Testing two existing fertilizer recommendation algorithms: Stanford's 1.2 rule for corn and site-specific nutrient management for irrigated rice

["Stanfords 1.2 Rule";site-specific nutrient management;Site-specific nutrient management (SSNM);nitrogen;fertilizer;fertilizer recommendation;fertilizer algorithm;nutrient management;corn;rice

Rodriguez, Divina Gracia

关键词: ["Stanfords 1.2 Rule";    site-specific nutrient management;    Site-specific nutrient management (SSNM);    nitrogen;    fertilizer;    fertilizer recommendation;    fertilizer algorithm;    nutrient management;    corn;    rice;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/49749/Divina%20Gracia_Rodriguez.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

This dissertation evaluates two existing fertilizer recommendation algorithms that are based on yield goal approach: Stanford’s 1.2 Rule for corn in the United States and site-specific nutrient management (SSNM) for rice, which uses an algorithm similar to the 1.2 Rule, in Southeast Asia.Fertilizer recommendations all over the world have relied heavily on an old but widely accepted rule of thumb from Stanford (1966, 1973): apply 1.2 pounds of nitrogen (N) fertilizer per bushel of corn expected. While algorithms similar to the “1.2 Rule” have been used for the past four decades all over the world to make fertilizer recommendations for various crops, little is known about the 1.2 Rule’s origin. I use microeconomic analysis to examine the historical origins of the 1.2 Rule and show that the 1.2 Rule only makes economic sense if the crop production satisfies two restrictions: (1) it is of the von Liebig functional form, i.e. the function has a “kink” and a “plateau,” and (2) the kinks of the von Liebig response curves for different growing conditions lie on a ray out of the origin with slope 1.2. To investigate if the 1.2 Rule satisfies these restrictions, I utilize the original dataset Stanford used in his analysis and the long-term corn experiment data from Illinois, Iowa, and Nebraska. I find no empirical evidence to support the 1.2 Rule using non-linear estimation techniques and a non-nested hypothesis framework. The crop production function and the critical concentration of N vary across and even within fields, and hence site-specificity matters in making fertilizer recommendations. I also critically discuss and evaluate the assumptions underlying the SSNM strategy for rice in the top rice producing countries in the world: India, Indonesia, Philippines, Thailand, and Vietnam. I find clear evidence that interaction among major nutrients matters in making fertilizer recommendations to farmers.The relationships among N, P, and K vary across sites --some inputs are complements, some are substitutes, and some are independent. I also find that soil organic matter, manifested in soil C stocks, significantly affect the economic returns to N fertilizer inputs.The marginal product on N is low on soils with low C content.These results suggest the SSNM strategy should explicitly account for the: (1) nutrient interactions and (2) relationship of N fertilizer and soil organic matter, as reflected in soil C stocks. In addition, input and output prices should also not be ignored in SSNM algorithm. The major challenge for SSNM strategy will be to retain the simplicity of the approach that is understandable to producers and extension agents while accounting for the relationship of NPK, soil organic matter, and prices.Clearly, this is an area of research in great need of interdisciplinary collaboration among agronomists and agricultural economists.