Scientia Horticulturae 182 (2015) 18–26 Contents lists available at ScienceDirect Scientia Horticulturae journal h om epa ge: www.elsevier.com/locate/scihorti Estimating nitrogen nutritional crop requirements of grafted tomatoes under field conditions Desire Djidonou a , Kenneth Lopiano b , Xin Zhao a,∗ , Eric H. Simonne a , John E. Erickson c , Karen E. Koch a a Horticultural Sciences Department, University of Florida, Gainesville, FL 32611, USA b Statistics Department, University of Florida, Gainesville, FL 32611, USA c Agronomy Department, University of Florida, Gainesville, FL 32611, USA a r t i c l e i n f o Article history: Received 7 July 2014 Received in revised form 10 October 2014 Accepted 15 October 2014 Keywords: Grafting N rates Solanum lycopersicum L Yield response functions Confidence interval Bootstrap method a b s t r a c t An experiment was conducted to test the hypothesis that the enhanced yield possible with grafted tomato (Solanum lycopersicum L.) under field conditions will also increase the nitrogen (N) crop nutri- tional requirement (CNR). Determinate ‘Florida 47’ tomatoes were grafted onto interspecific hybrid rootstocks (‘Multifort’ or ‘Beaufort’) and grown in a sandy soil with six N rates (56, 112, 168, 224, 280, and 336 kg ha -1 ) under plastic mulched bed and drip-irrigation systems during the spring seasons of 2010 (March–June) and 2011 (April–July). The N-CNR for grafted and non-grafted tomatoes was assessed using five yield response functions: exponential, linear-plateau, quadratic-plateau, quadratic, and square root. Over the two seasons, the estimated N-CNR ranged from 165 kg ha -1 with the quadratic-plateau model to 324 kg ha -1 with the square root model. Confidence intervals (CI) around these N-CNR ranged from 125 to 585 kg ha -1 using the bootstrap method and from 98 to 440 kg ha -1 using the delta method. Anal- ysis of these CIs gave N-CNR rates of 239–246 kg N ha -1 for grafted plants, and 196–197 kg N ha -1 for non-grafted plants. Predicted maximum marketable yields were similar between the models, ranging from 56 -71 Mg ha -1 for grafted plants, and 43–53 Mg ha -1 for non-grafted plants, over the two seasons. Overall, while the actual N-CNR is likely to vary with season, soil types, and management practices, the results indicated that grafted tomato plants had a greater N-CNR than non-grafted plants together with an increase in predicted marketable yield. The yield response curves also showed that at a fixed marketable yield goal within the estimated range, the N fertilization rate required was lower for the grafted tomato plants as compared with the non-grafted plants. This study demonstrated that N fertilization program for optimizing tomato production may be modified when grafted plants are used. © 2014 Elsevier B.V. All rights reserved. 1. Introduction The use of vigorous, interspecific hybrid rootstocks has shown enhanced growth and fruit yields for several solanaceous and cucurbitaceous vegetable crops, including tomato (Solanum lycop- ersicum L.), watermelon (Citrullus lanatus (Thunb.) Matsum. & Nakai), melon (Cucumis melo L.), cucumber (Cucumis sativus L.), pepper (Capsicum annuum L.) and eggplant (S. melongena L.) (Lee and Oda, 2003; Passam et al., 2005; Leonardi and Giuffrida, 2006; Colla et al., 2008, 2010; Di Gioia et al., 2010; Djidonou et al., 2013). For example, a 39% increase in total marketable yield of grafted versus non-grafted watermelon across three N rates (0, 50, ∗ Corresponding author. Tel.: +1 352 273 4773; fax: +1 352 392 6479. E-mail address: zxin@ufl.edu (X. Zhao). and 100 kg ha -1 ) was accompanied by a 21% relative increase in N uptake efficiency (Colla et al., 2011). However, yield potential for these grafted cucurbits at even higher N levels have not been examined. In tomatoes, increased N uptake has also been observed in grafted relative to non-grafted plants (Leonardi and Giuffrida, 2006). It is very likely that the N-CNR and recommended N fertil- ization rates may differ for grafted versus non-grafted tomatoes. In general, the CNR represents the total, seasonal rate of N required by a given crop to achieve maximum yields (Cantliffe et al., 2006). Specifically in Florida, recommended N fertilization rates for field-grown tomato are based on empirical studies that assessed the CNR across various tomato growing areas (Cantliffe et al., 2006; Hochmuth and Hanlon, 2010). Results of these various studies have led to a recommended N rate of 224 kg ha -1 for field-grown, round, plum-type, and grape tomato production on sandy soils in Florida (Olson et al., 2009). On the other hand, empirical values of the http://dx.doi.org/10.1016/j.scienta.2014.10.034 0304-4238/© 2014 Elsevier B.V. All rights reserved.