Soil iron (Fe) performs vital functions in the biogeochemical cycles of soil environments. The amount and profile allocation of various Fe parameters can be used as sensitive indicators of soil development and pedogenic processes. This study aimed to evaluate the potential of ground-based hyperspectral imaging (HSI: 400–1010 nm) spectroscopy to predict and map six Fe parameters indicative of pedogenic processes: total Fe (Fe_t), dithionite-citrate-bicarbonate (DCB)-extracted Fe (Fe_d), oxalate-extracted Fe (Fe_o), weathering index (Fe_W), active ratio (Fe_A) and crystallinity ratio (Fe_C). In total, 17 intact soil profiles at a depth of 100 ± 5 cm were collected to acquire HSI images. Four non-linear machine learning techniques, namely, random forest (RF), XGBoost, CatBoost and support vector machine regression (SVMR), were implemented and compared with linear partial least squares (PLS) to identify the models with the best performance for different soil Fe parameters. Our results indicate that the four non-linear machine learning models outperformed PLS for most soil Fe parameters, with low root mean square error (RMSE) and high Lin's concordance correlation coefficient (LCCC) values. Based on the testing set, SVMR showed better performance over the other tested models for Fe_t (RMSE_P = 2.645 g kg⁻¹, LCCC_P = 0.89), Fe_d (RMSE_P = 0.972 g kg⁻¹, LCCC_P = 0.97), Fe_o (RMSE_P = 0.273 g kg⁻¹, LCCC_P = 0.97), Fe_W (RMSE_P = 0.035, LCCC_P = 0.97), Fe_A (RMSE_P = 0.033, LCCC_P = 0.97) and Fe_C (RMSE_P = 0.031, LCCC_P = 0.97). According to the LCCC_P values, soil Fe_t was predicted to be in substantial agreement by SVMR, and the other soil Fe predictions were considered to be in near perfect agreement. Moreover, SVMR required lower computational costs. Given these results, the combination of HSI spectroscopy and SVMR is recommended due to its more reliable estimation and profile mapping of the selected soil Fe parameters than that of PLS, RF, XGBoost and CatBoost.

中文翻译：

---

通过完整土壤剖面的高光谱成像绘制指示成土过程的土壤铁参数的机器学习算法比较

土壤铁 (Fe) 在土壤环境的生物地球化学循环中发挥着重要作用。各种Fe参数的量和分布可作为土壤发育和成土过程的敏感指标。本研究旨在评估基于地面的高光谱成像（HSI：400–1010 nm）光谱预测和绘制指示成土过程的六个 Fe 参数的潜力：总 Fe (Fe _t )、连二亚硫酸盐-柠檬酸盐-碳酸氢盐 (DCB)-萃取铁 (Fe _d )、草酸盐萃取铁 (Fe _o )、风化指数 (Fe _W )、活性比 (Fe _A ) 和结晶度比 (Fe _C）。总共收集了 17 个 100 ± 5 cm 深度的完整土壤剖面以获取 HSI 图像。实现了四种非线性机器学习技术，即随机森林 (RF)、XGBoost、CatBoost 和支持向量机回归 (SVMR)，并与线性偏最小二乘法 (PLS) 进行比较，以确定在不同情况下性能最佳的模型土壤铁参数。我们的结果表明，对于大多数土壤 Fe 参数，四种非线性机器学习模型的性能优于 PLS，均方根误差 (RMSE) 较低，Lin 的一致性相关系数 (LCCC) 值较高。基于测试集，SVMR 表现出优于其他测试模型的 Fe _t (RMSE _P  = 2.645 g kg ^-1 , LCCC _P = 0.89), Fe _d (RMSE _P  = 0.972 g kg ^-1 , LCCC _P  = 0.97), Fe _o (RMSE _P  = 0.273 g kg ^-1 , LCCC _P  = 0.97), Fe _W (RMSE _P  = 0.035, LCCC _P  = 0.97)、Fe _A (RMSE _P  = 0.033, LCCC _P  = 0.97) 和 Fe _C (RMSE _P  = 0.031, LCCC _P  = 0.97)。根据 LCCC _P值，土壤 Fe _tSVMR 预测基本一致，其他土壤 Fe 预测被认为接近完美一致。此外，SVMR 需要更低的计算成本。鉴于这些结果，建议将 HSI 光谱和 SVMR 结合使用，因为它比 PLS、RF、XGBoost 和 CatBoost 更可靠地估计和绘制所选土壤 Fe 参数的剖面图。