CLM5-DART Regional Carbon Fluxes and Stocks over the Western US, 1998-2010

This dataset provides monthly estimates of biomass stocks and land-atmosphere carbon exchange across the western United States at 0.95 degrees longitude x 1.25 degrees latitude grid resolution from 1998 through 2010. The data include outputs from two types of model simulations: (1) a "free" simulation which used Community Land Model (CLM version 5.0) simulations forced with meteorology appropriate for complex mountainous terrain, and (2) "assimilation" runs using the land surface data assimilation system (CLM5-DART). In assimilation runs, the CLM5 vegetation state is constrained by remotely sensed observations of leaf area index and aboveground biomass, which influenced biomass stocks and carbon fluxes.

The CLM5-DART (CLM5 with Data Assimilation Research Testbed) land surface data assimilation system was run with an ensemble spread induced from 80 members of meteorological forcing from the Community Atmosphere Model (CAM4) (Raeder et al., 2012). The ensemble spread provided a means to assess uncertainty for the biomass stock and carbon exchange estimates. The simulated carbon fluxes (i.e., net ecosystem production, gross primary production, and ecosystem respiration) were compared against independent estimates of regional carbon flux from FLUXCOM (Jung et al., 2020).

There are 312 data files in netCDF-4 (*.nc4) format.

This dataset provides monthly estimates of biomass stocks and land-atmosphere carbon exchange across the western United States at 0.95 degrees longitude x 1.25 degrees latitude grid resolution from 1998 through 2010. The data include outputs from two types of model simulations: (1) a "free" simulation which used Community Land Model (CLM version 5.0) simulations forced with meteorology appropriate for complex mountainous terrain, and (2) "assimilation" runs using the land surface data assimilation system (CLM5-DART). In assimilation runs, the CLM5 vegetation state is constrained by remotely sensed observations of leaf area index and aboveground biomass, which influenced biomass stocks and carbon fluxes.

The CLM5-DART (CLM5 with Data Assimilation Research Testbed) land surface data assimilation system was run with an ensemble spread induced from 80 members of meteorological forcing from the Community Atmosphere Model (CAM4) (Raeder et al., 2012). The ensemble spread provided a means to assess uncertainty for the biomass stock and carbon exchange estimates. The simulated carbon fluxes (i.e., net ecosystem production, gross primary production, and ecosystem respiration) were compared against independent estimates of regional carbon flux from FLUXCOM (Jung et al., 2020).

Project: Carbon Monitoring System

The NASA Carbon Monitoring System (CMS) program is designed to make significant contributions in characterizing, quantifying, understanding, and predicting the evolution of global carbon sources and sinks through improved monitoring of carbon stocks and fluxes. The System uses NASA satellite observations and modeling/analysis capabilities to establish the accuracy, quantitative uncertainties, and utility of products for supporting national and international policy, regulatory, and management activities. CMS data products are designed to inform near-term policy development and planning.

Related Publications

Raczka, B., T.J. Hoar, H.F. Duarte, A.M. Fox, J.L. Anderson, D.R. Bowling, and J.C. Lin. 2021. Improving CLM5.0 biomass and carbon exchange across the western US using a data assimilation system. Journal of Advances in Modeling Earth Systems. https://doi.org/10.1029/2020MS002421

Related Datasets

Hagen, S., N. Harris, S.S. Saatchi, T. Pearson, C.W. Woodall, S. Ganguly, G.M. Domke, B.H. Braswell, B.F. Walters, J.C. Jenkins, S. Brown, W.A. Salas, A. Fore, Y. Yu, R.R. Nemani, C. Ipsan, and K.R. Brown. 2016. CMS: Forest Carbon Stocks, Emissions, and Net Flux for the Conterminous US: 2005-2010. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/1313.

Joiner, J., and Y. Yoshida. 2021. Global MODIS and FLUXNET-derived Daily Gross Primary Production, V2. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/1835.

Zhou, Y., C.A. Williams, T. Lauvaux, S. Feng, I.T. Baker, Y. Wei, A.S. Denning, K. Keller, and K.J. Davis. 2019. ACT-America: Gridded Ensembles of Surface Biogenic Carbon Fluxes, 2003-2019. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/1675.

Acknowledgments

This project received funding from NASA’s Carbon Monitoring System (grants NNX16AP33G, 80NSSC20K0010).

Spatial Coverage: Western United States

Spatial Resolution: 0.95 to 1.25 degrees

Temporal Coverage: 1998-01-01 to 2010-12-31

Temporal Resolution: Monthly

Study Area: All latitudes and longitudes given in decimal degrees

Site	Westernmost Longitude	Easternmost Longitude	Northernmost Latitude	Southernmost Latitude
Western United States	-130.625	-99.375	50.891	25.445

Data File Information

There are 312 data files in netCDF-4 (*.nc4) format. The files contain raster grids in geographic coordinates. Note that peripheral cells outside of the modeled area were filled with missing values (-9999). Details about the CLM5-DART model runs can be found in the global attributes saved in the netCDFs.

The naming convention is clm5_XXXXX_ensemble_YYYY-MM.nc4, where XXXX indicates the type of simulation (i.e., “assim” or “free”), YYYY = year, and MM = month. The “clm5” refers to the CLM5.0 model “ensemble” runs (n=80) per month.

Data File Details

Table 1. Variable names and descriptions for the model output files. See Lawrence et al. (2019) for technical descriptions.

Variable	Dimensions	Units	Description
BTRAN2	lon, lat, ensemble	1	root zone soil wetness factor
BTRANMN	lon, lat, ensemble	1	daily minimum of transpiration beta factor
CPOOL	lon, lat, ensemble	g m-2	temporary photosynthate C pool
DEADCROOTC	lon, lat, ensemble	g m-2	structural carbon, not biologically active, within roots
DEADCROOTN	lon, lat, ensemble	g m-2	non-biologically active nitrogen content of coarse roots
DEADSTEMC	lon, lat, ensemble	g m-2	dead stem C
DEADSTEMN	lon, lat, ensemble	g m-2	dead stem N
EFLX_LH_TOT	lon, lat, ensemble	W m-2	total latent heat flux [+ to atmosphere]
ER	lon, lat, ensemble	g m-2 s-1	total ecosystem respiration, autotrophic + heterotrophic, in grams carbon
FROOTC	lon, lat, ensemble	g m-2	biologically active carbon content of fine roots
FROOTN	lon, lat, ensemble	g m-2	biologically active nitrogen content of fine roots
FSH	lon, lat, ensemble	W m-2	sensible heat not including correction for land-use change and rain/snow conversion
FSIF	lon, lat, ensemble	W m-2	solar-induced fluorescence
GPP	lon, lat, ensemble	g m-2 s-1	gross primary production in grams carbon
H2OSNO	lon, lat, ensemble	mm	snow depth (liquid water)
H2OSOI	lon, lat, levsoi, ensemble	g cm-3	volumetric soil water (vegetated land units only)
HR	lon, lat, ensemble	g m-2 s-1	total heterotrophic respiration in grams carbon
HTOP	lon, lat, ensemble	m	canopy top
LEAFC	lon, lat, ensemble	g m-2	leaf C
LEAFN	lon, lat, ensemble	g m-2	leaf N
LITR1C_vr	lon, lat, levsoi, ensemble	g m-2	litter 1 C (vertically resolved)
LITR1N_vr	lon, lat, levdcmp, ensemble	g m-2	litter 1 N (vertically resolved)
LITR2C_vr	lon, lat, levsoi, ensemble	g m-2	litter 2 C (vertically resolved)
LITR2N_vr	lon, lat, levdcmp, ensemble	g m-2	litter 2 N (vertically resolved)
LITR3C_vr	lon, lat, levsoi, ensemble	g m-2	litter 3 C (vertically resolved)
LITR3N_vr	lon, lat, levdcmp, ensemble	g m-2	litter 3 N (vertically resolved)
LIVECROOTC	lon, lat, ensemble	g m-2	biologically active carbon content of coarse roots
LIVECROOTN	lon, lat, ensemble	g m-2	biologically active nitrogen content of coarse roots
LIVESTEMC	lon, lat, ensemble	g m-2	live stem C
LIVESTEMN	lon, lat, ensemble	g m-2	live stem N
NEP	lon, lat, ensemble	g m-2 s-1	net ecosystem production in g C, excludes fire, land use, and harvest flux, positive for sink
NPOOL	lon, lat, ensemble	g m-2	temporary plant N pool
Q2M	lon, lat, ensemble	1	2-m specific humidity
QDRAI	lon, lat, ensemble	mm s-1	sub-surface drainage
QINFL	lon, lat, ensemble	mm s-1	infiltration (=precipitation -runoff -evaporation from canopy)
QRUNOFF	lon, lat, ensemble	mm s-1	total liquid runoff not including correction for land-use change
QSOIL	lon, lat, ensemble	mm s-1	ground evaporation (soil/snow evaporation + soil/snow sublimation - dew)
QVEGE	lon, lat, ensemble	mm s-1	canopy evaporation
QVEGT	lon, lat, ensemble	mm s-1	canopy transpiration
RAIN	lon, lat, ensemble	mm s-1	atmospheric rain, after rain/snow repartitioning based on temperature
RH2M	lon, lat, ensemble	percent	relative humidity at 2 m
SNOW	lon, lat, ensemble	mm s-1	atmospheric snow, after rain/snow repartitioning based on temperature
SNOWDP	lon, lat, ensemble	m	mean snow height in grid cell
SNOWLIQ	lon, lat, ensemble	kg m-2	snow liquid water
SOIL1C_vr	lon, lat, levsoi, ensemble	g m-3	soil 1 C (vertically resolved)
SOIL1N_vr	lon, lat, levdcmp, ensemble	g m-3	soil 1 N (vertically resolved)
SOIL2C_vr	lon, lat, levsoi, ensemble	g m-3	soil 2 C (vertically resolved)
SOIL2N_vr	lon, lat, levdcmp, ensemble	g m-3	soil 2 N (vertically resolved)
SOIL3C_vr	lon, lat, levsoi, ensemble	g m-3	soil 3 C (vertically resolved)
SOIL3N_vr	lon, lat, levdcmp, ensemble	g m-3	soil 3 N (vertically resolved)
SOILLIQ	lon, lat, levsoi, ensemble	kg m-2	soil liquid water (vegetated land units only)
TG	lon, lat, ensemble	K	ground temperature
TLAI	lon, lat, ensemble	1	total projected leaf area index
TSA	lon, lat, ensemble	K	2 m air temperature
TSOI	lon, lat, levgrnd, ensemble	K	soil temperature (vegetated land units only)
TSOI_10CM	lon, lat, ensemble	K	soil temperature in top 10cm of soil
TV	lon, lat, ensemble	K	vegetation temperature
landfrac	lon, lat	1	land fraction

Table 2. Variables associated with file dimensions in the model output files. See Lawrence et al. (2019) for technical descriptions.

Variable	Units	Description
lat	degree_north	coordinate latitude (n = 27)
lon	degree_east	coordinate longitude (n = 25)
ensemble	1	ensemble number (n = 80)
levdcmp	m	coordinate soil levels (n = 25)
levgrnd	m	coordinate soil levels (n = 25)
levlak	m	coordinate lake levels (n = 10)

Biomass stocks, soil carbon, and carbon fluxes are difficult to monitor in complex, mountainous terrain. The CLM5 land surface biosphere model was forced with a bias-corrected meteorological dataset designed for complex terrain. CLM5 estimates key carbon and flux variables by representing important earth system processes. All models are subject to deviation from reality; therefore, the data assimilation system DART, was used to nudge the modeled vegetation closer to observed vegetation states (Fig. 2). In this case, remotely sensed leaf area index observations (Zhu et al., 2013) and aboveground biomass (Liu et al., 2015) constrained the CLM simulation. This approach improved the vegetation state's representation along with the associated carbon, water, and nitrogen cycling. This dataset includes both the unconstrained simulations (“free”) and the simulations constrained by remotely sensed vegetation states (“assim”).

The CLM5-DART carbon flux simulations were compared to FLUXCOM (Jung et al., 2020), an independent estimate of carbon flux. CLM5-DART is a mechanistic approach that represents vegetation states with biomass and leaf area index (LAI) observations to account for disturbances and land-use history. In contrast, FLUXCOM is a more empirical machine learning approach that relies primarily on flux tower observations to represent regional carbon fluxes.

In the simulations without vegetation constraints ("free"), CLM5-DART and FLUXCOM predicted similar levels of ecosystem respiration across the western U.S. However, estimates of gross primary production (GPP) were lower for CLM5-DART (38.5 g C m^-2) than FLUXCOM (43.6 g C m^-2). Across the region, FLUXCOM predicted a stronger carbon sink (8 g C m^-2) than CLM5-DART (1.3 g C m^-2). The vegetation-constrained assimilation model ("assim") had a greater agreement with FLUXCOM estimates; however, regional differences persisted. The greatest difference was in mountainous terrain where CLM5-DART predicted near-neutral carbon exchange while FLUXCOM predicted strong carbon update and storage (Raczka et al., 2021).

All simulations were performed within 11 states of the western U.S., with the model domain defined between the U.S. borders with Mexico and Canada, and between the Pacific Ocean and the eastern edge of Colorado within 49.00^oN–31.30^oN and 124.40^oW–102.05^oW (Fig. 1).

Estimates of biomass stocks and land-atmosphere carbon exchange were derived from the terrestrial biosphere Community Land Model (CLM5) (CLM version 5.0, Lawrence et al., 2019) across the western United States from 1998–2011. CLM5 is a mechanistic terrestrial biosphere model that can assimilate observational data (e.g., meteorology, vegetation states) to adjust model predictions to better match real-world states and dynamics (Fig. 2). CLM5-DART assimilates observational data using the Data Assimilation Research Testbed (DART; Anderson et al., 2009).

Figure 2. An overview of the assimilation system (CLM5-DART) that combines observations with model ensemble simulations from CLM5.0. Satellite-derived observations of aboveground biomass and leaf area index were used to adjust the simulated model state of CLM5.0. Source: Razcka et al. (2021)

This dataset includes outputs from two types of CLM5 model simulations. The first type was a "free" simulation which was forced with meteorology (Buotte et al., 2019) specifically designed to be run across complex, mountainous terrain. These simulations allowed land surface vegetation states to freely vary as modified by model dynamics. The second type of simulation was "assimilation" runs using the land surface data assimilation system (CLM5-DART). In these assimilation runs, the CLM5 vegetation state was constrained by remotely sensed leaf area index observations (Zhu et al., 2013) and aboveground biomass (Liu et al., 2015). These constraints influenced both the biomass stocks and carbon flux characteristics.

The CLM5-DART assimilation system was run with 80 ensemble members in which the ensemble spread is induced by meteorological forcing from the Community Atmosphere Model (CAM4) (Raeder et al., 2012). Variation among ensemble members provided a means to assess uncertainty for the biomass stock and carbon exchange estimates. The simulated carbon fluxes from CLM5-DART (net ecosystem production, gross primary production, and ecosystem respiration) were compared against an independent estimate of regional carbon flux from FLUXCOM (Jung et al., 2020).

More details about the methods can be found within Raczka et al. (2021).

Anderson, J., T. Hoar, K. Raeder, H. Liu, N. Collins, R. Torn, and A. Avellano. 2009. The Data Assimilation Research Testbed: a community facility. Bulletin of the American Meteorological Society 90:1283–1296. https://doi.org/10.1175/2009BAMS2618.1.

Buotte, P.C., S. Levis, B.E. Law, T.W. Hudiburg, D.E. Rupp, and J.J. Kent. 2019. Near-future forest vulnerability to drought and fire varies across the western United States. Global Change Biology 25:290–303. https://doi.org/10.1111/gcb.14490.

Jung, M., C. Schwalm, M. Migliavacca, S. Walther, G. Camps-Valls, S. Koirala, P. Anthoni, S. Besnard, P. Bodesheim, N. Carvalhais, F. Chevallier, F. Gans, D.S. Goll, V. Haverd, P. Köhler, K. Ichii, A.K. Jain, J. Liu, D. Lombardozzi, J.E.M.S. Nabel, J.A. Nelson, M. O'Sullivan, M. Pallandt, D. Papale, W. Peters, J. Pongratz, C. Rödenbeck, S. Sitch, G. Tramontana, A. Walker, U. Weber, and M. Reichstein. 2020. Scaling carbon fluxes from eddy covariance sites to globe: synthesis and evaluation of the FLUXCOM approach. Biogeosciences 17:1343–1365. https://doi.org/10.5194/bg-17-1343-2020.

Lawrence, D.M., R.A. Fisher, C.D. Koven, K.W. Oleson, S.C. Swenson, G. Bonan, N. Collier, B. Ghimire, L. van Kampenhout, D. Kennedy, E. Kluzek, P.J. Lawrence, F. Li, H. Li, D. Lombardozzi, W.J. Riley, W.J. Sacks, M. Shi, M. Vertenstein, W.R. Wieder, C. Xu, A.A. Ali, A.M. Badger, G. Bisht, M. van den Broeke, M.A. Brunke, S.P. Burns, J. Buzan, M. Clark, A. Craig, K. Dahlin, B. Drewniak, J.B. Fisher, M. Flanner, A.M. Fox, P. Gentine, F. Hoffman, G. Keppelaleks, R. Knox, S. Kumar, J. Lenaerts, L.R. Leung, W.H. Lipscomb, Y. Lu, A. Pandey, J.D. Pelletier, J. Perket, J.T. Randerson, D.M. Ricciuto, B.M. Sanderson, A. Slater, Z.M. Subin, J. Tang, R.Q. Thomas, M. Val Martin, and X. Zeng. 2019. The Community Land Model Version 5: description of new features, benchmarking, and impact of forcing uncertainty. Journal of Advances in Modeling Earth Systems, 11(12), 4245–4287. https://doi.org/10.1029/2018MS001583.

Liu, M., H. He, X. Ren, X. Sun, G. Yu, S. Han, H. Wang, and G. Zhou. 2015. The effects of constraining variables on parameter optimization in carbon and water flux modeling over different forest ecosystems. Ecological Modelling 303:30–41. https://doi.org/10.1016/j.ecolmodel.2015.01.027.

Raczka, B., T.J. Hoar, H.F. Duarte, A.M. Fox, J.L. Anderson, D.R. Bowling, and J.C. Lin. 2021. Improving CLM5.0 biomass and carbon exchange across the western US using a data assimilation system. Journal of Advances in Modeling Earth Systems. https://doi.org/10.1029/2020MS002421

Raeder, K., J.L. Anderson, N. Collins, T.J. Hoar, J E. Kay, P H. Lauritzen, and R. Pincus, R. 2012. DART/CAM: an ensemble data assimilation system for CESM atmospheric models. Journal of Climate 25:6304–6317. https://doi.org/10.1175/JCLI-D-11-00395.1.

Zhu, Z., J. Bi, Y. Pan, S. Ganguly, A. Anav, L. Xu, A. Samanta, S. Piao, R. R. Nemani, and R. B. Myneni. 2013. Global data sets of vegetation leaf area index (LAI)3g and fraction of photosynthetically active radiation (FPAR)3g derived from Global Inventory Modeling and Mapping Studies (GIMMS) Normalized Difference Vegetation Index (NDVI3g) for the period 1981 to 2011. Remote Sensing 5:927–948. https://doi.org/10.3390/rs5020927.