干旱区生态环境知识资源中心(Arid): Reconstructing Past Global Vegetation With Random Forest Machine Learning, Sacrificing the Dynamic Response for Robust Results

Arid

DOI	10.1029/2020MS002200
	Reconstructing Past Global Vegetation With Random Forest Machine Learning, Sacrificing the Dynamic Response for Robust Results
	Lindgren, Amelie; Lu, Zhengyao; Zhang, Qiong; Hugelius, Gustaf
通讯作者	Lindgren, A (corresponding author), Stockholm Univ, Dept Phys Geog, Stockholm, Sweden. ; Lindgren, A (corresponding author), Bolin Ctr Climate Res, Stockholm, Sweden. ; Lu, ZY (corresponding author), Lund Univ, Ctr Environm & Climate Res, Lund, Sweden. ; Lu, ZY (corresponding author), Lund Univ, Dept Phys Geog & Ecosyst Sci, Lund, Sweden.
来源期刊	JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS
EISSN	1942-2466
出版年	2021
卷号	13 期号:2
英文摘要	Vegetation is an important component in the Earth system, providing a direct link between the biosphere and atmosphere. As such, a representative vegetation pattern is needed to accurately simulate climate. We attempt to model global vegetation (biomes) with a data-driven approach, to test if this allows us to create robust global and regional vegetation patterns. This not only provides quantitative reconstructions of past vegetation cover as a climate forcing, but also improves our understanding of past land cover-climate interactions which have important implications for the future. By using a Random Forest (RF) machine learning tool, we train the vegetation reconstruction with available biomized pollen data of present and past conditions to produce broad-scale vegetation patterns for the preindustrial (PI), the mid-Holocene (MH, similar to 6,000 years ago), and the Last Glacial Maximum (LGM, similar to 21,000 years ago). We test the method's robustness by introducing a systematic temperature bias based on existing climate model spread and compare the result with that of LPJ-GUESS, an individual-based dynamic global vegetation model. The results show that the RF approach is able to produce robust patterns for periods and regions well constrained by evidence (the PI and the MH), but fails when evidence is scarce (the LGM). The apparent robustness of this method is achieved at the cost of sacrificing the ability to model dynamic vegetation response to a changing climate. Plain Language Summary This study introduces a simple but robust method to create global vegetation patterns. The distribution of forests, grasslands, deserts, and other vegetation types are important boundary conditions to model climate, because they have different albedo (affecting short-wave radiation), shape and height (affecting friction), evapotranspiration effects (affecting humidity and latent heat flux), and photosynthesis features (affecting atmospheric CO2 concentration). Our method uses artificial intelligence to produce these patterns using known examples of vegetation and modeled climate together. A great advantage of this method to reconstruct vegetation patterns compared to other models involving empirical or realistic physical and biological processes is that it can be less affected by biases in the climate forcing. For instance, because the model does not know the correct relationships beforehand, it is possible to introduce an error in the climate data and still produce the same vegetation patterns. We perturb the temperature, which drives vegetation changes in a complex dynamic vegetation model, but produces almost no change in our simple method. We show that it is possible to construct robust vegetation maps despite uncertainties in the climate forcing. However, the method needs more evidence than what is currently available to reconstruct vegetation for the last ice age.
类型	Article
语种	英语
开放获取类型	gold
收录类别	SCI-E
WOS记录号	WOS:000623792200012
WOS类目	Meteorology & Atmospheric Sciences
WOS研究方向	Meteorology & Atmospheric Sciences
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/350632
作者单位	[Lindgren, Amelie; Zhang, Qiong; Hugelius, Gustaf] Stockholm Univ, Dept Phys Geog, Stockholm, Sweden; [Lindgren, Amelie; Zhang, Qiong; Hugelius, Gustaf] Bolin Ctr Climate Res, Stockholm, Sweden; [Lu, Zhengyao] Lund Univ, Ctr Environm & Climate Res, Lund, Sweden; [Lu, Zhengyao] Lund Univ, Dept Phys Geog & Ecosyst Sci, Lund, Sweden
推荐引用方式 GB/T 7714	Lindgren, Amelie,Lu, Zhengyao,Zhang, Qiong,et al. Reconstructing Past Global Vegetation With Random Forest Machine Learning, Sacrificing the Dynamic Response for Robust Results[J],2021,13(2).
APA	Lindgren, Amelie,Lu, Zhengyao,Zhang, Qiong,&Hugelius, Gustaf.(2021).Reconstructing Past Global Vegetation With Random Forest Machine Learning, Sacrificing the Dynamic Response for Robust Results.JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS,13(2).
MLA	Lindgren, Amelie,et al."Reconstructing Past Global Vegetation With Random Forest Machine Learning, Sacrificing the Dynamic Response for Robust Results".JOURNAL OF ADVANCES IN MODELING EARTH SYSTEMS 13.2(2021).