Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

A differentiable, physics-informed ecosystem modeling and learning framework for large-scale inverse problems: demonstration with photosynthesis simulations

Journal Article · · Biogeosciences (Online)
 [1];  [2];  [3];  [1];  [4];  [5];  [1];  [1]
  1. Pennsylvania State University, University Park, PA (United States)
  2. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
  3. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States); University of Tennessee, Knoxville, TN (United States)
  4. SciML, Open Source Software Organization, Cambridge, MA (United States)
  5. Massachusetts Institute of Technology (MIT), Cambridge, MA (United States)
+ Show Author Affiliations
Photosynthesis plays an important role in carbon, nitrogen, and water cycles. Ecosystem models for photosynthesis are characterized by many parameters that are obtained from limited in situ measurements and applied to the same plant types. Previous site-by-site calibration approaches could not leverage big data and faced issues like overfitting or parameter non-uniqueness. Here we developed an end-to-end programmatically differentiable (meaning gradients of outputs to variables used in the model can be obtained efficiently and accurately) version of the photosynthesis process representation within the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) model. As a genre of physics-informed machine learning (ML), differentiable models couple physics-based formulations to neural networks (NNs) that learn parameterizations (and potentially processes) from observations, here photosynthesis rates. We first demonstrated that the framework was able to correctly recover multiple assumed parameter values concurrently using synthetic training data. Then, using a real-world dataset consisting of many different plant functional types (PFTs), we learned parameters that performed substantially better and greatly reduced biases compared to literature values. Further, the framework allowed us to gain insights at a large scale. Our results showed that the carboxylation rate at 25 °C (Vc,max25) was more impactful than a factor representing water limitation, although tuning both was helpful in addressing biases with the default values. This framework could potentially enable substantial improvement in our capability to learn parameters and reduce biases for ecosystem modeling at large scales.
Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States); Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
Grant/Contract Number:
AC05-00OR22725; SC0021979
OSTI ID:
1997685
Alternate ID(s):
OSTI ID: 2568878
Report Number(s):
LA-UR--23-25521
Journal Information:
Biogeosciences (Online), Journal Name: Biogeosciences (Online) Journal Issue: 13 Vol. 20; ISSN 1726-4189
Publisher:
Copernicus Publications, EGUCopyright Statement
Country of Publication:
United States
Language:
English

References (75)

Prolongation of SMAP to Spatiotemporally Seamless Coverage of Continental U.S. Using a Deep Learning Neural Network journal November 2017
Developing and Testing a Long Short-Term Memory Stream Temperature Model in Daily and Continental Scale posted_content December 2020
Deep learning approaches for improving prediction of daily stream temperature in data‐scarce, unmonitored, and dammed basins journal November 2021
Encyclopedia of Plant Physiology, New Series. Editors: Pirson, A.; Zimmermann, M.H., Vol. 12, Part A (in 4 parts) Physiological Plant Ecology I. Responses to the Physical Environment, Editors: Lange, O.L.; Nobel, P.S.; Osmond, C.B.; Ziegler, H., Springer‐Verlag, Berlin–Heidelberg–New York, 1981, 110 figs. XV, 625 pages. Cloth DM 239,– journal January 1983
Modelling of Photosynthetic Response to Environmental Conditions book January 1982
A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species journal June 1980
Aleatoric and epistemic uncertainty in machine learning: an introduction to concepts and methods journal March 2021
Parameter inversion estimation in photosynthetic models: Impact of different simulation methods journal June 2014
The use and misuse of V c,max in Earth System Models journal April 2013
River flow forecasting through conceptual models part I — A discussion of principles journal April 1970
The thermal properties of soils in cold regions journal September 1981
Equifinality, data assimilation, and uncertainty estimation in mechanistic modelling of complex environmental systems using the GLUE methodology journal August 2001
A manifesto for the equifinality thesis journal March 2006
C3 and C4 photosynthesis models: An overview from the perspective of crop modelling journal December 2009
A deep learning-based novel approach to generate continuous daily stream nitrate concentration for nitrate data-sparse watersheds journal June 2023
From Hydrometeorology to River Water Quality: Can a Deep Learning Model Predict Dissolved Oxygen at the Continental Scale? journal February 2021
Uncertainties in global terrestrial biosphere modeling, Part II: Global constraints for a process-based vegetation model journal March 2001
Equifinality in parameterization of process-based biogeochemistry models: A significant uncertainty source to the estimation of regional carbon dynamics: EQUIFINALITY IN REGIONAL CARBON DYNAMICS journal October 2008
Deep Learning: A Next-Generation Big-Data Approach for Hydrology journal April 2018
The Community Land Model Version 5: Description of New Features, Benchmarking, and Impact of Forcing Uncertainty journal December 2019
Enhancing Streamflow Forecast and Extracting Insights Using Long‐Short Term Memory Networks With Data Integration at Continental Scales journal September 2020
Transferring Hydrologic Data Across Continents – Leveraging Data‐Rich Regions to Improve Hydrologic Prediction in Data‐Sparse Regions journal April 2021
Mitigating Prediction Error of Deep Learning Streamflow Models in Large Data‐Sparse Regions With Ensemble Modeling and Soft Data journal July 2021
A Multiscale Deep Learning Model for Soil Moisture Integrating Satellite and In Situ Data journal March 2022
Machine Learning‐Based Modeling of Vegetation Leaf Area Index and Gross Primary Productivity Across North America and Comparison With a Process‐Based Model journal October 2021
The Data Synergy Effects of Time‐Series Deep Learning Models in Hydrology journal April 2022
Differentiable, Learnable, Regionalized Process‐Based Models With Multiphysical Outputs can Approach State‐Of‐The‐Art Hydrologic Prediction Accuracy journal October 2022
Empirical equations for some soil hydraulic properties journal August 1978
Deep learning journal May 2015
Optimal stomatal behaviour around the world journal March 2015
Global variation in the fraction of leaf nitrogen allocated to photosynthesis journal August 2021
From calibration to parameter learning: Harnessing the scaling effects of big data in geoscientific modeling journal October 2021
Increasing impacts of extreme droughts on vegetation productivity under climate change journal November 2019
Deep learning and process understanding for data-driven Earth system science journal February 2019
Temperature outweighs light and flow as the predominant driver of dissolved oxygen in US rivers journal March 2023
C4 photosynthesis in a single C3 cell is theoretically inefficient but may ameliorate internal CO2 diffusion limitations of C3 leaves journal August 2003
Direct and Indirect Climate Change Effects on Photosynthesis and Transpiration journal May 2004
Coupled Photosynthesis-Stomatal Conductance Model for Leaves of C4 Plants journal January 1992
Parametrization of peatland hydraulic properties for the Canadian land surface scheme journal March 2000
Application of Deep Learning Neural Networks for Nitrate Prediction in the Klokot River, Bosnia and Herzegovina conference February 2021
Matplotlib: A 2D Graphics Environment journal January 2007
Integration of Multisource Data to Estimate Downward Longwave Radiation Based on Deep Neural Networks journal January 2022
Automatic Fruit Classification Using Deep Learning for Industrial Applications journal February 2019
Leaf chlorophyll content as a proxy for leaf photosynthetic capacity journal January 2017
Towards physiologically meaningful water‐use efficiency estimates from eddy covariance data journal September 2017
Vegetation demographics in Earth System Models: A review of progress and priorities journal October 2017
TRY plant trait database – enhanced coverage and open access journal December 2019
Reconciling the optimal and empirical approaches to modelling stomatal conductance: RECONCILING OPTIMAL AND EMPIRICAL STOMATAL MODELS journal January 2011
Plant carbon metabolism and climate change: elevated CO 2 and temperature impacts on photosynthesis, photorespiration and respiration journal July 2018
Steady-state models of photosynthesis: Steady-state models of photosynthesis journal April 2013
Toward dynamic global vegetation models for simulating vegetation–climate interactions and feedbacks: recent developments, limitations, and future challenges journal December 2010
High-performance symbolic-numerics via multiple dispatch journal September 2021
Near-Real-Time Forecast of Satellite-Based Soil Moisture Using Long Short-Term Memory with an Adaptive Data Integration Kernel journal March 2019
Machine learning models for net photosynthetic rate prediction using poplar leaf phenotype data journal February 2020
Global-scale environmental control of plant photosynthetic capacity journal December 2015
A Climate-Vegetation Interaction Model: Simulating Physical and Biological Processes at the Surface journal March 1995
Effects of Elevated Carbon Dioxide on Photosynthesis and Carbon Partitioning: A Perspective on Root Sugar Sensing and Hormonal Crosstalk journal August 2017
Assessing the Effects of Water Deficit on Photosynthesis Using Parameters Derived from Measurements of Leaf Gas Exchange and of Chlorophyll a Fluorescence journal December 2017
Assessment of a Spatiotemporal Deep Learning Approach for Soil Moisture Prediction and Filling the Gaps in Between Soil Moisture Observations journal March 2021
Plant Disease Detection and Classification by Deep Learning journal October 2019
Deep learning approaches for improving prediction of daily stream temperature in data-scarce, unmonitored, and dammed basins dataset January 2021
Impacts of trait variation through observed trait–climate relationships on performance of an Earth system model: a conceptual analysis journal January 2013
Benchmarking and parameter sensitivity of physiological and vegetation dynamics using the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) at Barro Colorado Island, Panama journal January 2020
C3 plants converge on a universal relationship between leaf maximum carboxylation rate and chlorophyll content preprint July 2019
Global datasets of leaf photosynthetic capacity for ecological and earth system research journal September 2022
Parameter calibration and stomatal conductance formulation comparison for boreal forests with adaptive population importance sampler in the land surface model JSBACH journal January 2019
Evaluating a global soil moisture dataset from a multitask model (GSM3 v1.0) with potential applications for crop threats journal March 2023
Taking off the training wheels: the properties of a dynamic vegetation model without climate envelopes, CLM4.5(ED) journal January 2015
Linking hydraulic traits to tropical forest function in a size-structured and trait-driven model (TFS v.1-Hydro) journal January 2016
A global scale mechanistic model of photosynthetic capacity (LUNA V1.0) journal January 2016
HESS Opinions: Incubating deep-learning-powered hydrologic science advances as a community journal January 2018
Groundwater level forecasting with artificial neural networks: a comparison of long short-term memory (LSTM), convolutional neural networks (CNNs), and non-linear autoregressive networks with exogenous input (NARX) journal April 2021
Automatically delineating the calving front of Jakobshavn Isbræ from multitemporal TerraSAR-X images: a deep learning approach journal January 2019
DeepBedMap: a deep neural network for resolving the bed topography of Antarctica journal November 2020
Global-scale environmental control of plant photosynthetic capacity dataset August 2016

Similar Records

Inferring Plant Acclimation and Improving Model Generalizability With Differentiable Physics‐Informed Machine Learning of Photosynthesis
Journal Article · Wed Jun 25 20:00:00 EDT 2025 · Journal of Geophysical Research. Biogeosciences · OSTI ID:2575575

Related Subjects

54 ENVIRONMENTAL SCIENCES
59 BASIC BIOLOGICAL SCIENCES