Publication: Site specific agriculture for the Andean lulo
A group of researchers from DAPA and other institutions have for several years been developing the idea that information obtained from commercial production systems, which are well characterized in terms of climate, weather, soils and overall management, can be used to provide guidelines on how to better manage crops according to the specific conditions of the sites where they are grown. The success of this approach depends on collecting large amounts of data as the number of variables and the noise within the data sets tends to be large. Furthermore, with noisy and often incomplete data sets, with a large number of variables, the analysis and interpretation of the data and its presentation in a form that can help producers make better decisions is a challenge.
A new paper published in Agricultural Systems addresses that challenge, using information obtained from small scale fruit producers who cultivate lulo, a mid altitude Solanaceous tropical fruit. This paper builds on the experience of a previous paper, published in 2009, that developed data analysis strategies for site-specific agriculture for under-researched tropical fruit species using data obtained from Andean blackberry fields.
Small scale fruit growers compiled information on soils, and commercial production, in a standardized format, from 2005 to 2007. Information on climate and weather linked to the individual farmers’ fields was obtained from public databases. We found that an iterative approach, guided by expert opinion and various analytical methods, was the most appropriate means of drawing conclusions that farmers could use to help them make decisions. The methodology relied on using expert opinion to identify variables that cannot be managed by the farmer such as weather, inherent soil characteristics and topography, which were associated with variation in crop response. Once these variables were identified, clusters of Homogeneous Environmental Conditions (HECs) were determined and the effects of management or controllable variables were then analyzed with the HECs being used as a categorical variable. As the data sets did not describe most of the individual management practices of each farmer, farms were used as a categorical proxy for management variation. This methodology successfully identified the most appropriate environmental conditions for obtaining high yields quantifying the yield differences attributable to different HECs. Furthermore, farms were identified that outperformed the majority of farms in each HEC. Although there was not sufficient data on individual management practices to associate particular practices with high productivity, the authors suggest that through further visits to the well managed farms it should be possible to identify those management practices that are associated with high levels of productivity, and conversely those practices which are inappropriate.
This information is extremely valuable as visits to superior farms by farmers with lower productivity, but similar environmental conditions, could provide them with guidelines for improving their yields.
Read the paper from Agricultural Systems here: