Partitioning Global Surface Energy and Their Controlling Factors Based on Machine Learning

Abstract

As two competitive pathways of surface energy partitioning, latent (LE) and sensible (H) heat fluxes are expected to be strongly influenced by climate change and wide spread of global greening in recent several decades. Quantifying the surface energy fluxes (i.e., LE and H) variations and controlling factors is still a challenge because of the discrepancy in existing models, parameterizations, as well as driven datasets. In this study, we assessed the ability of random forest (RF, a machine learning method) and further predicted the global surface energy fluxes (i.e., LE and H) by combining FLUXNET observations, climate reanalysis and satellite-based leaf area index (LAI). The results show that the surface energy fluxes variations can be highly explained by the established RF models. The coefficient of determination (R2) ranges from 0.66 to 0.89 for the LE, and from 0.53 to 0.90 for the H across 10 plant functional types (PFTs), respectively. Meanwhile, the root mean square error (RMSE) ranges from 12.20 W/m2 to 21.94 W/m2 for the LE and from 12.05 W/m2 to 22.34 W/m2 for the H at a monthly scale, respectively. The important influencing factors in building RF models are divergent with respect to LE and H, but the solar radiation is common to both LE and H and to all 10 PFTs in this study. We also found a contrasting trend of LE and H: a positive trend in LE and a negative trend in H during 1982–2016 and these contrasting trends are dominated by the elevated CO2 concentration level. Our study suggested an important role of the CO2 concentration in determining surface energy partitioning which is needed to be considered in future studies.

Machine learning and internet of things applications in enterprise architectures: Solutions, challenges, and open issues

“…A computer program is said to learn about a job T and a metric of success P from an E experience (Mitchell (1997)). While ML originated in computer science, numerous methods of vector quantification (Zou et al (2021)) have been introduced for telecommunications and data transmission (Li & Yuan et al (2020)), encoding and compression (Liu, Yao, and Du (2021), Spanias et al (2006)). In computer science and data science, learning is focused on examples (data samples) and practice.…”

“…Figure 1 presents an overview of ML significance and role in IoT-based EA. The most paradigmatic algorithms in ML have been used in many areas, including decisionmaking (Zhao et al (2020), Seyedzadeh et al (2020), Valdivia et al (2021)) and intelligent control (Matei et al (2021), Yuan et al (2020), Tofighbakhsh (2020)), computer graphics (Zoni (2020), Janai et al (2020), Arya et al (2020)), voice recognition (Honnavalli and Shylaja (2021), Ku smierczyk et al (2020), Shankar et al (2020)), Natural Language Processing (NLP) (Moon et al (2021), Van Rousselt (2021), Prudhvi et al (2021)), and computer vision (Tien et al (2021), Willman (2021), Falk et al (2020)). Likewise, ML will also give computer networks a potential advantage.…”