Decision-making in regard to scheduling transplanting date is one of the least studied aspects of sweetpotato production. It is well documented that transplanting dates can potentially influence storage root yield (Edmond and Ammerman, 1971). Recently, we have documented that up to 85% of adventitious roots extant at 5 to 7 d after transplanting (DAT) have the potential to become storage roots (Villordon et al., 2009b). The uniform and consistent initiation of adventitious roots has been shown to be a critical step in the determination of final yield (Kokubu, 1973; Togari, 1950; Villordon et al., 2009b). Thus, it is important to identify agrometeorological and management variables that exert influence on this specific stage (5 to 7 DAT) to optimize decision-making. In Louisiana, a calendar-based system is used for recommending transplanting dates, i.e., 15 Apr. to 30 June for south Louisiana and 1 May to 30 June for north Louisiana (Boudreaux, 2005). In North Carolina, the recommendations include a provision for soil temperature to be at least 18 °C at a depth of 10 cm for 4 consecutive days before transplanting (North Carolina Sweet Potato Commission, 2009). Some commercial growers in Louisiana are known to temporarily stop transplanting operations if there is a prevailing “northeast wind,” which is typically a cold, dry wind that predisposes transplants to desiccation (Cannon, personal communication). Precise information on the relative importance and interactions of agrometeorological variables on transplant establishment and subsequent storage root yield will potentially benefit researchers, growers, and crop consultants.
The objective of this study was to identify consensus variables at transplant time that were related to U.S. #1 yield outcome using least squares-based linear regression and machine learning approaches. Machine learning generally refers to the class of computational methods for deriving insightful knowledge (including heuristics, strategies, or structure) from data, observations, or past solutions (Shaw, 1993). Models derived from machine learning approaches are also referred to as adaptive models and are characterized by learning by example to solve problems. Adaptive modeling techniques are increasingly being used in areas where there is little or incomplete understanding of the problem to be solved but where training data are available (Park et al., 2005).
Duval, J.R. 2002 Use of prohexadione-CA to increase early yield and reduce establishment irrigation of strawberry (Fragaria ×ananassa) Proc. Fla. Hort. Soc. 115 220 222
Eguchi, T., Kitano, M. & Eguchi, H. 1994 Effect of root temperature on sink strength of tuberous root in sweetpotato plants (Ipomoea batatas Lam.) Biotronics 23 75 80
Golden, E.A., Duval, J.R., Albregts, E.E. & Howard, C.M. 2003 Intermittent sprinkler irrigation for establishment of bare root strawberry transplants Document HS947 Florida Cooperative Extension Service, Institute of Food and Agricultural Sciences Univ., Florida
Kokubu, T. 1973 Thremmatological studies on the relationship between the structure of tuberous root and its starch accumulating function in sweet potato varieties Bull. Fac. Agr. Kagoshima Univ. 22 1 126
Louisiana Agriclimatic Information 2009 Louisiana agriclimatic information 1 Mar. 2009 <http://www.lsuagcenter.com/weather/index.asp>.
North Carolina Sweet Potato Commission 2009 How to grow commercial sweetpotatoes 1 Mar. 2009 <http://www.ncsweetpotatoes.com/content/view/53/78/>.
Pale, S., Mason, S.C. & Galusha, T.D. 2003 Planting time for early-season pearl millet and grain sorghum in Nebraska Agron. J. 95 1047 1053
Park, S.J., Hwang, C.S. & Vlek, P.L.G. 2005 Comparison of adaptive techniques to predict crop yield response under varying soil and land management conditions Agr. Syst. 85 59 81
Riha, S.J., Wilks, D.S. & Simoens, P. 1996 Impact of temperature and precipitation variability on crop model predictions Clim. Change 32 293 311
The University of Waikato 2009 Waikato Environment for Knowledge Analysis (WEKA). Version 3.5.8 The University of Waikato Hamilton, New Zealand
Villordon, A., Clark, C.A., Ferrin, D. & LaBonte, D. 2009a Using growing degree days, agrometeorological variables, linear regression, and data mining methods to help improve prediction of sweetpotato harvest date in Louisiana HortTechnology 19 133 144
Villordon, A., LaBonte, D., Firon, N., Kfir, Y., Pressmn, E. & Schwartz, A. 2009b Characterization of adventitious root development in sweetpotato HortScience 44 651 655
Villordon, A., LaBonte, D.R. & Firon, N. 2009c Development of a simple thermal time method for describing the onset of morpho-anatomical features related to sweetpotato storage root formation Sci. Hort. 121 374 377