Deep Learning Optimizes Data-Driven Representation of Soil Organic Carbon in Earth System Model Over the Conterminous United States.

Feng Tao,Yiqi Luo,Zhehao Ren,Yishuang Liang,Xiaomeng Huang,Yuanyuan Huang,Russell Doughty,Gustaf Hugelius,Zhenghu Zhou,Shuang Ma,Xingjie Lu,Qianyu Li,Lifen Jiang

doi:10.3389/fdata.2020.00017

Abstract

Soil organic carbon (SOC) is a key component of the global carbon cycle, yet it is not well-represented in Earth system models to accurately predict global carbon dynamics in response to climate change. This novel study integrated deep learning, data assimilation, 25,444 vertical soil profiles, and the Community Land Model version 5 (CLM5) to optimize the model representation of SOC over the conterminous United States. We firstly constrained parameters in CLM5 using observations of vertical profiles of SOC in both a batch mode (using all individual soil layers in one batch) and at individual sites (site-by-site). The estimated parameter values from the site-by-site data assimilation were then either randomly sampled (random-sampling) to generate continentally homogeneous (constant) parameter values or maximally preserved for their spatially heterogeneous distributions (varying parameter values to match the spatial patterns from the site-by-site data assimilation) so as to optimize spatial representation of SOC in CLM5 through a deep learning technique (neural networking) over the conterminous United States. Comparing modeled spatial distributions of SOC by CLM5 to observations yielded increasing predictive accuracy from default CLM5 settings (R2 = 0.32) to randomly sampled (0.36), one-batch estimated (0.43), and deep learning optimized (0.62) parameter values. While CLM5 with parameter values derived from random-sampling and one-batch methods substantially corrected the overestimated SOC storage by that with default model parameters, there were still considerable geographical biases. CLM5 with the spatially heterogeneous parameter values optimized from the neural networking method had the least estimation error and less geographical biases across the conterminous United States. Our study indicated that deep learning in combination with data assimilation can significantly improve the representation of SOC by complex land biogeochemical models.

Highlights

Soil is a vast carbon reservoir in terrestrial ecosystems
We have demonstrated that massive observational datasets can be assimilated into Earth system models (ESMs)
Bayesian data assimilation has been applied to integration of observations with process-oriented models to improve simulation performance at individual sites (Xu et al, 2006; Li et al, 2016), the complexity of ESMs and the computational cost in the Markov Chain Monte Carlo (MCMC) process hindered the progress of using big data to inform models at a continental or global scale

Summary

Introduction

Soil is a vast carbon reservoir in terrestrial ecosystems. It stores more than three times as much organic carbon as terrestrial vegetation does (Ciais et al, 2014). Due to its large reserve, a small change in soil organic carbon (SOC) potentially results in strong regulation of the global carbon cycle and its feedbacks to climate change (Friedlingstein et al, 2006; Luo et al, 2015). Soil carbon dynamics simulated by ESMs are highly variable and fit poorly with observations (Luo et al, 2015). Modeled global soil carbon storage differs by up to 6-fold among 11 models in the Coupled Model Intercomparison Project phase 5 (CMIP5) (Todd-Brown et al, 2013).

Methods

Results

Conclusion