Search for: All records

Abstract. Land models are essential tools for understanding and predicting terrestrial processes and climate–carbon feedbacks in the Earth system, but uncertainties in their future projections are poorly understood. Improvements in physical process realism and the representation of human influence arguably make models more comparable to reality but also increase the degrees of freedom in model configuration, leading to increased parametric uncertainty in projections. In this work we design and implement a machine learning approach to globally calibrate a subset of the parameters of the Community Land Model, version 5 (CLM5) to observations of carbon and water fluxes. We focus on parameters controlling biophysical features such as surface energy balance, hydrology, and carbon uptake. We first use parameter sensitivity simulations and a combination of objective metrics including ranked global mean sensitivity to multiple output variables and non-overlapping spatial pattern responses between parameters to narrow the parameter space and determine a subset of important CLM5 biophysical parameters for further analysis. Using a perturbed parameter ensemble, we then train a series of artificial feed-forward neural networks to emulate CLM5 output given parameter values as input. We use annual mean globally aggregated spatial variability in carbon and water fluxes as our emulation and calibration targets. Validation and out-of-sample tests are used to assess the predictive skill of the networks, and we utilize permutation feature importance and partial dependence methods to better interpret the results. The trained networks are then used to estimate global optimal parameter values with greater computational efficiency than achieved by hand tuning efforts and increased spatial scale relative to previous studies optimizing at a single site. By developing this methodology, our framework can help quantify the contribution of parameter uncertainty to overall uncertainty in land model projections. 

Land surface models (LSMs) are a vital tool for understanding, projecting, and predicting the dynamics of the land surface and its role within the Earth system, under global change. Driven by the need to address a set of key questions, LSMs have grown in complexity from simplified representations of land surface biophysics to encompass a broad set of interrelated processes spanning the disciplines of biophysics, biogeochemistry, hydrology, ecosystem ecology, community ecology, human management, and societal impacts. This vast scope and complexity, while warranted by the problems LSMs are designed to solve, has led to enormous challenges in understanding and attributing differences between LSM predictions. Meanwhile, the wide range of spatial scales that govern land surface heterogeneity, and the broad spectrum of timescales in land surface dynamics, create challenges in tractably representing processes in LSMs. We identify three “grand challenges” in the development and use of LSMs, based around these issues: managing process complexity, representing land surface heterogeneity, and understanding parametric dynamics across the broad set of problems asked of LSMs in a changing world. In this review, we discuss progress that has been made, as well as promising directions forward, for each of these challenges.

 

The response of tropical ecosystems to elevated carbon dioxide (CO₂) remains a critical uncertainty in projections of future climate. Here, we investigate how leaf trait plasticity in response to elevated CO₂alters projections of tropical forest competitive dynamics and functioning. We use vegetation demographic model simulations to quantify how plasticity in leaf mass per area and leaf carbon to nitrogen ratio alter the responses of carbon uptake, evapotranspiration, and competitive ability to a doubling of CO₂in a tropical forest. Observationally constrained leaf trait plasticity in response to CO₂fertilization reduces the degree to which tropical tree carbon uptake is affected by a doubling of CO₂(up to −14.7% as compared to a case with no plasticity; 95% confidence interval [CI_95%] −14.4 to −15.0). It also diminishes evapotranspiration (up to −7.0%, CI_95%−6.4 to −7.7), and lowers competitive ability in comparison to a tree with no plasticity. Consideration of leaf trait plasticity to elevated CO₂lowers tropical ecosystem carbon uptake and evapotranspirative cooling in the absence of changes in plant‐type abundance. However, “plastic” responses to high CO₂which maintain higher levels of plant productivity, many of which fall outside of the observed range of response, are potentially more competitively advantageous, thus, including changes in plant type abundance may mitigate these decreases in ecosystem functioning. Models that explicitly represent competition between plants with alternative leaf trait plasticity in response to elevated CO₂are needed to capture these influences on tropical forest functioning and large‐scale climate.

 

In tropical forests, both vegetation characteristics and soil properties are important not only for controlling energy, water, and gas exchanges directly but also determining the competition among species, successional dynamics, forest structure and composition. However, the joint effects of the two factors have received limited attention in Earth system model development. Here we use a vegetation demographic model, the Functionally Assembled Terrestrial Ecosystem Simulator (FATES) implemented in the Energy Exascale Earth System Model (E3SM) Land Model (ELM), ELM‐FATES, to explore how plant traits and soil properties affect tropical forest growth and composition concurrently. A large ensemble of simulations with perturbed vegetation and soil hydrological parameters is conducted at the Barro Colorado Island, Panama. The simulations are compared against observed carbon, energy, and water fluxes. We find that soil hydrological parameters, particularly the scaling exponent of the soil retention curve (B_sw), play crucial roles in controlling forest diversity, with higherB_swvalues (>7) favoring late successional species in competition, and lowerB_swvalues (1 ∼ 7) promoting the coexistence of early and late successional plants. Considering the additional impact of soil properties resolves a systematic bias of FATES in simulating sensible/latent heat partitioning with repercussion on water budget and plant coexistence. A greater fraction of deeper tree roots can help maintain the dry‐season soil moisture and plant gas exchange. As soil properties are as important as vegetation parameters in predicting tropical forest dynamics, more efforts are needed to improve parameterizations of soil functions and belowground processes and their interactions with aboveground vegetation dynamics.

 

Mechanistic photosynthesis models are at the heart of terrestrial biosphere models (TBMs) simulating the daily, monthly, annual and decadal rhythms of carbon assimilation (A). These models are founded on robust mathematical hypotheses that describe howAresponds to changes in light and atmospheric CO₂concentration. Two predominant photosynthesis models are in common usage: Farquhar (FvCB) and Collatz (CBGB). However, a detailed quantitative comparison of these two models has never been undertaken. In this study, we unify the FvCB and CBGB models to a common parameter set and use novel multi‐hypothesis methods (that account for both hypothesis and parameter variability) for process‐level sensitivity analysis. These models represent three key biological processes: carboxylation, electron transport, triose phosphate use (TPU) and an additional model process: limiting‐rate selection. Each of the four processes comprises 1–3 alternative hypotheses giving 12 possible individual models with a total of 14 parameters. To broaden inference, TBM simulations were run and novel, high‐resolution photosynthesis measurements were made. We show that parameters associated with carboxylation are the most influentialparametersbut also reveal the surprising and marked dominance of the limiting‐rate selectionprocess(accounting for 57% of the variation inAvs. 22% for carboxylation). The limiting‐rate selection assumption proposed by CBGB smooths the transition between limiting rates and always reducesAbelow the minimum of all potentially limiting rates, by up to 25%, effectively imposing a fourth limitation onA. Evaluation of the CBGB smoothing function in three TBMs demonstrated a reduction in globalAby 4%–10%, equivalent to 50%–160% of current annual fossil fuel emissions. This analysis reveals a surprising and previously unquantified influence of a process that has been integral to many TBMs for decades, highlighting the value of multi‐hypothesis methods.

 

Numerous current efforts seek to improve the representation of ecosystem ecology and vegetation demographic processes within Earth System Models (ESMs). These developments are widely viewed as an important step in developing greater realism in predictions of future ecosystem states and fluxes. Increased realism, however, leads to increased model complexity, with new features raising a suite of ecological questions that require empirical constraints. Here, we review the developments that permit the representation of plant demographics inESMs, and identify issues raised by these developments that highlight important gaps in ecological understanding. These issues inevitably translate into uncertainty in model projections but also allow models to be applied to new processes and questions concerning the dynamics of real‐world ecosystems. We argue that stronger and more innovative connections to data, across the range of scales considered, are required to address these gaps in understanding. The development of first‐generation land surface models as a unifying framework for ecophysiological understanding stimulated much research into plant physiological traits and gas exchange. Constraining predictions at ecologically relevant spatial and temporal scales will require a similar investment of effort and intensified inter‐disciplinary communication.