR269, MR297 and UPUTRA. The specific objectives are (i) to investigate crop physiological responses of MR219, MR220CL2, MR269, MR297 and UPUTRA treated with three N rates under field conditions and (ii) to calibrate and validate ORYZA(v3) using the rice crop physiological data obtained from the field experiment conducted in IADA KETARA, Terengganu. In this thesis, a field experiment was carried out to investigate the physiological responses of Malaysian rice varieties to three N treatments using a Split-plot Randomized Complete Block Design experimental design. The main factor in this study was N rates arranged in four blocks while the subplot factor was the five selected rice varieties. The three N treatments are T1 (76 kg N ha-1), farmer’s practice T2 (109 kg N ha-1), and T3 (142 kg N ha-1). Statistical analysis of two-way ANOVA was conducted to investigate the effects of N and varieties on the rice crop physiological traits. The ORYZA(v3) was calibrated and validated using the field experimental data sets. In this study, the Nash-Sutcliffe efficiency (NSE) was used as an indicator of the model performance as the NSE reflects the overall fit of model simulations against the observations. Except for the dry biomass of panicles, the two-way ANOVA shows that there is lack of N responses on most of plant physiological traits especially in relative chlorophyll content, leaf area index and dry biomass of green leaves. Each of the variety has different biomass partitioning percentage for different plant organs at different growth phases, but overall, no distinct trend was observed. Calibration of ORYZA(v3) using T1 data consistently resulted in acceptable NSE, where 25 out of 30 rice crop physiological traits of the T1 treatment had NSE of above 0.6. ORYZA(v3) was poorly calibrated T2 and T3 treatments. Except for MR220, validation of the model shows better fit between simulations and observations for the T2 treatment compared to the T3 treatment for the other four varieties. In this thesis, calibrated crop files improved the performance of ORYZA(v3) in simulating the local Malaysian rice varieties for the low N treatment when compared to using a default crop file. Confidence in ORYZA(v3)’s performance and its robustness can be increased with more data in future. |
---|