Clean quality control of agricultural and non-agricultural lime by rapid and accurate assessment of calcium and magnesium contents via proximal sensors.

Ca and Mg are the most important chemical elements in lime. Properly measuring Ca and Mg contents is essential to assess the quality of lime products. Quality control guarantees the adequate use of lime in industrial processes, in soils, and helps avoiding adulteration. Proximal sensors can aid in this process by determining Ca and Mg contents easily, rapidly and without producing chemical waste. The objective of this study was to evaluate the use an environmentally-friendly method of analyzing the quality of lime. We studied 1) the use of portable X-ray fluorescence (pXRF) to predict concentrations of Ca and Mg in lime, 2) tested if NixPro™ sensor can improve prediction accuracy and 3) tested if sample preparation methods (grinding) affect analyses. 74 samples of lime were analyzed by two different laboratories (lab. 1 = 38, lab. 2 = 36). All samples submitted to pXRF and NixPro™ analyses. Sensor analyses were done in whole (CP) and ground (AQ) samples to test the effect of sample preparation in prediction performance. High correlation was found between Ca and Mg contents measured via pXRF and laboratory analyses. Mg-CP presented the highest correlation coefficient (r = 0.81); Mg-AQ, the lowest (0.57). Predictions presented good performance (R2 > 0.68); Mg had the best results (0.86). Separating models per laboratory showed that some datasets are harder to model, probably due to variability in the source material (limestone). The addition of NixPro™ data contributed to improve prediction accuracy, although slightly. Predictions using CP samples presented the best results, especially for Mg, indicating that grinding is not necessary. This pioneer study demonstrated that fused proximal sensors can be used to rapidly and easily determine contents of Ca and Mg in soil amendments without producing chemical waste.

Proximal sensing provides clean, fast, and accurate quality control of organic and mineral fertilizers.

Farms use large quantities of fertilizers from many sources, making quality control a challenging task, as the traditional wet-chemistry analyses are expensive, time consuming and not environmentally-friendly. As an alternative, this work proposes the use of portable X-ray fluorescence (pXRF) spectrometry and machine learning algorithms for rapid and low-cost estimation of macro and micronutrient contents in mineral and organic fertilizers. Four machine learning algorithms were tested. Whole (i.e., as delivered by the manufacturer) (CP) and ground (AQ) samples (429 in total) were analyzed to test the effect of fertilizer granulometry in prediction performance. Model validation indicated highly accurate predictions of macro (N: R2 = 0.92; P: 0.97; K: 0.99; Ca: 0.94, Mg: 0.98; S: 0.96) and micronutrients (B: 0.99; Cu: 0.99; Fe: 0.98; Mn: 0.91; Zn: 0.94) for both organic and mineral fertilizers. RPD values ranged from 2.31 to 9.23 for AQ samples, and Random Forest and Cubist Regression were the algorithms with the best performances. Even samples analyzed as they were received from the manufacturer (i.e., no grinding) provided accurate predictions, which accelerate the confirmation of nutrient contents contained in fertilizers. Results demonstrated the potential of pXRF data coupled with machine learning algorithms to assess nutrient composition in both mineral and organic fertilizers with high accuracy, allowing for clean, fast and accurate quality control. Sensor-driven quality assessment of fertilizers improves soil and plant health, crop management efficiency and food security with a reduced environmental footprint.