An analysis of the factors that influence sugarcane yield in Northern Argentina using classification and regression trees

FERRARO, DIEGO OMAR; RIVERO, D.E.; GHERSA, C. M.

FIELD CROPS RESEARCH

Elsevier

Año: 2009 vol. 112 p. 149 - 157

0378-4290

Multi-location trials are commonly used to estimate the effects of different explanatory factors on cropyield. Conversely, the analysis of production databases could also be useful for exploring andunderstanding such effects. These data require flexible and robust methods for dealing withmultivariate, non-linear and unbalanced data structures, high-order interactions and missing values.In this paper, we explore the issue of crop yield explanation using a 5-year period (1999–2005) ofsugarcane (Saccharum officinarum L.) yield data from Northern Argentina. Using a data mining techniquesuch as classification and regression trees (CART) we show that farm membership (FARM) was amongthe main splitting factors for total cane per hectare (TCH) cluster variability. Crop class (AGE) was at thesecond level in the hierarchy and values of AGE higher than 2,5 splitted low and medium from the highTCH clusters. Sugarcane cultivar (VAR) was the most important explanatory factor regarding total sugarper hectare (TSH), and crop class (AGE) was second in importance. In this case, farm membership did notappear among the main splitting factors. The growth period duration, field area and precipitation did notshow remarkable importance values for explaining final TCH and TSH values. By-year CART models alsoshowed low values of importance of weather related variables across the years analyzed suggesting thatother environmental conditions than precipitation is controlling yearly variation in sugar and cane yield(e.g. radiation, water-use efficiency or temperature regime). The CART analysis developed here is the firstsystematic analysis for explanatory factors of biomass and sugar content in Argentina’s cane mostproductive region. However, we believe this methodology could be applicable for a wider geographicarea and other sugarcane production regions as well as other cropping systems. Although regressiontrees provide less formal statistical inference, its results could be added as an additional analytical tool totraditional experimental analyses that use mixed models. Also, they could be useful for elaboratinghypotheses and suggest mechanistic studies to test them.