Simulating shrubs and their energy and carbon dioxide fluxes in Canada's Low Arctic with the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC)

Climate change in the Arctic is leading to shifts in vegetation communities, permafrost degradation and alteration of tundra surface–atmosphere energy and carbon (C) fluxes, among other changes. However, year-round C and energy flux measurements at high-latitude sites remain rare. This poses a challenge for evaluating the impacts of climate change on Arctic tundra ecosystems and for developing and evaluating process-based models, which may be used to predict regional and global energy and C feedbacks to the climate system. Our study used 14 years of seasonal eddy covariance (EC) measurements of carbon dioxide (CO2), water and energy fluxes, and winter soil chamber CO2 flux measurements at a dwarf-shrub tundra site underlain by continuous permafrost in Canada’s Southern Arctic ecozone to evaluate the incorporation of shrub plant functional types (PFTs) in the Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC), the land surface component of the Canadian Earth System Model. In addition to new PFTs, a modification of the efficiency with which water evaporates from the ground surface was applied. This modification addressed a high ground evaporation bias that reduced model performance when soils became very dry, limited heat flow into the ground, and reduced plant productivity through water stress effects. Compared to the grass and tree PFTs previously used by CLASSIC to represent the vegetation in Arctic permafrost-affected regions, simulations with the new shrub PFTs better capture the physical and biogeochemical impact of shrubs on the magnitude and seasonality of energy and CO2 fluxes at the dwarf-shrub tundra evaluation site. The revised model, however, tends to overestimate gross primary productivity, particularly in spring, and overestimated late-winter CO2 emissions. On average, annual net ecosystem CO2 exchange was positive for all simulations, suggesting this site was a net CO2 source of 18 ± 4 g C m−2 yr−1 using shrub PFTs, 15 ± 6 g C m−2 yr−1 using grass PFTs, and 25 ± 5 g C m−2 yr−1 using tree PFTs. These results highlight the importance of using appropriate PFTs in process-based models to simulate current and future Arctic surface–atmosphere interactions.

CLASSIC v1.0: the open-source community successor to the Canadian Land Surface Scheme (CLASS) and the Canadian Terrestrial Ecosystem Model (CTEM) – Part 2: Global benchmarking

The Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC) is an open-source community model designed to address research questions that explore the role of the land surface in the global climate system. Here, we evaluate how well CLASSIC reproduces the energy, water, and carbon cycle when forced with quasi-observed meteorological data. Model skill scores summarize how well model output agrees with observation-based reference data across multiple statistical metrics. A lack of agreement may be due to deficiencies in the model, its forcing data, and/or reference data. To address uncertainties in the forcing, we evaluate an ensemble of CLASSIC runs that is based on three meteorological data sets. To account for observational uncertainty, we compute benchmark skill scores that quantify the level of agreement among independent reference data sets. The benchmark scores demonstrate what score values a model may realistically achieve given the uncertainties in the observations. Our results show that uncertainties associated with the forcing and observations are considerably large. For instance, for 10 out of 19 variables assessed in this study, the sign of the bias changes depending on what forcing and reference data are used. Benchmark scores are much lower than expected, implying large observational uncertainties. Model and benchmark score values are mostly similar, indicating that CLASSIC performs well when considering observational uncertainty. Future model development should address (i) a positive albedo bias and resulting shortwave radiation bias in parts of the Northern Hemisphere (NH) extratropics and Tibetan Plateau, (ii) an out-of-phase seasonal gross primary productivity cycle in the humid tropics of South America and Africa, (iii) a lacking spatial correlation of annual mean net ecosystem exchange with site-level measurements, (iv) an underestimation of fractional area burned and corresponding emissions in the boreal forests, (v) a negative soil organic carbon bias in high latitudes, and (vi) a time lag in seasonal leaf area index maxima in parts of the NH extratropics. Our results will serve as a baseline for guiding and monitoring future CLASSIC development.

Simulating shrubs and their energy and carbon dioxide fluxes in Canada's Low Arctic with the Canadian Land Surface Scheme Including biogeochemical Cycles (CLASSIC)

The Arctic is warming more rapidly than other regions of the world leading to ecosystem change including shifts in vegetation communities, permafrost degradation and alteration of tundra surface-atmosphere energy and carbon (C) fluxes, among others. However, year-round C and energy flux measurements at high-latitude sites remain rare. This poses a challenge for evaluating the impacts of climate change on Arctic tundra ecosystems and for developing and evaluating process-based models, which may be used to predict regional and global energy and C feedbacks to the climate system. Our study used 14 years of seasonal eddy covariance (EC) measurements of carbon dioxide (CO2), water and energy fluxes and winter soil chamber CO2 flux measurements at a dwarf-shrub tundra site underlain by continuous permafrost in Canada's Southern Arctic ecozone to evaluate the incorporation of shrub plant functional types (PFTs) in the Canadian Land Surface Scheme Including biogeochemical Cycles (CLASSIC), the land surface component of the Canadian Earth System Model. In addition to new PFTs, a modification of the efficiency with which water evaporates from the ground surface was applied. This modification addressed a high ground evaporation bias that reduced model performance when soils became very dry, limited heat flow into the ground and reduced plant productivity through water stress effects. Compared to the grass and tree PFTs previously used by CLASSIC to represent the vegetation in Arctic permafrost-affected regions, simulations with the new shrub PFTs better capture the physical and biogeochemical impact of shrubs on the magnitude and seasonality of energy and CO2 fluxes at the dwarf-shrub tundra evaluation site. The revised model, however, tends to overestimate gross primary productivity, particularly in spring, and overestimated late winter CO2 emissions. On average, annual net ecosystem CO2 exchange was positive for all simulations, suggesting this site was a net CO2 source of 18 ± 4 g C m−2 year−1 using shrub PFTs, 15 ± 6 g C m−2 year−1 using grass PFTs, and 25 ± 5 g C m−2 year−1 using tree PFTs. These results highlight the importance of using appropriate PFTs in process-based models to simulate current and future Arctic surface-atmosphere interactions.

CLASSIC v1.0: the open-source community successor to the Canadian Land Surface Scheme (CLASS) and the Canadian Terrestrial Ecosystem Model (CTEM) – Part 2: Global Benchmarking

The Canadian Land Surface Scheme Including Biogeochemical Cycles (CLASSIC) is an open source community model designed to address research questions that explore the role of the land surface in the global climate system. Here we evaluate how well CLASSIC reproduces the energy, water, and carbon cycle when forced with quasi-observed meteorological data. Model skill scores summarize how well model output agrees with observation-based reference data across multiple statistical metrics. A lack of agreement may be due to deficiencies in the model, its forcing data, and/or reference data. To address uncertainties in the forcing we evaluate an ensemble of CLASSIC runs that is based on three meteorological data sets. To account for observational uncertainty, we compute benchmark skill scores that quantify the level of agreement among independent reference data sets. The benchmark scores demonstrate what score values a model may realistically achieve given the uncertainties in the observations. Our results show that uncertainties associated with the forcing and observations are considerably large. For instance, for 10 out of 19 variables assessed in this study, the sign of the bias changes depending on what forcing and reference data are used. Benchmark scores are much lower than expected, implying large observational uncertainties. Model and benchmark score values are mostly similar, indicating that CLASSIC performs well when considering observational uncertainty. Using the difference between model and benchmark scores as a measure of performance shows that model skill increases in the following order: fractional area burned, runoff, soil heat flux, leaf area index, net shortwave radiation, net ecosystem exchange, above-ground biomass, gross primary productivity, surface albedo, snow water equivalent, net surface radiation, sensible heat flux, net longwave radiation, latent heat flux, and ecosystem respiration. Our results will serve as a baseline for guiding and monitoring future CLASSIC development.

Comparing the Assimilation of SMOS Brightness Temperatures and Soil Moisture Products on Hydrological Simulation in the Canadian Land Surface Scheme

Soil moisture is a key variable used to describe water and energy exchanges at the land surface/atmosphere interface. Therefore, there is widespread interest in the use of soil moisture retrievals from passive microwave satellites. In the assimilation of satellite soil moisture data into land surface models, two approaches are commonly used. In the first approach brightness temperature (TB) data are assimilated, while in the second approach retrieved soil moisture (SM) data from the satellite are assimilated. However, there is not a significant body of literature comparing the differences between these two approaches, and it is not known whether there is any advantage in using a particular approach over the other. In this study, TB and SM L2 retrieval products from the Soil Moisture and Ocean Salinity (SMOS) satellite are assimilated into the Canadian Land Surface Scheme (CLASS), for improved soil moisture estimation over an agricultural region in Saskatchewan. CLASS is the land surface component of the Canadian Earth System Model (CESM), and the Canadian Seasonal and Interannual Prediction System (CanSIPS). Our results indicated that assimilating the SMOS products improved the soil moisture simulation skill of the CLASS. Near surface soil moisture assimilation also resulted in improved forecasts of root zone soil moisture (RZSM) values. Although both techniques resulted in improved forecasts of RZSM, assimilation of TB resulted in the superior estimates.

CLASSIC v1.0: the open-source community successor to the Canadian Land Surface Scheme (CLASS) and the Canadian Terrestrial Ecosystem Model (CTEM) – Part 1: Model framework and site-level performance

Recent reports by the Global Carbon Project highlight large uncertainties around land surface processes such as land use change, strength of CO2 fertilization, nutrient limitation and supply, and response to variability in climate. Process-based land surface models are well suited to address these complex and emerging global change problems but will require extensive development and evaluation. The coupled Canadian Land Surface Scheme and Canadian Terrestrial Ecosystem Model (CLASS-CTEM) framework has been under continuous development by Environment and Climate Change Canada since 1987. As the open-source model of code development has revolutionized the software industry, scientific software is experiencing a similar evolution. Given the scale of the challenge facing land surface modellers, and the benefits of open-source, or community model, development, we have transitioned CLASS-CTEM from an internally developed model to an open-source community model, which we call the Canadian Land Surface Scheme including Biogeochemical Cycles (CLASSIC) v.1.0. CLASSIC contains many technical features specifically designed to encourage community use including software containerization for serial and parallel simulations, extensive benchmarking software and data (Automated Model Benchmarking; AMBER), self-documenting code, community standard formats for model inputs and outputs, amongst others. Here, we evaluate and benchmark CLASSIC against 31 FLUXNET sites where the model has been tailored to the site-level conditions and driven with observed meteorology. Future versions of CLASSIC will be developed using AMBER and these initial benchmark results to evaluate model performance over time. CLASSIC remains under active development and the code, site-level benchmarking data, software container, and AMBER are freely available for community use.

On the Performance of the Canadian Land Surface Scheme Driven by the ERA5 Reanalysis over the Canadian Boreal Forest

The Canadian Land Surface Scheme (CLASS) has been applied over the years in coupled and uncoupled (offline) modes at local, regional, and global scales using various forcing datasets. In this study, CLASS is applied at a local scale in the offline configuration to evaluate its performance when driven by the ERA5 reanalysis. Simulated surface energy fluxes, as well as several other water balance components, are investigated at four sites across the Canadian boreal biome. The results from CLASS driven by ERA5 (CLASS-RNL) are compared with available in situ measurements, as well as with results from CLASS driven by observations (CLASS-CTL). Additional simulations are conducted to evaluate the effects of biases in the ERA5 precipitation, where CLASS is forced by ERA5 data, but with ERA5 precipitation being replaced by observed precipitation (CLASS-RNL-ObsP). The results show that simulated surface variables in CLASS-RNL are in good agreement with observations as well as with those simulated in CLASS-CTL. The CLASS-RNL captures well the observed annual cycles of the surface energy and water fluxes, as well as the year-to-year variation of snow depth, soil temperature, and soil moisture. A strong correlation is found between the observed and CLASS-RNL simulated snow depth and soil temperature. Biases in the ERA5 precipitation did not affect the simulation of soil state variables, whereas the simulated surface heat and water fluxes, as well as the snow depth, were significantly affected. For instance, the simulated runoff in CLASS-RNL is much higher than in CLASS-RNL-ObsP and CLASS-CTL at the most humid sites due to significant positive bias in ERA5 precipitation.

Importance of shaded leaf contribution to the total GPP of high latitude ecosystems: evaluation of MODIS GPP

<p>The Moderate Resolution Imaging Radiometer (MODIS) is a primary instrument in the NASA Earth Observing System (EOS) which was designed for monitoring global terrestrial vegetation. MODIS provides global estimates of 8-day mean gross primary productivity (GPP) at 1-km spatial resolution. In this study, the MODIS GPP algorithm using light use efficiency (LUE) approach and the Integrated Carbon-Canadian Land Surface Scheme (IC-CLASS) based on Farquhar photosynthetic model and a sunlit and shaded leaf separation scheme was evaluated against eddy covariance (EC) measured GPP in a variety of ecosystems in Canada. Although GPP simulated by the two models agreed well when they were averaged over Canadian landmass, there were systematic differences between them in spatial distribution patterns. These differences were due to inherent shortcomings of the LUE modeling approach. When a constant maximum LUE value is specified for each biome type, this simplification cannot appropriately deal with the shaded leaf contribution to total canopy GPP. When GPP was simulated by IC-CLASS with the separation of sunlit and shaded leaves, the biases were minimized. Compared with daily and annual GPP derived from EC flux data at 7 Fluxnet Canada sites, IC-CLASS performed better than the MODIS GPP algorithm. The differences between IC-CLASS and MODIS GPP were larger in more clumped canopies (i.e. forests), resulting from the increase in the fraction of shaded leaves. Thus, the LUE models should be improved to consider different LUEs in sunlit and shaded portions of the canopy for their effective and reliable estimation of GPP at regional scale.</p>