Non-destructive prediction of soluble solids and dry matter content using NIR spectroscopy and its relationship with sensory quality in sweet cherries

•Near-infrared models were built to predict quality traits in sweet cherries.•Predictions were accurate and robust for soluble solids and dry matter content.•Dry matter was equal to or superior to soluble solids to predict flavor intensity. Near infrared spectroscopy (NIR) models developed in the spectral region between 729 and 975nm for a handheld instrument have been demonstrated to be accurate, robust and versatile for the rapid and non-destructive evaluation of soluble solids (SSC) and dry matter (DMC) content in sweet cherries at two different temperatures, 0 and 23°C. Coefficient of determination (R2) values for model calibrations ranged from 0.922 to 0.946 for SSC and from 0.910 to 0.933 for DMC, with standard errors of calibration from 0.612 to 0.792 and from 0.687 to 0.911% for SSC and DMC, respectively, depending on the variety and the temperature of the fruit. External validation of the models demonstrated they could be confidently used for ‘Chelan’ and ‘Bing’ sweet cherries; R2 values for SSC ranged from 0.726 to 0.891 and for DMC from 0.670 to 0.725 between the predicted and the observed values in the external validation set, respectively. Consumer preference evaluation for ‘Bing’ and ‘Chelan’ sweet cherries with differing SSC (16.38–28.45) and DMC (17.67–31.62%) revealed higher acceptance by consumers for cherries with high DMC as well as for cherries with high SSC. DMC was equal to or superior to SSC for predicting flavor intensity and balance of sweet to sour taste as assessed by consumers. Considering the revealed importance in sensory evaluation of DMC, as well as SSC, in the eating quality of sweet cherry, the use of handheld NIR devices, which are rapid, lightweight and user-friendly should be considered for routine, non-destructive analysis of sweet cherry quality.