Abstract. Information on soils' composition and physical, chemical and biological properties is paramount to elucidate agroecosystem functioning in space and over time. For this purposes we developed a national Swiss soil spectral library (SSL; n = 4374) in the mid-infrared (mid-IR), calibrating 17 properties from legacy measurements on soils from the Swiss biodiversity monitoring program (n = 3778; 1094 sites) and the Swiss long-term monitoring network (n = 596; 71 sites). General models were trained with the interpretable rule-based learner CUBIST, testing combinations of {5, 10, 20, 50, 100} committees of rules and {2, 5, 7, 9} neighbors to localize predictions with repeated by location grouped ten-fold cross-validation. To evaluate the information in spectra to facilitate long-term soil monitoring at a plot-level, we conducted 71 model transfers for the NABO sites to induce locally relevant information from the SSL, using the data-driven sample selection method rs-local. Eleven soil properties were estimated with discrimination capacity suitable for screening (R² > 0.6), out of which total carbon (C), organic C (OC), total N, organic matter content, pH, and clay showed accuracy eligible for accurate diagnostics (R² > 0.8). Cubist and the spectra estimated total C accurately with RMSE = 0.84 % while the measured range was 0.1–⁠58.3 %, and OC with RMSE = 1.20 % (measured range 0.0–⁠27.3 %). Compared to general estimates of properties from Cubist, local modeling on average reduced the root mean square error of total C per site fourfold. We found that the selected SSL subsets were highly dissimilar in terms of both their spectral input space and the measured values. This suggests that data-driven selection with RS-LOCAL leverages chemical diversity in composition rather than similarity. Our results suggest that mid-IR soil estimates were sufficiently accurate to support many soil applications that require a large volume of input data, such as precision agriculture, soil C accounting and monitoring, and digital soil mapping. This SSL can be updated continuously, for example with samples from deeper profiles and organic soils, so that the measurement of key soil properties becomes even more accurate and efficient in the near future.

中文翻译：

---

开发瑞士土壤光谱库以进行本地估计和监测

摘要。有关土壤成分，物理，化学和生物学特性的信息对于阐明空间和时间上的农业生态系统功能至关重要。为此，我们开发了国家 红外土壤光谱库（SSL；n = 4374），用于中红外（mid-IR），校准了来自瑞士生物多样性监测计划对土壤进行的遗留测量的17个属性（n  = 3778； 1094个站点） ）和瑞士的长期监控网络（n = 596; 71个网站）。通用模型由可解释的基于规则的学习者CUBIST训练，测试{5、10、20、50、100}规则委员会和{2、5、7、9}邻居的组合以定位预测，并按位置分组十个重复位置折交叉验证。为了评估光谱中的信息以促进在样地一级进行长期土壤监测，我们使用数据驱动的样本选择方法rs-local对NABO站点进行了71种模型转换，以从SSL导出本地相关信息。估计了11种土壤特性，并具有适合筛选的判别能力（R ²  > 0.6），其中总碳（C），有机碳（OC），总氮，有机质含量，pH和黏土显示出可以进行准确诊断的准确度（R ² > 0.8）。Cubist和光谱在RMSE = 0.84％的情况下准确地估算了总C，而测量范围为0.1–58.3％，而RMSE = 1.20％的OC（测量范围为0.0-27.3％）。与来自Cubist的属性的一般估计相比，局部建模平均将每个站点的总C的均方根误差减小了四倍。我们发现，所选的SSL子集在其频谱输入空间和测量值方面都极为不同。这表明使用RS-LOCAL进行数据驱动的选择会利用成分的化学多样性而不是相似性。我们的结果表明，中红外土壤估算值足够准确，可以支持许多需要大量输入数据的土壤应用，例如精密农业，土壤碳核算和监测以及数字土壤测绘。