Phosphorus (P) deficiency is a major constraint for rice production in the tropics. Field-specific P management is key for resource-limited farmers to increase yields with minimal inputs. We used soil P fractionation analysis to identify the relevant factors controlling P uptake and the responses to P fertilization of rice in flooded and highly weathered soils.

As a proximal soil sensing technique, laboratory visible and near-infrared (Vis-NIR) spectroscopy is a promising tool for the quantitative estimation of soil properties. However, there remain challenges for predicting soil phosphorus (P) content and availability, which requires a reliable model applicable for different land-use systems to upscale. Recently, a one-dimensional convolutional neural network (1D-CNN) corresponding to the spectral information of soil was developed to considerably improve the accuracy of soil property predictions. The present study investigated the predictive ability of a 1D-CNN model to estimate soil available P (oxalate-extractable P; Pox) content in soils by comparing it with partial least squares (PLS) and random forest (RF) regressions using soil samples (n = 318) collected from natural (forest and non-forest) and cultivated (upland and flooded rice fields) systems in Madagascar. Overall, the 1D-CNN model showed the best predictive accuracy (R² = 0.878) with a highly accurate prediction ability (ratio of performance to the interquartile range = 2.492). Compared to the PLS model, the RF and 1D-CNN models indicated 4.37% and 23.77% relative improvement in root mean squared error values, respectively. Based on a sensitivity analysis, the important wavebands for predicting soil Pox were associated with iron (Fe) oxide, organic matter (OM), and water absorption, which were previously known wavelength regions for estimating P in soil. These results suggest that 1D-CNN corresponding spectral signatures can be expected to significantly improve the predictive ability for estimating soil available P (Pox) from Vis-NIR spectral data. Rapid and accurate estimation of available P content in soils using our results can be expected to contribute to effective fertilizer management in agriculture and the sustainable management of ecosystems. However, the 1D-CNN model will require a large dataset to extend its applicability to other regions of Madagascar. Thus, further updates should be tested in future studies using larger datasets from a wide range of ecosystems in the tropics.

As a laboratory proximal sensing technique, the capability of visible and near-infrared (Vis-NIR) diffused reflectance spectroscopy with partial least squares (PLS) regression to determine soil properties has previously been demonstrated. However, the evaluation of the soil phosphorus (P) content—a major nutrient constraint for crop production in the tropics—is still a challenging task. PLS regression with waveband selection can improve the predictive ability of a calibration model, and a genetic algorithm (GA) has been widely applied as a suitable method for selecting wavebands in laboratory calibrations. To develop a laboratory-based proximal sensing method, this study investigated the potential to use GA-PLS regression analyses to estimate oxalate-extractable P in upland and lowland soils from laboratory Vis-NIR reflectance data. In terms of predictive ability, GA-PLS regression was compared with iterative stepwise elimination PLS (ISE-PLS) regression and standard full-spectrum PLS (FS-PLS) regression using soil samples collected in 2015 and 2016 from the surface of upland and lowland rice fields in Madagascar (n = 103). Overall, the GA-PLS model using first derivative reflectance (FDR) had the best predictive accuracy (R² = 0.796) with a good prediction ability (residual predictive deviation (RPD) = 2.211). Selected wavebands in the GA-PLS model did not perfectly match wavelengths of previously known absorption features of soil nutrients, but in most cases, the selected wavebands were within 20 nm of previously known wavelength regions. Bootstrap procedures (N = 10,000 times) using selected wavebands also confirmed the improvements in accuracy and robustness of the GA-PLS model compared to those of the ISE-PLS and FS-PLS models. These results suggest that soil oxalate-extractable P can be predicted from Vis-NIR spectroscopy and that GA-PLS regression has the advantage of tuning optimum bands for PLS regression, contributing to a better predictive ability.

Soil phosphorus (P) forms have been practically defined as chemically fractionated pools. A knowledge of the abundance and diversity of P forms in soil, and the factors affecting them, will lead to better soil management. However, little is known about the differences in P forms among soils with different geological properties in tropical Africa. The aim of this study was to investigate the P forms in soils with different physicochemical properties formed under different geological conditions in southern Tanzania and to identify the factors affecting the P forms in these soils. In total, 37 surface soil samples were collected from three geological groups; the plutonic (mainly granite) rock (PL) group, the sedimentary and metamorphic rock (SM) group, and the volcanic ash (V) group. Soil P was sequentially extracted by NH₄Cl, NH₄F, NaHCO₃, NaOH + NaCl, and HCl, and inorganic (P_i) and organic P (P_o) in each fraction were determined. The lowest total P was in the PL group (average, 360 mg P kg^-1) because of the high sand content. Iron (Fe)-P (NaOH-P_i) was the major form in this group, accounting for 8.4% of total P. In the SM group (average total P, 860 mg P kg^-1), Fe-P was the major form in most, accounting for 7.8% of total P. Soils in the SM group occasionally had high calcium (Ca)-P due to application of chemical fertilizer at the collection site. The V group had the highest total P (average, 1600 mg P kg^-1) and its major P form was Ca-P, which was possibly derived from primary minerals (i.e., apatite), accounting for 14% of total P. In addition, the high oxalate-extractable Al possibly caused the accumulation of Al-P in the V group. Oxalate-extractable Fe generally increased with increasing Fe-P_i, while oxalate-extractable Al increased with increasing organic P and Al-P_i in soils in all three geological groups. These results demonstrate that the soil P forms differ greatly among sites in southern Tanzania with different geological conditions and associated soil properties.

Applied phosphorus (P) use efficiency is generally low due to the low mobility of P in soil and its affinity to form insoluble complexes. Localized P application nearby the root zone is a potential approach to overcome this issue in crop production. However, the interaction with soil conditions is little understood, which results in less effective application of this approach. Using root-box experiments and changing P-retention capacity of soils, we revealed that applied P use efficiency of rice can be substantially improved by dipping seedlings in P-enriched slurry at transplanting (P-dipping) even in highly P-fixing soils. Spatial analysis of soluble P in soils indicated that P-dipping creates a P hotspot because the P-enriched slurry is transferred with seedling roots. The P hotspot could have induced vigorous surface root and facilitated further P uptake from the spot. In contrast, the effect of conventional P incorporation depended on P-retention capacity of soils; no increases in soluble P content in soils or plant P uptakes were observed when P-retention capacity was high. Our finding of significant interaction between localized P application and a specific soil property should help improving applied P use efficiency and achieving sustainable rice production against depleting P fertilizer resources.