Abstract:

Sugar beet (Beta vulgaris L.) yields in England are predicted to increase in the future, due to the advances in plant breeding and agronomic progress, but the intra-field variations in yield due to the variability in soil properties is considerable. This paper explores the within-field spatial variation in environmental variables and crop development during the growing season and their link to spatial variation in sugar beet yield. A further research question was: could spatial variation in the yield of the crop which preceded sugar beet be used for more precise management of the sugar beet. The overall objective is to develop information systems for more precise management of inputs such as water and nutrients in order to increase sugar beet yields and especially yield uniformity within fields and decrease the impacts of adverse environmental conditions.

This study was conducted in three commercial sugar beet fields in the east of England, two in 2012 and one in 2013. A nested sampling scheme with an irregular grid of sample locations was applied in each field although sampling intervals and number differed between the fields. Measurements of biomass and sugar yields, growth, microenvironment (moisture, humidity, air and soil temperature, solar radiation) and soil variables were taken in these plots at different growth stages. Crop management and field operations were uniform across each field, the farmer being responsible for all operations. Final harvest samples were removed from the fields shortly before the farmer harvested the crops. Differential GPS was used for geo-referencing the plots in 2012, while in the 2013 season an RTK GPS was used. The data were analyzed and the variograms were created using GenStat software, while ArcGIS used for Kriging and creating the maps. In addition, spatial variation in the yields of crops preceding sugar beet (winter wheat) were mapped in two of the fields in order to investigate the correlation of spatial variation in sugar beet yield with that of the previous crop.

Geostatistical analysis of results available to date has indicated spatial variation in the majority of studied variables at different growth stages. The sampling protocol accounted for the majority of the variation in two of the three fields. In the third field, a high nugget variance implied a need for denser sampling. The spatial variation differed between the fields and from one variable to another and kriging maps showed that the variation in most properties was distributed as small patches of low and high value. In one field in 2012, the value of the economic yield varied from £1121 to £2994 per hectare evidencing the potential benefits that could accrue for more precise management and the extent of non-uniformity of yields within fields. Based on the kriging interpolation methods and the results of multiple regression analysis, the within field variation in sugar beet yield could be accounted for by the combined effects of different environmental variables, the most influential being soil texture which in turn was correlated with soil moisture and soil temperature. Areas of low soil moisture, leaf area index (LAI), solar radiation interception and crop cover early in the growing season occurred in approximately the same areas of each of the three fields and were closely associated with the zones in each field where there were low sugar yields and crop economic value at harvest.

It is concluded that variation in sugar beet LAI or crop cover early in the season is a good predictor of the within-field spatial variation in final economic yield of sugar. It is therefore suggested that varying fertilizer and perhaps irrigation inputs in accordance with variation in LAI or crop cover could increase the uniformity of sugar yield and crop value. Results will be discussed further in terms of correlations with yield variation of the preceding winter wheat crop.