A super learner ensemble to map potassium fixation in California vineyard soils

Abstract

Potassium (K) deficiency in wine grapes results in reduced vine growth, premature leaf drop, and yield and color loss. K can be fixed in the interlayer of clay minerals in a process called K fixation, which leads to high spatial variability in soil K. In the Lodi American Viticulture Area (AVA) management of winegrapes is complicated by a mix of K fixing soils and non-K fixing soils. Here, we leverage a digital soil mapping (DSM) framework to identify the spatial distribution of K fixation and availability, to disentangle the complexity of K management in the region. Soil samples (n = 107) were collected, analyzed for the K fixation index, K availability and cation exchange capacity (CEC), and aggregated into two depths (0–30 cm and 30–100 cm). Soil samples were intersected with remotely sensed proxies for the soil forming factors and existing soil survey data and used to train a “super learner” ensemble or combination of base models, including random forest (RF), extreme gradient boosting (XGB) and cubist. Base models were combined via model averaging (each model weighted by its R2) or model stacking (linear combination of base models via OLS regression), and model performance was compared. We generated mapped uncertainties from a super learning framework by utilizing bootstrapped realizations of each base model and weighting each bootstrapped base model map via the β-coefficients generated in the ensemble fitting step. Bootstrapped maps of the super learner were utilized to generate upper and lower 90% prediction limits.For the K fixation index at the 0–30 cm depth, RF outperformed other models (R2 = 0.42), whereas a linear combination of all base models performed best in the 30–100 cm depth (R2 = 0.41). Results improved for K availability in the 0–30 cm horizons (R2 = 0.48) and the 30–100 cm horizons (R2 = 0.46). Overall, predictions for CEC were superior to both the K fixation index and K availability at both depths (0–30 cm; R2 = 0.71) and (30–100 cm; R2 = 0.51). We conclude that K fixation and availability can be predictively mapped with marginal success, while CEC is amenable to a DSM framework. CEC is tied more to soil genesis and formation, while K fixation and availability are affected by K fertilization. Finally, we compared the DSM K fixation index map to an existing soil landscape model K fixation index map, to facilitate discussion of the connection between pedogenic state factors, soil forming processes and soil properties in soil mapping. Results inform the pedogenic foundations of DSM, as well as global efforts to utilize DSM to map K fixing soils and manage K-based crop interventions.