|Tubiello, Francesco - NASA-GODDARD|
|Rosenzweig, Cynthia - NASA-GISS|
|Pinter Jr, Paul|
|Garcia, Richard - LI-COR INC LINCOLN NE|
Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: September 15, 1998
Publication Date: N/A
Interpretive Summary: In order to predict the consequences of present and future global environmental changes on the security of world food production and on future irrigation requirements, efforts are underway to develop the capability to predict the growth, yield, and water use of major food crops. These global changes entail an increasing concentration of atmospheric carbon dioxide (CO2) which is expected to double sometime during the next century. Climate modelers have predicted that the elevated CO2 will cause the earth to warm and that precipitation patterns will change. Elevated CO2 is also known to alter the growth of plants and may affect their water requirements. Accordingly, a computer model called CERES-Wheat was modified, which is capable of predicting the growth of a wheat crop day by day. This paper describes a specific validation test of the model comparing its predictions with actual data from a free-air CO2-enrichment experiment on wheat at CO2 concentrations of 550 ppm and present-day levels of about 370 ppm. The results showed that the model could predict plant phenology, dry matter production, and the subsequent partitioning of the dry matter into leaf, stem, root, and grain reasonably well. This work should help growers develop optimum management strategies for increased CO2 concentrations predicted in the next century. The model can also aid in determining the impact of global change on the world's future food supply.
Technical Abstract: Free Air Carbon-dioxide Enrichment (FACE) experiments were conducted in 1992-93 and 1993-94 at Maricopa, AZ, to investigate the dynamics of CO2- water interactions in spring wheat (Triticum aestivum L., cv. Yecora Rojo). We use these experimental data to evaluate a modified version of the CERES-What model. Simulations are compared to observations of plant phenology, plant dry matter production and its subsequent partitioning to leaf, stem, root, and grain organs during the crop life-cycle. Four simulations for each experimental season are performed, representing combinations of two irrigation treatments (well-watered and water- stressed); and two atmospheric CO2 concentrations (ambient, 350 ppm and elevated, 550 ppm). We use the well-watered, ambient CO2 1992-93 dataset for model development and evaluation, including testing of photosynthesis equations developed in previous work. Our results indicate that the model can predict above-ground dry matter production and grain yield satisfactorily, to within 10% of the observed values; but that its simulations of component plant organs are incorrect over the crop life cycle. Leaf and root dry matter at harvest are overpredicted by as much as 150%, while final stem dry matter is underpredicted by more than 40%. Modifications to the model's carbon partitioning routines allow for much improved prediction of leaf, root dry matter over the life cycle. The modified model is used to simulate the remaining treatments in the 1992-93 and 1993-94 FACE datasets, predicting the effects of CO2 concentration and water management on total above-ground, leaf, stem, root and grain dry matter production in good agreement with observations.