时间调整方法用于对土壤有机碳的高分辨率大陆规模建模
Laxman Bokati1, Anil Somenahally2,3, Saurav Kumar4
1School of Sustainable Engineering and Built Environment, Arizona State University, 777 E University Dr., 85287, AZ, Tempe, USA.
Scientific reports
|February 22, 2025
概括
这项研究引入了一种新的方法来调整土壤有机碳 (SOC) 数据以适应时间变化,提高机器学习模型的准确性. 这种方法提高了SOC股票预测,有利于碳激励计划.
科学领域:
- 环境科学 环境科学
- 土壤科学 土壤科学
- 数据科学数据科学数据科学
背景情况:
- 有限且分布不均的开源遗留数据阻碍了土壤有机碳 (SOC) 模型培训.
- 大规模数据驱动的SOC建模往往忽略了时间变化,影响了准确性.
- SOC数据的时间漂移需要为有效的机器学习应用程序进行调整.
研究的目的:
- 开发和评估一种创新方法,用于创建基于近距离的距离矩阵.
- 为调整 SOC 观测生成空间解析的时间转移预测.
- 通过考虑时间动态来提高SOC股票预测的准确性.
主要方法:
- 利用来自连续美国 (CONUS) 的遗留数据来构建基于近距离的距离矩阵.
- 生成空间解析的时间转移预测,以调整SOC观测到目标日期.
- 对1980年和2020年参考年进行时间调整和没有时间调整的SOC库存估计进行了比较.
主要成果:
- 在有时间调整和没有时间调整的SOC股票预测之间观察到显著差异 (SOCno-adj与SOC2020).
- 在CONUS农田的基线SOC库存估计高于SOCno-adj (14.49 Pg C),而不是SOC2020 (13.29 Pg C).
- 牧场土地显示较高的SOCno-adj (15.49 Pg C) 与SOC2020 (14.22 Pg C) 相比,而森林土地显示SOCno-adj (39.52 Pg C) 与SOC2020 (40.83 Pg C) 相比.
结论:
- 该研究证实了提高SOC库存预测的拟议方法论的有效性和有效性.
- 时间调整方法显著提高了用于机器学习模型的SOC数据集的准确性.
- 结果对碳激励计划的利益相关者来说是有价值的,包括农民,科学家,政策制定者和工业.
更多相关视频
07:32Monitoring Pedogenic Inorganic Carbon Accumulation Due to Weathering of Amended Silicate Minerals in Agricultural Soils.
Published on: June 4, 2021
5.1K
08:09Measuring and Mapping Patterns of Soil Erosion and Deposition Related to Soil Carbonate Concentrations Under Agricultural Management
Published on: September 12, 2017
11.7K
相关概念视频
The Carbon Cycle
36.9K
Carbon is the basis of all organic matter on Earth, and is recycled through the ecosystem in two primary processes: one in which carbon is exchanged among living organisms, and one in which carbon is cycled over long periods of time through fossilized organic remains, weathering of rocks, and volcanic activity. Human activities, including increased agricultural practices and the burning of fossil fuels, has greatly affected the balance of the natural carbon cycle.
36.9K
Global Climate Change
24.1K
Throughout its ~4.5 billion year history, the Earth has experienced periods of warming and cooling. However, the current drastic increase in global temperatures is well outside of the Earth’s cyclic norms, and evidence for human-caused global climate change is compelling. Paleoclimatology, the study of ancient climate conditions, provides ample evidence for human-caused global climate change by comparing recent conditions with those in the past.
24.1K
