Key genotype and soil parameters for sugarcane growth: A sensitivity analysis of the SAMUCA model in Brazilian agroecosystems

Author(s)

Pereira, Rodolfo Armando de Almeida ; de Melo, Marina Luciana Abreu ; Vianna, Murilo dos Santos ; van Lier, Quirijn de Jong ; Brunetti, Henrique Bauab ; Marin, Fabio Ricardo

• Document type: Journal article
• Source: FIELD CROPS RESEARCH; v. 322, p. 14-pg., 2025-01-24.

Abstract

Context: Climate constraints drive Brazilian farmers to improve production efficiency. Process-based crop models are crucial for evaluating cultivars, management strategies, and environmental scenarios. To optimize model efficiency and enhance interpretations, it is essential to identify key parameters across diverse production environments. Objective: We aimed at identifying the key genotype and soil parameters for the SAMUCA sugarcane crop model in Brazil's main producing regions, under irrigated and rainfed conditions. Methods: A global sensitivity analysis (GSA) was performed using two methods: the extended Fourier Amplitude Sensitivity Test (eFAST) and the Partial Rank Correlation Coefficient (PRCC). A total of 31 parameters of the SAMUCA model were analyzed, including 24 related to genotype and 7 to soil. The GSA was applied across 8 representative sugarcane production environments in Brazil, including both rainfed and irrigated conditions. Results: Under irrigated conditions, key genotype parameters were tillochron and popmat, which explained over 90 % of the variance in tillering. The leaf area index (LAI) was notably affected by genotype parameters, with minimal impact from soil parameters. Under rainfed conditions, both genotype and soil parameters were influential. Over 85 % of the variance in sucrose concentration was explained by the genotype parameter mid_tt_it_gro. For stalk mass, plastochron and eff were the primary parameters. While mla was the main parameter for LAI in irrigated scenarios, the soil parameters were influential in site-specific rainfed scenarios. Conclusions: In irrigated environments, genotype parameters are more influential, whereas in rainfed conditions, soil parameters become more important. This highlights the need for targeted parameter selection to enhance crop model accuracy under varying climates and water availability in sugarcane production environments. Implications: Our findings highlight the need for integrated crop management in sugarcane, emphasizing both genotype and soil factors, especially in rainfed systems. We recommend refining crop models to incorporate more parameters, supporting better decision-making in agriculture. (AU)

• FAPESP's process: 21/00720-0 - Advancing databases and modeling techniques for yield-gap analysis in Brazilian agricultural systems
• Grantee: Fabio Ricardo Marin
• Support Opportunities: Regular Research Grants
• FAPESP's process: 24/02381-6 - Future scenarios of crop yield and water use by sugarcane in Brazil: a multi-model approach for soil management in adaptation to climate change
• Grantee: Marina Luciana Abreu de Melo
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• FAPESP's process: 21/10573-4 - Center for Carbon Research in Tropical Agriculture (CCARBON)
• Grantee: Carlos Eduardo Pellegrino Cerri
• Support Opportunities: Research Grants - Research, Innovation and Dissemination Centers - RIDC