当前位置: X-MOL 学术Geoderma › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Assessing countrywide soil organic carbon stock using hybrid machine learning modelling and legacy soil data in Cameroon
Geoderma ( IF 6.1 ) Pub Date : 2020-05-01 , DOI: 10.1016/j.geoderma.2020.114260
Francis B.T. Silatsa , Martin Yemefack , Fritz O. Tabi , Gerard B.M. Heuvelink , Johan G.B. Leenaars

Abstract Countrywide estimates of soil organic carbon stock (SOCS) are useful to set up national strategies for sustainable land use management as well as to enhance the accuracy of global SOCS inventories. We appraised the spatial distribution of SOCS at five depth layers (0–15 cm, 15–30 cm, 30–100 cm, 0–30 cm and 0–100 cm) in Cameroon at 100 m spatial resolution, using a national harmonized legacy soil database (Camsodat 0.1) with 1432 georeferenced soil profiles. We assessed the prediction performances of random forest (RF) and generalized boosted regression (GBR), combined with two hybridization approaches of spatial interpolation of the residuals using ordinary kriging (OK) and inverse distance weighting (IDW). The estimates were compared to two global estimates derived from the Harmonized World Soil Database (HWSD) and SoilGrids250m. The SOCS distribution across the country showed a moderate spatial heterogeneity at all depth layers with coefficients of variation between 35% and 47%, and values ranging from 6 to 108 Mg C ha−1 at 0–15 cm, from 4 to 107 Mg C ha−1 at 15–30 cm, from 10 to 276 Mg C ha−1 at 30–100 cm, from 11 to 210 Mg C ha−1 at 0–30 cm and from 21 to 468 Mg C ha−1 at the 0–100 cm layer. Of the selected environmental covariates, terrain and climate attributes were the most relevant to predict the SOCS spatial distribution at country level. The RF model outperformed the GBR model, with about 10% improvement on prediction performance (R2) for most soil depths. The hybridization further slightly improved performance. However, OK was only slightly better than IDW in the overall assessment. Compared to national estimates, SoilGrids overestimated the SOCS by 15% at 0–30 cm depth, while HWSD underestimated SOCS by 26% at the same depth. Overall, about 5.7 Pg C are stored in the top 1 m of soils in Cameroon, with about 50% of that in the top 30 cm. The national distribution of SOCS is consistent with the pattern of agro-ecological zones. Our assessment provides baseline information for sustainable land management and climate change mitigation, as well as for improving the understanding of the spatial distribution of SOCS in Cameroon.

中文翻译:

在喀麦隆使用混合机器学习模型和遗留土壤数据评估全国土壤有机碳储量

摘要 全国土壤有机碳储量 (SOCS) 的估算有助于制定可持续土地利用管理的国家战略以及提高全球 SOCS 清单的准确性。我们使用国家统一遗产评估了喀麦隆五个深度层(0–15 cm、15–30 cm、30–100 cm、0–30 cm 和 0–100 cm)的 SOCS 空间分布,空间分辨率为 100 m土壤数据库 (Camsodat 0.1) 具有 1432 个地理参考土壤剖面。我们评估了随机森林 (RF) 和广义增强回归 (GBR) 的预测性能,并结合使用普通克里金法 (OK) 和逆距离加权 (IDW) 的残差空间插值两种混合方法。这些估计值与来自协调世界土壤数据库 (HWSD) 和 SoilGrids250m 的两个全球估计值进行了比较。全国 SOCS 分布在所有深度层均表现出中等空间异质性,变异系数在 35% 至 47% 之间,值范围为 6 至 108 Mg C ha-1,0-15 cm,4 至 107 Mg C ha−1 在 15–30 cm,从 10 到 276 Mg C ha−1,在 30–100 cm,从 11 到 210 Mg C ha−1,在 0–30 cm,从 21 到 468 Mg C ha−1,在0-100 厘米层。在选定的环境协变量中,地形和气候属性与预测国家级 SOCS 空间分布最相关。RF 模型优于 GBR 模型,对大多数土壤深度的预测性能 (R2) 提高了约 10%。杂交进一步略微提高了性能。然而,OK 在整体评估中仅略好于 IDW。与国家估计相比,SoilGrids 将 0-30 cm 深度的 SOCS 高估了 15%,而 HWSD 将相同深度的 SOCS 低估了 26%。总体而言,喀麦隆土壤表层 1 m 中储存了约 5.7 Pg C,其中约 50% 储存在表层 30 cm 中。SOCS的全国分布与农业生态区格局一致。我们的评估为可持续土地管理和减缓气候变化提供了基线信息,以及提高对喀麦隆 SOCS 空间分布的了解。
更新日期:2020-05-01
down
wechat
bug