Author
LA, WOO JUNG - Gyeongsang National University | |
Sudduth, Kenneth - Ken | |
KIM, HAK-JIN - Seoul National University | |
CHUNG, SUN-OK - Chungnam National University |
Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 4/26/2016 Publication Date: 8/15/2016 Citation: La, W., Sudduth, K.A., Kim, H., Chung, S. 2016. Fusion of spectral and electrochemical sensor data for estimating soil macronutrients. Transactions of the ASABE. 59(4):787-794. doi: 10.13031/trans.59.11562. Interpretive Summary: Measuring the variation in soil properties within fields is an important component of precision agriculture. For many soil properties, it is difficult to obtain enough data to accurately characterize their spatial variation, due to the cost of traditional sampling and laboratory analysis. Sensors that can estimate soil properties without the need for sampling are a promising alternative. One technology that has received considerable attention in this regard is optical reflectance sensing in the visible and near infrared wavelength bands. Another technology for which prototype systems have been developed is rapid electrochemical analysis using ion-selective electrodes (ISE). In a previous study, we developed an ISE system and evaluated its performance compared to standard laboratory methods using soil samples from Missouri and Illinois. Our goal in this study was to apply reflectance sensing to the same samples and evaluate the results in comparison to the ISE results and standard methods. We found that optical reflectance sensing worked well to estimate a number of soil physical properties, but did not provide good results for chemical properties – soil pH, phosphorus (P) and potassium (K). However, when we combined reflectance sensing data with data from our prototype ISE system, we obtained very good estimates of P and K. This study shows that combining the outputs of multiple sensors, sometimes called “sensor fusion”, has potential for improving soil chemical property estimates, and should be investigated further. If proven, this combination approach has the potential to benefit producers by providing them with a rapid, accurate method of quantifying variation in soil chemical properties, as needed for fertility management in precision agriculture. Technical Abstract: Rapid and efficient quantification of plant-available soil phosphorus (P) and potassium (K) is needed to support variable-rate fertilization strategies. Two methods that have been used for estimating these soil macronutrients are diffuse reflectance spectroscopy in visible and near-infrared (VNIR) wavelength ranges and electrochemical sensing using an automated ion-selective electrode (ISE) system. The goal of this research was to compare P and K estimates with VNIR spectroscopy, ISE analysis, and a sensor fusion approach that combined the two methods. Diffuse reflectance spectra of air-dried, sieved samples were obtained in the laboratory. Estimates of P and K based on ISE measurements and standard laboratory analysis methods were available for the same samples from a previous study. Calibrations relating spectra to P, K and other soil properties determined by standard laboratory methods were developed using partial least squares regression. Although good estimates (R2 = 0.83 to 0.92) were obtained for soil texture fractions, organic matter, and CEC, estimates of pH, P, and K were not good (R2 < 0.7). Accuracy of ISE-based P and K estimates was better (R2 = 0.87), but both P and K were systematically under-estimated by over 40%. Including both spectral and ISE information in P and K calibrations improved the accuracy (R2 = 0.93), and reduced underestimation to 7% or less, a considerable improvement over ISE data alone. In P and K models including ISE data and spectrally-estimated soil properties, significant variables included texture, organic matter, and cation exchange capacity, suggesting that these soil properties may affect the relationship of ISE-measured P and K to values determined by standard laboratory methods. These improved results obtained with a sensor fusion approach using laboratory-collected data are promising and warrant further investigation with additional datasets. |