Using ceres-maize model for maize sown off-season yield forecast.

Author(s)

Cecilia Manuela Tojo Soler

• Document type: Doctoral Thesis
• Press: Piracicaba.
• Institution: Universidade de São Paulo (USP). Escola Superior de Agricultura Luiz de Queiroz (ESALA/BC)
• Defense date: 2004-08-13

Examining board members

Marcos Vinicius Folegatti; Luiz Roberto Angelocci; Marcelo Bento Paes de Camargo; Durval Dourado Neto; Mario Jose Pedro Junior

• Advisor: Paulo Cesar Sentelhas

Abstract

The objectives of the present work were: 1) to study the phenology and development of four maize hybrids: AG9010, DAS CO32, Exceler and DKB 333B sown off-season in Piracicaba region, State of São Paulo, Brazil; 2) to calibrate and test the CERES-Maize model to estimate the productivity and development for the four maize hybrids sown off-season for the mentioned region; 3) to apply the model for determination of sowing dates with less risk, for irrigated and rainfed conditions, for the four maize studied hybrids; 4) to apply the CERES-Maize model for long term climatic conditions aiming yield forecast; 5) to verify the association between climatic parameter and ENSO phases (El Niño Southern Oscillation) for Piracicaba region; 6) to analyze the influence of ENSO phases in maize productivity. Three field experiments that included four hybrids with different maturity durations were conducted during the autumn and winter growing seasons of 2001 and 2002 at the Escola Superior de Agricultura "Luiz de Queiroz", University of São Paulo, in Piracicaba, SP, Brazil. One experiment was conducted in 2001 under irrigated conditions, and two experiments were conducted in 2002, one under rainfed and another under irrigated conditions. All experiments had a randomized complete block design. The maize hybrids used in this study were: AG9010 (very short season), DAS CO32 and Exceler (short season) and DKB 333B (normal season). The sowing dates were March 15th, for the 2001 experiment, and March 13th, for the 2002 experiments. CERES-Maize model was calibrated and tested with data obtained in these experiments. The results suggested that maize sown off-season under rainfed conditions, presented in general, lower values of LAI, plant height and less dry matter accumulation than under irrigated conditions. An increase in root water uptake on deep layers of the soil was evident on maize crop under soil water limitations conditions. The CERES-Maize model simulated with accuracy the maize phenology and yield for off-season conditions. In general, for the two cropping systems, irrigated and rainfed, as more delayed the planting date is, less simulate yield for the four hybrids. The CERES-Maize was an excellent tool for maize off-season yield forecast using climatic historic data combined with data from the year of study. This procedure permitted to obtain a good estimation of yield 45 days before harvest for the hybrids sown off-season in Piracicaba. The ENSO phases had influence on maize sown off-season yield, as result of the precipitation patterns during April and May months. The highest yields were simulated for "El Niño" years, being higher than yields for "La Niña" years, which were higher than for neutral years. (AU)

• FAPESP's process: 00/09050-0 - Using Ceres-Maize model for maize sown off-season yield forecast in the region of Vale do Paranapanema (SP)
• Grantee: Cecilia Manuela Tojo Soler
• Support Opportunities: Scholarships in Brazil - Doctorate