Abstract
How well could one predict the growth of a leafy crop from reflectance spectra from the soil and how might a grower manage the crop in the light of those predictions? Topsoil from two fields was sampled and analysed for various nutrients, particle-size distribution and organic carbon concentration. Crop measurements (lettuce diameter) were derived from aerial-imagery. Reflectance spectra were obtained in the laboratory from the soil in the near- and mid-infrared ranges, and these were used to predict crop performance by partial least squares regression (PLSR). Individual soil properties were also predicted from the spectra by PLSR. These estimated soil properties were used to predict lettuce diameter with a linear model (LM) and a linear mixed model (LMM): considering differences between lettuce varieties and the spatial correlation between data points. The PLSR predictions of the soil properties and lettuce diameter were close to observed values. Prediction of lettuce diameter from the estimated soil properties with the LMs gave somewhat poorer results than PLSR that used the soil spectra as predictor variables. Predictions from LMMs were more precise than those from the PLSR using soil spectra. All model predictions improved when the effects of variety were considered. Predictions from the reflectance spectra, via the estimation of soil properties, can enable growers to decide what treatments to apply to grow lettuce and how to vary their treatments within their fields to maximize the net profit from the crop.
Highlights
Leafy crops such as lettuce and brassicas are important commercial crops in the UK
To the authors’ knowledge, there are no studies that seek to explain variance in crop-yield metrics from soil properties estimated by reflectance spectra, work has been done using soil reflectance spectra directly to determine crop characteristics. These include predictions of grain yield in rice (Van Groenigen et al 2003) and plant N uptake (Börjesson et al 1999; Stenberg et al 2005; Terhoeven-Urselmans 2008; Wetterlind et al 2008). This study investigated both methods of predicting crop performance from near- and mid-infrared soil reflectance spectra, i.e. directly from the soil spectra and indirectly using soil properties estimated by reflectance spectra (Fig. 1)
The pH and P were somewhat skewed, but as residuals from the partial least squares regression (PLSR) were approximately normally distributed for pH and P these were not transformed
Summary
Leafy crops such as lettuce and brassicas are important commercial crops in the UK. The value of these crops depends on quality indicators such as size and weight. If growers have dense information on soil variation within their fields, they are likely to be sufficiently well informed to make two main decisions They should be able to recognize a priori where their crops will not reach a saleable quality and so where not to waste time and resources on production. They should be equipped to decide how best to vary fertilizer and irrigation spatially (precision application) to maximize growth without applying excess of either. That has generally been adequate to estimate mean values and average fertilizer requirements It is too coarse, for mapping the variation within individual fields in a way that enables growers to vary their applications of fertilizers and water rationally. Estimated soil properties from reflectance spectra need to be sufficiently accurate to explain variance in crop performance and so inform management decisions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
