Application of NASA-Unified WRF model to carbon dioxide simulation- model development and evaluation
Introduction
Approximately one half of the anthropogenic emission of carbon dioxide (CO2) to the atmosphere is currently removed from the atmosphere by the air-sea flux and land-atmosphere flux. Interannual variability in this flux is large, and the magnitude, location, and mechanisms producing the CO2 sink are not well determined (IPCC, 2018; USGCRP, 2018). As a result, carbon-climate interaction is among the leading sources of uncertainty in prediction of future climate (Cox et al., 2000; Dufresne et al., 2002; Friedlingstein et al., 2014). In particular, the inferred terrestrial carbon sink and its variability, spatial distribution, and dependence on environmental conditions must be better characterized in order to assess how surface uptake and release of CO2 will evolve with changing fossil fuel emissions, land use, and climate in coming decades (IPCC, 2018 and references within).
Part of the problem in understanding these mechanisms is that key processes controlling CO2 fluxes and, hence, local CO2 mixing ratio variability, occur at relatively small spatial and/or rapid temporal scales. Examples include photosynthetic dependence on sunlight and soil moisture, temperature dependence of above-ground and soil respiration, and plant cover and phenology. Small scales are similarly important for fossil fuel emissions from point sources, urban and/or industrial areas, and transportation, as well as biomass burning. Because surface fluxes are generally inferred from gradients in atmospheric CO2, the problem is further complicated by transport in atmospheric boundary layers and frontal systems that can concentrate or reduce mixing ratio gradients produced by fluxes, which are themselves often correlated with the local weather (e.g., Parazoo et al., 2011). CO2 flux and abundance heterogeneity are so large that it becomes very difficult to integrate the net of these processes in time and space to global, regional, or even landscape scales. Yet it is precisely the small shifts in balance between photosynthesis and respiration during the mid-latitude seasonal cycle that are hypothesized to drive a large part of the global net terrestrial sink (Friedlingstein et al., 1995). As a result, global prognostic models, which have been developed from sparse observational evidence, diverge greatly, and bottom up flux inferences don't agree with top-down ones (e.g., Gourdji et al., 2012). Inferring fluxes from observations in either forward or inverse models is hindered by the ‘stiffness’ of the system owing to limited model resolution.
In addressing the aforementioned scale problems, efforts have been made to develop high-resolution regional or mesoscale carbon flux/transport models. Ahmadov et al. (2007) coupled a diagnostic biosphere model – Vegetation Photosynthesis and Respiration Model (VPRM, Mahadevan et al., 2008) – to the community Weather Research and Forecasting (WRF) model (Skamarock et al., 2008) to investigate the CO2 flux and transport at resolutions of a few tens of kilometers (km). VPRM is an empirical model calibrated to measurements of eddy covariance flux towers and extrapolated across the landscape. In WRF-VPRM, the CO2 net ecosystem exchange (NEE) flux is derived using the enhanced vegetation index and land surface water index from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the shortwave radiation and air temperature from WRF. WRF-VPRM has shown better performance than two global models in reproducing the coastal tower CO2 measurements (Ahmadov et al., 2009). Using WRF as the modeling framework, Ballav et al. (2012) simulated the CO2 concentration over East Asia at the 27 × 27 km horizontal resolution, and Diaz Isaac et al. (2014) compared the simulated CO2 at a 10 × 10 km resolution with the measurements from the Mid-Continental Intensive (MCI) field campaign, both of which utilized the terrestrial carbon flux from stand-alone biosphere models. Similar efforts employing other mesoscale transport models to investigate CO2 at high spatial resolutions include the online atmosphere-biosphere coupling works by Sarrat et al. (2007a, b), and the offline coupling works by Ter Maat et al. (2010) and Uebel et al. (2017), all of which are based on short episodic simulations geared toward various field campaigns. Inverse modeling has also been progressing toward higher resolutions, especially for regional models. For example, Wang et al. (2014) used a North American regional inversion system with meteorology from WRF run at 40 km resolution coupled to the Stochastic Time-Inverted Lagrangian Transport (STILT) model run at 1 ° × 1 ° flux resolution to evaluate the flux uncertainty reduction that could be provided by the proposed Active Sensing of CO2 Emissions over Nights, Days, and Seasons (ASCENDS) satellite mission.
The works by different research teams demonstrate the value of high-resolution CO2 simulation in characterizing carbon sources/sinks and improving carbon flux inversions. This paper presents a model development and evaluation effort to integrate the Carnegie-Ames-Stanford Approach (CASA) biogeochemical model (Randerson et al., 1996, 1997; van der Werf et al., 2006, 2010) to NASA Unified WRF (NU-WRF, Z. Tao et al., 2013, 2016; Peters-Lidard et al., 2015). It is similar to the studies by Ballav et al. (2012) and Diaz Isaac et al. (2014) but with a greater spatial/temporal scale and scope of analysis. The goal is to extend the capability of NU-WRF, an institutional regional modeling and assimilation system, to investigate carbon fluxes and transport at satellite resolved spatial scales (1–20 km).
The paper is organized as follows. Section 2 describes the modeling system and development, followed by model evaluation in section 3. Different from most existing studies whose experiments focus on relatively small regions at weeks to months’ temporal scales, this work evaluates the model performance over a relatively large region covering various landscapes, topographies, and climate zones over a relatively long period (3 years) including various environmental conditions such as drought. Conclusions are summarized in section 4.
Section snippets
Model development
There are four major components in the modeling system: 1) NASA's Goddard Earth Observing System Model, version 5 (GEOS-5) that includes the production of NASA's Modern-Era Retrospective Analysis for Research and Applications, version 2 (MERRA-2); 2) CASA and Global Fire Emissions Database (CASA-GFED); 3) Parameterized Chemistry Transport Model (PCTM); and 4) NU-WRF. Supplement Figure 1s shows the relationship of four components, in which GEOS-5/MERRA-2 provides meteorology to drive CASA-GFED
Model evaluation
NU-WRF has been applied to many investigations (e.g., Santanello et al., 2013a, b; Z. Tao et al., 2013, 2015, 2016, 2018; Zaitchik et al., 2013; Shi et al., 2014; W.-K. Tao et al., 2013, 2016; Wu et al., 2016), and has performed well in meteorological, hydrological, and atmospheric chemistry simulations. Therefore, this study focused on the evaluation of NU-WRF-CASA performance in simulating spatial and temporal (including diurnal, seasonal, and interannual) CO2 distributions. Correlation
Summary
Modeling CO2 at fine spatial resolution assists in better understanding the mechanisms and processes controlling carbon sources/sinks and atmospheric CO2 transport/variability since many of these processes occur at relatively small spatial and/or rapid temporal scales. Knowledge gained from such mechanism/process studies is anticipated to lead to reduction of uncertainty in carbon-climate interactions. In this study, we expanded the capability of the NU-WRF model to simulate CO2 transport and
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
The development and evaluation of NU-WRF-CASA was funded by NASA's Modeling, Analysis, and Prediction (MAP) program (Solicitation NNH12ZDA001N-MAP). The authors would like to thank NASA Center for Climate Simulation (NCCS) for supercomputing and data storage support, as well as NOAA's Cooperative Global Atmospheric Data Integration Project (NCGADIP) who organizes and distributes CO2 data. Thanks are also due to each individual CO2 data providers who submit their data to NCGADIP, including Beth
References (76)
- et al.
Development of high-resolution dynamic dust source function – a case study with a strong dust storm in a regional model
Atmos. Environ.
(2017) - et al.
Land information system, an interoperable framework for high resolution land surface modeling
Environ. Model. Software
(2006) - et al.
Integrated modeling of aerosol, cloud, precipitation and land processes at satellite-resolved scales
Environ. Model. Software
(2015) - et al.
Climatological mean and decadal changes in surface ocean pCO2, and net sea-air CO2 flux over the global oceans
Deep-Sea Res. II
(2009) - et al.
Impact of transpacific aerosol on air quality over the United States: a perspective from aerosol-cloud-radiation interactions
Atmos. Environ.
(2016) - et al.
Mesoscale covariance of transport and CO2 fluxes: evidence from observations and simulations using the WRF-VPRM coupled atmosphere-biosphere model
J. Geophys. Res.
(2007) - et al.
Comparing high resolution WRF-VPRM simulations and two global CO2 transport models with coastal tower measurements of CO2
Biogeosciences
(2009) - et al.
Monthly fossil-fuel CO2 emissions: mass of emissions gridded by one degree latitude by one degree longitude
Carbon Dioxide. Inf. Anal. Cent.Oak Ridge Natl. Lab.U.S. Dep. Energy
(2016) - et al.
TransCom 3 inversion intercomparison: impact of transport model errors on the interannual variability of regional CO2 fluxes, 1988–2003
Global Biogeochem. Cycles
(2006) - et al.
Carbon source/sink information provided by column CO2 measurements from the Orbiting Carbon Observatory
Atmos. Chem. Phys.
(2010)
Simulation of CO2 concentration over East Asia using the regional transport model WRF-CO2
J. Meteoro. Soc. Japan
Using continental observations in global atmospheric inversions of CO2: north American carbon sources and sinks
Tellus
Tropospheric aerosol optical thickness from the GOCART model and comparisons with satellite and Sun photometer measurements
J. Atmos. Sci.
Multi-laboratory compilation of atmospheric carbon dioxide data for the period 1957-2016; obspack_co2_1_GLOBALVIEWplus_v3.1_2017_10_18
NOAA Earth. Syst. Res.Lab.GlobaLMonit.Div.
Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model
Nature
Model-data comparison of MCI field campaign atmospheric CO2 mole fractions
J. Geophys. Res. Atmos.
On the magnitude of positive feedback between future climate change and the carbon cycle
Geophys. Res. Lett.
Implementation of noah land surface model advances in the national centers for environmental prediction operational mesoscale eta model
J. Geophys. Res.
Crop production under drought and heat stress: plant responses and management options
Front. Plant Sci.
Evaluating a 3-D transport model of atmospheric CO2 using ground-based, aircraft, and space-borne data
Atmos. Chem. Phys.
On the contribution of CO2 to the missing biospheric sink
Global Biogeochem. Cycles
Uncertainties in CMIP5 climate projections due to carbon cycle feedbacks
J. Clim.
The Modern-Era retrospective analysis for research and Applications, version 2 (MERRA-2)
J. Clim.
Global distribution and seasonality of active fires as observed with the terra and aqua moderate resolution imaging spectroradiometer (MODIS) sensors
J. Geophys. Res.
Assessing variability and long-term trends in burned area by merging multiple satellite fire products
Biogeosciences
North American CO2 exchange: intercomparison of modeled estimates with results from a fine-scale atmospheric inversion
Biogeosciences
A generalized approach to parameterizing convection combining ensemble and data assimilation techniques
Geophys. Res. Lett.
A new vertical diffusion package with an explicit treatment of entrainment processes
Mon. Weather Rev.
A new stable boundary layer mixing scheme and its impact on the simulated East Asian summer monsoon
Q. J. R. Meteorol. Soc.
Climate Change 2013: the Physical Science Basis
Global CO2 transport simulations using meteorological data from the NASA data assimilation system
J. Geophys. Res.
Simulation studies for a space-based CO2 lidar mission
Tellus
The impact of soil moisture modifications on CBL characteristics in West Africa: a case-study from the AMMA campaign
Q. J. R. Meteorol. Soc.
TransCom model simulations of hourly atmospheric CO2: experimental overview and diurnal cycle results for 2002
Global Biogeochem. Cycles
Precipitation thresholds regulate net carbon exchange at the continental scale
Nature Comunications
Detecting drought impact on terrestrial biosphere carbon fluxes over contiguous US with satellite observations
Environ. Res. Lett.
A global 9-yr biophysical land surface dataset from NOAA AVHRR data
J. Hydrometeorol.
Cited by (4)
Carbon dioxide exchange in an idealized valley
2024, Environmental Modelling and SoftwareSimulation and forecasting of atmospheric CO2 using RAMS
2023, Remote Sensing Applications: Society and EnvironmentCitation Excerpt :Their results showed a maximum mean bias of about 2.4 ppm and an RMSE of 4.7 ppm. In another study, Tao et al. (2020) used the NU-WRF model for CO2 estimation. The maximum mean bias and RMSE in their simulations are about 4 ppm and 8 ppm, respectively.
Influencing factors of PM<inf>2.5</inf> and O<inf>3</inf> from 2016 to 2020 based on DLNM and WRF-CMAQ
2021, Environmental PollutionCitation Excerpt :WRF-CMAQ, as a typical physical-chemical model, can simulate the transformation and transport of airborne pollutants with unified and detailed meteorological and topographical fields. Except for the elements in conventional meteorological observations, the meteorological fields that WRF-CMAQ applied also include zonal wind, meridional wind, precipitation, boundary layer heights, geopotential height et al. (Tao et al., 2020). The uniformity and completeness of the input meteorological fields helps decrease the errors, but errors caused by the relatively large uncertainty in air pollutant emission inventory and the relatively poor performance in simulating heavy pollution cannot be avoided at present.