Peng Wang

The openness and sensitivity of karst water systems make the geochemical processes of groundwater in karst areas of great importance. In this study, the recharge processes have been identified based on physical-chemical indicators (T, DO, EC, HCO3−, SO42−, Cl-, Na+, Ca2+, and Mg2+), δ2H, δ18O, and δ13CDIC isotopes. The results are as follows: (1) Ions exchange is active in karst groundwater, particularly during the dry season. (2) Evaporation and precipitation are significant but seasonally varied processes governing the recharge of the Lihu underground river. Spring and summer show significant groundwater recharge driven by rainfall, with intense evaporation in summer and the isotopic and d-excess values in autumn reflecting the reduced rainfall and cooler weather. Despite the combined influence of surface water and rainfall in winter, low precipitation produces predominantly old groundwater, resulting in pronounced evaporative effects on groundwater isotopic composition. (3) δ13CDIC values are generally heavy (-10.74‰ to -4.34‰) in the underground water, indicating rapid surface water infiltration with short residence times and abbreviated karstification processes in summer versus longer residence times and prolonged karstification in winter. This study highlights chemical and isotope analysis to understand karst recharge and hydrodynamics in Southwest China and beyond.

Hyperspectral analysis is useful in monitoring and evaluating soil heavy metal pollution. The difficulty in estimation of heavy metals by hyperspectral analysis is to extract the spectral features related to soil heavy metal concentration which may be interfered by signals resulting from other soil components. In this study, we established a comprehensive framework for prediction of Cr and Ni from hyperspectral data combined with soil Al-Fe minerals using advanced machine learning methods. The Al-Fe minerals were measured in laboratory and predicted from the spectral data, respectively. The results show that (1) from the original hyperspectral data, the model performances for Cr (R² = 0.61, RMSE = 16.96) and Ni (R² = 0.34, RMSE = 6.92) are not satisfactory. (2) With the incorporation of the measured Al-Fe minerals as the predictors, the optimal machine learning method for prediction of Cr and Ni is XgBoost, and the model performances are improved obviously (Cr: R² = 0.85, RMSE = 10.67; Ni: R² = 0.71, RMSE = 4.55). (3) With the incorporation of the Al-Fe minerals predicted from the spectral data, the optimal machine learning method for prediction of Cr and Ni is GdBoost, and the model performances are acceptable (Cr: R² = 0.74, RMSE = 13.93; Ni: R² = 0.68, RMSE = 4.79). This study provides a preliminary analysis for mapping soil heavy metal content and monitoring heavy metal pollution over a large spatial extent.