Carbon stock map and texture map for CAPV soils.

Maps obtained as part of the LURCARBONTEXT project (KLIMATEK call 2017). Sand, silt and clay maps have been obtained based on 6,736 georeferenced soil analyses. Since the texture is expressed as % of the three classes (sand, clay and silt) and its sum must be 100 %, the additive log-ratio transformation proposed by Aitchison (1982) has been carried out for compositional data. In this way, the adjusted model chosen will consist of a multivariate response model for the transformed sand and clay variables, and the third (lime) will be the result of the difference between 100 % and the sum of the other two. The statistical model used is the Generalised Additive Models (GAM) subject to a softening penalty.The modelling of the organic carbon content in the first 30 cm of mineral soils (Mg C ha-1) of the Basque Country was made with 12,122 georeferenced analytical samples.本数据集的土壤质地分布图源自LURCARBONTEXT项目（KLIMATEK 2017年度资助征集项目）。基于6736个带地理坐标的土壤分析样本，生成了砂粒、粉粒与黏粒的含量分布图。由于土壤质地以三类组分（砂粒、黏粒、粉粒）的百分比形式表征，且三者之和恒为100%，因此针对该成分数据采用了Aitchison（1982）提出的加性对数比变换。据此，最终选用的校正模型包含针对经变换后的砂粒与黏粒变量的多响应回归模型，第三组分（原文标注为“石灰”，结合土壤组分逻辑应为粉粒）的含量可通过100%减去其余两组分之和得到。所用统计模型为带平滑惩罚项的广义加性模型（Generalised Additive Models，GAM）。本数据集基于12122个带地理坐标的分析样本，对巴斯克地区0~30 cm矿质土壤的有机碳含量（单位：Mg C ha⁻¹）开展了建模工作。