ArticlePDF Available

WRF v.3.9 sensitivity to land surface model and horizontal resolution changes over North America

January 2022
Geoscientific Model Development 15(2):413-428

January 2022
15(2):413-428

DOI:10.5194/gmd-15-413-2022

License
CC BY 4.0

Authors:

Francisco José Cuesta-Valero

Helmholtz-Zentrum für Umweltforschung

J. Fidel González Rouco

Complutense University of Madrid

Show all 5 authorsHide

Understanding the differences between regional simulations of land–atmosphere interactions and near-surface conditions is crucial for a more reliable representation of past and future climate. Here, we explore the effect of changes in the model's horizontal resolution on the simulated energy balance at the surface and near-surface conditions using the Weather Research and Forecasting (WRF) model. To this aim, an ensemble of 12 simulations using three different horizontal resolutions (25, 50 and 100 km) and four different land surface model (LSM) configurations over North America from 1980 to 2013 is developed. Our results show that finer resolutions lead to higher surface net shortwave radiation and maximum temperatures at mid and high latitudes. At low latitudes over coastal areas, an increase in resolution leads to lower values of sensible heat flux and higher values of latent heat flux, as well as lower values of surface temperatures and higher values of precipitation, and soil moisture in summer. The use of finer resolutions leads then to an increase in summer values of latent heat flux and convective and non-convective precipitation and soil moisture at low latitudes. The effect of the LSM choice is larger than the effect of horizontal resolution on the near-surface temperature conditions. By contrast, the effect of the LSM choice on the simulation of precipitation is weaker than the effect of horizontal resolution, showing larger differences among LSM simulations in summer and over regions with high latent heat flux. Comparison between observations and the simulation of daily maximum and minimum temperatures and accumulated precipitation indicates that the CLM4 LSM yields the lowest biases in maximum and minimum mean temperatures but the highest biases in extreme temperatures. Increasing horizontal resolution leads to larger biases in accumulated precipitation over all regions particularly in summer. The reasons behind this are related to the partition between convective and non-convective precipitation, specially noticeable over western USA.

Subregions employed for the bias analysis, adapted from Giorgi and Francisco (2000): Central America, CAM; western North America, WNA; central North America, CNA; eastern North America, ENA; Alaska, ALA; and Greenland, GRL.

…

Regional mean annual and seasonal biases in extreme maximum and minimum temperatures ( • C) and in extreme precipitation (mm d −1 ) for all experiments and the DAYMET data product relative to the CRU database from 1980 to 2013. Extremes are calculated as the 95th percentile of the annual and seasonal temporal series at each grid cell. Biases are estimated averaging over six subregions (Fig. 1) adapted from Giorgi and Francisco (2000): Central America, CAM; western North America, WNA; central North America, CNA; eastern North America, ENA; Alaska, ALA; and Greenland, GRL.

…

Summary of the regional simulations performed in this analysis.

…

Figures - available via license: Creative Commons Attribution 4.0 International

Content may be subject to copyright.

Available via license: CC BY 4.0

Content may be subject to copyright.

Geosci. Model Dev., 15, 413–428, 2022

https://doi.org/10.5194/gmd-15-413-2022

the Creative Commons Attribution 4.0 License.

WRF v.3.9 sensitivity to land surface model and horizontal

resolution changes over North America

Almudena García-García1,2, Francisco José Cuesta-Valero1,2, Hugo Beltrami1, J. Fidel González-Rouco3, and

Elena García-Bustamante4

1Climate & Atmospheric Sciences Institute, St. Francis Xavier University, Antigonish, Nova Scotia, Canada

2Department of Remote Sensing, Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany

3Physics of the Earth and Astrophysics Department, IGEO (UCM-CSIC), Universidad Complutense de Madrid,

Madrid, Spain

4Department of Energy, Research Center for Energy, Environment and Technology (CIEMAT), Madrid, Spain

Correspondence: Hugo Beltrami (hugo@stfx.ca)

Received: 19 July 2021 – Discussion started: 26 August 2021

Revised: 26 November 2021 – Accepted: 1 December 2021 – Published: 18 January 2022

Abstract. Understanding the differences between regional

simulations of land–atmosphere interactions and near-

surface conditions is crucial for a more reliable representa-

tion of past and future climate. Here, we explore the effect of

changes in the model’s horizontal resolution on the simulated

energy balance at the surface and near-surface conditions us-

ing the Weather Research and Forecasting (WRF) model. To

this aim, an ensemble of 12 simulations using three different

horizontal resolutions (25, 50 and 100 km) and four different

land surface model (LSM) conﬁgurations over North Amer-

ica from 1980 to 2013 is developed. Our results show that

ﬁner resolutions lead to higher surface net shortwave radia-

tion and maximum temperatures at mid and high latitudes.

At low latitudes over coastal areas, an increase in resolution

leads to lower values of sensible heat ﬂux and higher values

of latent heat ﬂux, as well as lower values of surface temper-

atures and higher values of precipitation, and soil moisture in

summer. The use of ﬁner resolutions leads then to an increase

in summer values of latent heat ﬂux and convective and non-

convective precipitation and soil moisture at low latitudes.

The effect of the LSM choice is larger than the effect of hori-

zontal resolution on the near-surface temperature conditions.

By contrast, the effect of the LSM choice on the simulation

of precipitation is weaker than the effect of horizontal resolu-

tion, showing larger differences among LSM simulations in

summer and over regions with high latent heat ﬂux. Compar-

ison between observations and the simulation of daily max-

imum and minimum temperatures and accumulated precip-

itation indicates that the CLM4 LSM yields the lowest bi-

ases in maximum and minimum mean temperatures but the

highest biases in extreme temperatures. Increasing horizon-

tal resolution leads to larger biases in accumulated precipi-

tation over all regions particularly in summer. The reasons

behind this are related to the partition between convective

and non-convective precipitation, specially noticeable over

western USA.

1 Introduction

Most assessments of climate change impacts on ecosys-

tems and societies are based on projections performed by

regional climate models (RCMs) and/or earth system mod-

els (ESMs; IPCC, 2013, 2019). Exploring inter-model dif-

ferences in present climate simulations is necessary to un-

derstand their contribution to the spread in future climate

projections and ultimately to better characterize or even re-

duce the uncertainty in the simulation of the response to a

given scenario (Cubasch et al., 2013). Understanding inter-

model differences is also important for paleoclimatic stud-

ies relying on regional climate model simulations to bridge

the gap between the local character of proxy reconstructions

and ESM global simulations (e.g., PALEOLINK; Gómez-

Navarro et al., 2018).

The representation of land–atmosphere interactions within

climate models has received considerable attention during

Published by Copernicus Publications on behalf of the European Geosciences Union.

414 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

the last decades due to their inﬂuence on surface climate,

vegetation and soil hydrology, and therefore, on climate vari-

ability (e.g., Lorenz et al., 2016; Vogel et al., 2017). For ex-

ample, energy and water exchanges between the lower atmo-

sphere and the ground surface have been shown to alter sur-

face conditions, particularly during extreme weather events

in summer (Seneviratne et al., 2006; Hirschi et al., 2011; Mi-

ralles et al., 2012; Hauser et al., 2016). Land–atmosphere

interactions have also been studied in the evaluation of cli-

mate model simulations, applying several metrics to char-

acterize surface energy ﬂuxes and near-surface conditions

(Koven et al., 2013; Dirmeyer et al., 2013; Sippel et al., 2017;

García-García et al., 2019).

The representation of near-surface conditions (e.g., air and

soil temperatures, soil moisture, etc.) and energy and water

exchanges at the land surface in a climate model depends on

the processes simulated by the atmospheric and soil model

components, as well as on the degree of coupling imple-

mented between both model components (Koster et al., 2006;

Melo-Aguilar et al., 2018). Different land surface models

(LSMs) include varying levels of realism in the represen-

tation of soil physics, land cover type, soil water content,

snow cover, drip, runoff or inﬁltration. Thus, each LSM sim-

ulates somewhat different surface water and energy ﬂuxes

(e.g., Lawrence et al., 2019). For example, the representa-

tion of surface albedo, evaporative resistance and aerody-

namic roughness by each LSM alters the simulation of the

surface energy balance and consequently affects the evolu-

tion of surface temperatures (Laguë et al., 2019; MacDougall

and Beltrami, 2017). The dependence of the simulated land–

atmosphere interactions on the LSM has been shown in pre-

vious studies using global (García-García et al., 2019) and

regional climate model simulations over North America (Pei

et al., 2014; García-García et al., 2020). There are examples

of these studies at local scales (Mooney et al., 2013; Whar-

ton et al., 2013; Chen et al., 2014; Van Den Broeke et al.,

2018; Liu et al., 2019; Zhuo et al., 2019) and at continen-

tal scales for Europe (Davin and Seneviratne, 2012; Mooney

et al., 2013).

Horizontal resolution is another factor to take into ac-

count in the study of land–atmosphere interactions in cli-

mate model simulations. ESMs are limited by computational

resources, using horizontal resolutions from approximately

250 to 100 km (e.g., CMIP5 models; Taylor et al., 2012),

while RCMs allow for using much ﬁner resolutions. Anal-

ogously, the range of horizontal resolutions employed in

RCMs for climate studies is usually limited to approximately

50–25 km (e.g., CORDEX models; Giorgi and Gutowski,

2015). In spite of this enhanced resolution, an RCM’s abil-

ity to reproduce temporal variability like that of precipitation

at daily timescales is still limited, being greatly improved

by using resolutions of ∼4 km and convection-resolving

RCMs (Sun et al., 2016). Previous studies have shown some

resolution-induced improvements in the simulation of pre-

cipitation, wind and high-altitude temperatures at local and

regional scales with possible implications for the simula-

tion of climate dynamics (Ban et al., 2014; Gómez-Navarro

et al., 2015; Messmer et al., 2017; Hahmann et al., 2020;

Vegas-Cañas et al., 2020). Small-scale weather phenomena

such as sea breezes, snowstorms induced by the presence

of lakes, local winds, tropical cyclones and mesoscale con-

vective systems can be better represented with increased res-

olution (Wehner et al., 2010). Some studies have also sug-

gested a resolution-induced improvement in the representa-

tion of interactions between small- and large-scale dynami-

cal processes, ultimately leading to better large-scale atmo-

spheric ﬂow (Lucas-Picher et al., 2017). Thus, the difference

in the resolution employed in RCMs and ESMs is expected to

improve the representation of land–atmosphere interactions

in RCMs through a more adequate discretization of equa-

tions, as well as through a more realistic representation of

small-scale processes and topographical features (Xue et al.,

2014; Rummukainen, 2016; Vegas-Cañas et al., 2020). The

improved representation of land–atmosphere interactions as-

sociated with ﬁner resolutions is also expected to induce an

improvement in the simulation of near-surface conditions,

especially in the simulation of extreme events (Prein et al.,

2013; Di Luca et al., 2015; Rummukainen, 2016; Demory

et al., 2014).

Although the literature on the impact of the LSM choice

and changes in resolution on climate simulations is extensive,

most studies are focused on small domains and meteorolog-

ical events, providing little information about the impact of

the LSM choice and horizontal resolution changes on long-

term climatological variability. Here, we evaluate and com-

pare the inﬂuence of both factors, the LSM choice and hori-

zontal resolution, on the representation of energy and water

ﬂuxes at the surface and consequently on the simulation of

near-surface conditions over North America for a climato-

logical period spanning the time interval 1980–2013. We use

the Weather Research and Forecasting (WRF) model (ver-

sion 3.9; Skamarock et al., 2008) that allows for testing a

number of LSM schemes, each varying in physical parame-

terization complexities. An ensemble of 12 simulations was

generated using three different horizontal resolutions (25, 50

and 100 km), four different LSM conﬁgurations and two veg-

etation options, i.e., prescribed or dynamical vegetation.

The WRF representation of climate at different resolutions

is expected to affect the simulation of atmospheric and sur-

face phenomena in several ways, through a different rep-

resentation of cloud formation or through a different level

of details in the description of orography and land cover.

Simulations with different LSMs provide information about

the advantages of using a comprehensive LSM, such as the

CLM4 LSM (Oleson et al., 2010), or a simpler LSM com-

ponent, such as the NOAH LSM (Tewari et al., 2004). Ad-

ditionally, the comparison of two identical model conﬁgura-

tions except for the prescribed or dynamical vegetation mode

provides information about the effect of a realistic evolution

of vegetation cover on the energy and water balance at the

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 415

surface. Thus, soil and near-surface variables from these 12

simulations are compared and also evaluated against temper-

ature and precipitation observations to explore two questions.

(1) How do changes in horizontal resolution within WRF af-

fect the simulation of land–atmosphere interactions and near-

surface conditions? (2) How do LSM differences in the simu-

lation of land–atmosphere interactions translate into different

near-surface conditions?

The descriptions of the WRF experiments and the method-

ology applied for the analysis are included in Sects. 2 and 3,

respectively. Section 4 presents the results of the analysis,

which are discussed using the available literature in Sect. 5.

The conclusions and importance of this work are summarized

in Sect. 6.

2 Description of the modelling experiment

We performed three sets of regional simulations (12 simula-

tions in total) over North America from 1980 to 2013 using

the Advanced Research WRF model (WRF v3.9; Skamarock

et al., 2008) with initial and boundary conditions from

the North American Regional Reanalysis product (NARR;

Mesinger et al., 2006). The NARR product was generated by

the National Centers for Environmental Prediction (NCEP)

Eta atmospheric model (Janjic, 1997), the NOAH LSM com-

ponent (Mitchell, 2005) and the Regional Data Assimilation

System (RDAS; Mesinger et al., 2006). The NARR data were

obtained from the National Center for Environmental Infor-

mation (NOAA) archive and provides data over a 32 km grid

with a 3-hourly temporal resolution. No nudging techniques

were applied within the domain of the simulation. The use

of nudging techniques imposes large-scale variability within

the inner scales of the domain of simulation, thus partially

muting the generation of local- and regional-scale dynami-

cal responses in favour of representing the large scale ﬂow

of the driving conditions. Avoiding nudging, therefore, al-

lows for more clearly expressing the inﬂuences of increased

resolution and of changing the LSM component (von Storch

et al., 2000).

The three sets of simulations were performed using a

Lambert-type projection with resolutions of 25×25 km, 50×

50 km and 100 ×100 km. Each set includes four simulations

using three different LSM components: the NOAH LSM

(NOAH; Tewari et al., 2004), the NOAH LSM with multi-

parameterizations options (NOAH-MP; Niu et al., 2011) and

the Community Land Model version 4 LSM (CLM4; Oleson

et al., 2010). The fourth simulation included in each set was

performed using the NOAH-MP LSM with dynamic vegeta-

tion (NOAH-MP-DV), while vegetation was prescribed for

the other simulations. The rest of the WRF options remained

the same for all simulations, employing 27 atmospheric lev-

els, land cover categories from the Moderate Resolution

Imaging Spectroradiometer (MODIS; Barlage et al., 2005),

the WRF single moment (WSM) 6-class graupel scheme for

the microphysics (Hong and Lim, 2006), the Grell–Freitas

ensemble scheme (Grell and Freitas, 2014), the Yonsei Uni-

versity scheme for the description of the planetary bound-

ary layer scheme (YSU; Hong et al., 2006), the revised

MM5 Monin–Obukhov scheme (Jiménez et al., 2012) and

the Community Atmosphere Model (CAM) scheme (Collins

et al., 2004). The use of different horizontal resolutions re-

quires the use of different time steps for performing our

WRF simulations, as well as different time intervals for com-

puting radiation physics (radt option in WRF namelist). Ta-

ble 1 summarizes the differences between all simulations em-

ployed in this analysis.

The LSM schemes used in this study differ in parameteri-

zation complexity, the NOAH being the most basic amongst

the three selected LSMs. The NOAH LSM describes soil and

vegetation processes for the closure of the water and energy

budgets, discretizing the soil into four soil layers that reach a

total of 2 m of depth (Tewari et al., 2004). Some limitations

have been associated in this soil model with the represented

bulk layer of canopy, snow and soil, its system to drain water

at the bottom of the soil column, and its simple snowmelt dy-

namics (Wharton et al., 2013). The NOAH-MP version of the

NOAH LSM improves soil hydrology and the representation

of terrestrial biophysical processes (Niu et al., 2011). This

scheme includes a separate vegetation canopy with a compre-

hensive description of vegetation properties. The NOAH-MP

also includes a multi-layer snow pack description with liquid

water storage and melting and refreezing capabilities. It in-

cludes the same soil structure as the NOAH LSM (four soil

layers, down to 2 m). The CLM4 LSM incorporates a com-

prehensive representation of biogeophysics and hydrology,

including a single-layer vegetation canopy, a ﬁve-layer snow-

pack and a 10-layer soil column down to a depth of 3.802m.

This scheme also characterizes each grid cell into ﬁve pri-

mary sub-grid land cover types (glacier, lake, wetland, urban

and vegetated), using up to four plant functional types (PFTs)

to describe vegetation physiology and structure.

3 Methodology

We study the impact of changing horizontal resolution on

surface energy ﬂuxes and near-surface conditions as simu-

lated by the WRF model with different LSM components.

For this purpose, we estimated the temporal averages of sur-

face energy ﬂuxes for the analysis period (1980–2013) fo-

cusing on the following energy components: net shortwave

radiation (SNET, Wm−2), net longwave radiation (LNET,

Wm−2), net radiation absorbed by the soil (RNET, Wm−2),

latent heat ﬂux (LH, Wm−2), sensible heat ﬂux (HFX,

Wm−2) and ground heat ﬂux (GHF, Wm−2). The temporal

averages of near-surface conditions are estimated using out-

puts of 2 m air temperatures (SAT, ◦C), daily maximum SAT

(TASMAX, ◦C), daily minimum SAT (TASMIN, ◦C), soil

temperature at 1 m depth (GST 1 m, ◦C), accumulated con-

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

416 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

Table 1. Summary of the regional simulations performed in this analysis.

Name LSM Resolution Vegetation mode Simulation time step Radiation time step

NOAH 25 km 25 km Prescribed 2.5 min 6 min

NOAH 50 km NOAH 50 km Prescribed 5 min 20 min

NOAH 100 km 100 km Prescribed 10 min 20 min

NOAH-MP 25 km 25 km Prescribed 2.5 min 6 min

NOAH-MP 50 km NOAH-MP 50 km Prescribed 5 min 20 min

NOAH-MP 100 km 100 km Prescribed 10 min 20 min

NOAH-MP-DV 25 km 25 km Dynamic 2.5 min 6 min

NOAH-MP-DV 50 km NOAH-MP 50 km Dynamic 5 min 20 min

NOAH-MP-DV 100 km 100 km Dynamic 10 min 20 min

CLM4 25 km 25 km Prescribed 2.5 min 6 min

CLM4 50 km CLM4 50 km Prescribed 5 min 20 min

CLM4 100 km 100 km Prescribed 10 min 20 min

vective and non-convective precipitation at the surface (PRE-

CIP C and PRECIP NC, mmd−1), soil moisture contained in

the ﬁrst soil metre (SM 1 m, m3m−3), and total cloud fraction

(TCLDFR, %). All values are computed using the annual and

seasonal (boreal winter, DJF; summer, JJA) averages over the

34-year period (1980–2013) after discarding the ﬁrst year of

the simulation (1979) as spin-up, which is enough to avoid

the effect of initial conditions (García-García et al., 2020).

Thus, we estimated the anomalies of all outputs for each

LSM simulation relative to the multi-model mean (the mean

of CLM4, NOAH, NOAH-MP and NOAH-MP-DV outputs)

for each set of simulations with different resolution (25, 50

and 100 km). Similarly, we estimated the change in the sim-

ulation of all variables between the 100 and 50 km simula-

tions and between the 50 and 25 km simulations for all LSM

conﬁgurations. When required, outputs of all WRF experi-

ments were mapped to the grid of the observational refer-

ence employed in this study by selecting the nearest model

grid point. A Student’s ttest considering autocorrelation was

used to identify signiﬁcant differences between simulations

with different LSMs and horizontal resolutions at the 95 %

conﬁdence level.

Additionally, we evaluate the WRF performance in the

simulation of maximum and minimum air temperature and

accumulated precipitation against the Climatic Research

Unit Time-Series product version 4.03 (CRU; Harris et al.,

2014). The CRU database provides monthly data at a res-

olution of approximately 50 km. Previous studies have re-

ported inconsistencies between different observational prod-

ucts, particularly important for the evaluation of model sim-

ulations with different resolutions (e.g., Iles et al., 2020). To

avoid this issue, we included another observational database

in the analysis; the Daily Surface Weather Data version 3

(DAYMET; Thornton et al., 2016), with daily data at approx-

imately 1 km resolution over North America. This allows for

the comparison of uncertainties associated with the choice of

Figure 1. Subregions employed for the bias analysis, adapted from

Giorgi and Francisco (2000): Central America, CAM; western

North America, WNA; central North America, CNA; eastern North

America, ENA; Alaska, ALA; and Greenland, GRL.

the observational product and the uncertainties arising from

the model conﬁguration. We calculated annual and seasonal

WRF and DAYMET biases in these variables relative to the

CRU data for the analysis period averaged over six subre-

gions due to the large climate differences over North Amer-

ica. These six subregions cover Central and North Amer-

ica (NA) and are adapted from Giorgi and Francisco (2000):

Central America, CAM; western North America, WNA; cen-

tral North America, CNA; eastern North America, ENA;

Alaska, ALA; and Greenland, GRL (Fig. 1). The impact of

horizontal resolution is expected to be larger on the simula-

tion of extreme events than on surface climatologies (Prein

et al., 2013; Di Luca et al., 2015; Rummukainen, 2016). We

examined this by calculating the bias in the 95th percentile of

maximum and minimum temperatures and accumulated pre-

cipitation within all experiments using the DAYMET product

as reference.

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 417

4 Results

4.1 LSM inﬂuences on surface energy ﬂuxes and

near-surface conditions

The net radiation absorbed by the ground surface enhances

turbulent (latent and sensible) ﬂuxes at the surface and/or

warms the soil surface, which leads to an increase in the

emitted longwave radiation (Bonan, 2002). The relationship

between these variables is shown by their corresponding en-

semble mean of LSM simulations, indicating similar lati-

tudinal patterns in net radiation, turbulent ﬂuxes and near-

surface temperatures with higher ﬂuxes and temperatures at

lower latitudes (see for example Fig. 2 for the LSM ensem-

ble mean of the 50 km experiments). The net radiation results

from adding net shortwave and longwave radiations, whose

mean values have similar spatial distributions but with op-

posite sign (Fig. 2a). This indicates that more shortwave ra-

diation reaches the land surface than is reﬂected due to sur-

face albedo, while the radiation emitted from the soil due to

its surface temperature, is higher than the longwave radia-

tion reaching the soil surface (Fig. 2a). The energy propor-

tion of net radiation that is propagated through the soil is

much smaller than the rest of surface energy ﬂuxes (Fig. 2b).

Areas with high latent heat ﬂux coincide with areas with

high convective precipitation rates at low and middle lati-

tudes (Fig. 2b and d). The soil moisture map shows high val-

ues southward in the Great Lakes region and low values in

dry areas such as Florida and the southwest USA, in agree-

ment with both convective and non-convective precipitation

(Fig. 2d and e).

The use of a different LSM component in WRF affects

the representation of soil and vegetation properties and pro-

cesses in the simulation, resulting in noticeable differences

in the simulated energy ﬂuxes across the domain (Fig. 3).

For instance, the spatial pattern of the LSM anomalies rel-

ative to the multi-model mean in longwave net radiation is

similar to the minimum temperature anomalies (Figs. 3b and

4b), and the anomalies in latent heat ﬂux show similar LSM

differences to the convective precipitation maps at low lati-

tudes (Figs. 3d and 5a). The spatial patter of the LSM anoma-

lies in sensible and latent heat ﬂuxes shows opposite val-

ues around the mean. For example over the boreal forest,

the LSMs reaching the highest values of latent heat ﬂux also

reach the lowest values of sensible heat ﬂux at the same lo-

cations (Fig. 3d and f).

LSM differences in the simulation of surface ﬂuxes and

near-surface conditions are similar among the experiments

with different resolutions (Figs. 3–5 and S2, S4 and S6 in the

Supplement). For example, based on the differences between

each LSM simulation with 50 km resolution and the 50 km

multi-model mean (Fig. 2), we can identify the CLM4 as the

LSM component simulating the highest net shortwave radia-

tion over most of North America for the annual, DJF and JJA

means (SNET in Figs. 3a and S1 in the Supplement). Mean-

Figure 2. Ensemble mean of the LSM simulations for the sur-

face energy ﬂuxes and near-surface conditions: net shortwave radia-

tion (SNET), net longwave radiation (LNET), surface net radiation

(RNET) (a); latent heat ﬂux (LH), sensible heat ﬂux (HFX), ground

heat ﬂux (GHF) (b); maximum temperature (TASMAX), minimum

temperature (TASMIN), surface air temperature (SAT), soil tem-

perature at 1 m depth (GST 1 m) (c); accumulated convective and

non-convective precipitation at the surface (PRECIP C and PRECIP

NC) (d); soil moisture contained in the ﬁrst soil metre (SM 1 m) (e);

and total cloud cover fraction (TCLDFR) (f) for the WRF ensem-

ble mean. GHF results are also represented using its own colour

scale (g). Mean values are estimated as the temporal average for the

period 1980–2013 using simulations performed with 50 km resolu-

tion.

while, the NOAH-MP-DV simulation reaches the lowest net

shortwave radiation over the same areas (Fig. 3a). The WRF

simulation of net longwave radiation reaches negative values

(Fig. 2a) with the maximum values simulated by the NOAH

LSM and the minimum values simulated by the CLM4 LSM

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

418 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

Figure 3. Anomalies of energy ﬂuxes at the surface: (a) net short-

wave radiation (SNET), (b) net longwave radiation (LNET), (c) soil

net radiation (RNET), (d) latent heat ﬂux (LH), (f) sensible heat ﬂux

(HFX) and (g) ground heat ﬂux (GHF) for each LSM simulation

relative to the LSM ensemble mean. Mean values are estimated as

the temporal average for the period 1980–2013 using simulations

performed with 50 km resolution. Grid cells with a non-signiﬁcant

change at the 5 % signiﬁcance level are masked in grey.

(Fig. 3b). The upward (negative) component of the net long-

wave radiation results from the Stefan–Boltzmann equation,

from which the outgoing longwave radiation LW ↑is propor-

tional to σ T 4

s, where Tsis surface temperature and σis the

Stefan–Boltzmann constant. Thus, the CLM4 simulation pro-

duces the largest outgoing longwave radiation (see Fig. S7

in the Supplement), shown in Fig. 3b by the largest nega-

tive anomalies and therefore the highest air and soil temper-

atures (Fig. 4). The opposite behaviour is performed by the

NOAH LSM, yielding the lowest upward longwave radia-

tion (Figs. 3b and S7), and one of the coldest temperature

climatologies relative to the multi-model mean (Fig. 4). The

relationship between the radiation and temperature anoma-

Figure 4. Anomalies of near-surface temperature conditions: (a)

daily maximum temperature (TASMAX), (b) daily minimum tem-

perature (TASMIN), (c) surface air temperature (SAT) and (d) soil

temperature at 1 m depth (GST 1 m) for each LSM simulation rel-

ative to the LSM ensemble mean. Mean values are estimated as

the temporal average for the period 1980–2013 using simulations

performed with 50 km resolution. Grid cells with a non-signiﬁcant

change at the 5 % signiﬁcance level are masked in grey.

lies is also supported by high spatial correlation coefﬁcients

(Table S1 in the Supplement). These correlation coefﬁcients

show the link of maximum temperatures to both shortwave

and longwave net radiation, particularly in summer, while

minimum temperatures show higher correlation with long-

wave net radiation than with shortwave net radiations in most

of the LSM simulations (Table S1).

The simulation of sensible heat ﬂux reaches the highest

values using the CLM4 conﬁguration over the boreal for-

est and the lowest values in western USA. Meanwhile, the

NOAH simulation reaches the lowest sensible heat ﬂuxes

over the boreal forest and the highest values in western USA

(Fig. 3e). The spatial patterns of LSM anomalies in sensi-

ble heat ﬂux are similar to the LSM anomalies in net short-

wave radiation and daily maximum temperatures (Figs. 3a

and 4a), which is also shown by high spatial correlation coef-

ﬁcients, particularly in summer (Table S1). LSM differences

in ground heat ﬂux are smaller than for the rest of the en-

ergy ﬂuxes due to the small magnitude of the GHF in com-

parison with the rest of energy components (Fig. 2g). The

NOAH LSM reaches the lowest ground heat ﬂux values for

the annual mean in most of the domain, showing nonethe-

less the highest ground heat ﬂux values in summer at high

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 419

Figure 5. Anomalies of near-surface humid conditions: (a) accumu-

lated convective and (b) non-convective precipitation at the surface

(PRECIP C and PRECIP NC), (c) soil moisture contained in the

ﬁrst soil metre (SM 1 m), and (d) total cloud fraction (TCLDFR)

for each LSM simulation relative to the LSM ensemble mean. Mean

values are estimated as the temporal average for the period 1980–

2013 using simulations performed with 50 km resolution. Grid cells

with a non-signiﬁcant change at the 5 % signiﬁcance level are

masked in grey.

latitudes (Figs. 3f and S1). The spatial pattern of LSM dif-

ferences in ground heat ﬂux differ from the soil temperature

results, whose spatial correlation coefﬁcients are higher with

the longwave net radiation mainly in summer. LSM differ-

ences are larger for the simulation of soil temperatures than

for the simulation of air temperatures particularly at high lati-

tudes in summer where LSM differs largely in the simulation

of shortwave net radiation probably due to different estimates

of surface albedo under different land cover and soil moisture

values (Figs. 2a, 4c and d, 5c, and S3 in the Supplement).

The CLM4 simulation produces the highest latent heat

ﬂux values and convective precipitation rates, particularly

over southwestern NA, while the NOAH simulation provides

the lowest latent heat ﬂux and convective precipitation val-

ues over the same areas (Figs. 3d and 5a). LSM differences

in latent heat ﬂux, convective and non-convective precipita-

tion are larger in summer (Figs. S1 and S5 in the Supple-

ment). LSM differences in JJA convective precipitation rates

are particularly large at low latitudes, where the CLM4 LSM

produces high latent heat ﬂux over the western part of the

domain and low latent heat ﬂux over the eastern NA rela-

tive to the multi-model mean (Figs. S1 and S5). These differ-

ences between the east and west in the CLM4 simulation of

latent heat ﬂux are not reﬂected in the values of convective

precipitation rates, hence the low spatial correlation coefﬁ-

cients between both variables (Table S2 in the Supplement).

There are also LSM differences in the non-convective term

of precipitation, with larger anomalies in summer than in

winter (Figs. 5b and S5). Thus, the NOAH LSM produces

the highest precipitation anomaly at mid-latitudes, where the

same LSM produces high values of total cloud fraction rela-

tive to the multi-model mean (Fig. 5b and d). This relation-

ship between non-convective precipitation and cloud cover

is also shown by high spatial correlation coefﬁcients (Ta-

ble S2). There are large LSM differences in the simulation

of soil moisture, with the CLM4 LSM generating the driest

conditions and the NOAH-MP experiments the wettest soils

(Figs. 5c and S5). Although the LSM differences are large in

the representation of humid conditions, the relationship be-

tween precipitation, latent heat ﬂux, soil moisture and cloud

fraction is weaker than the relationship between radiation and

temperatures in the WRF experiments.

Larger LSM differences in the simulation of shortwave

and longwave radiations and temperatures are found over

very vegetated areas such as the boreal forest and eastern

USA (Figs. 3 and 4). The comparison of the NOAH-MP

experiments with prescribed and dynamic vegetation also

shows larger differences in air temperatures over these areas.

This suggests that the different representation of vegetation

in each LSM yields to different estimates of soil properties,

such as surface albedo, evaporative resistance and aerody-

namic roughness, affecting the simulation of radiation and

temperatures. The LSM differences in latent heat ﬂux and

convective precipitation are larger at low latitudes, with the

NOAH LSM yielding the most different values relative to the

rest of LSMs. This can be associated with the issues identi-

ﬁed in the NOAH LSM to simulate soil hydrology (Wharton

et al., 2013).

4.2 Resolution impact on surface energy ﬂuxes and

near-surface conditions

The response of surface energy ﬂuxes and near-surface con-

ditions to changes in spatial resolution varies considerably

with the season, while they behave similarly using different

LSM components (Figs. 6 and S8–S13 in the Supplement).

We consider the results of the NOAH-MP-DV experiments

as an example to minimize the number of ﬁgures in the pa-

per. Changes in resolution alter the surface energy ﬂuxes in

DJF mainly over regions of complex topography and coastal

areas, while in JJA the simulation of energy ﬂuxes is affected

by resolution over the whole domain except over a region in

the central USA and northern areas of Hudson Bay (Fig. 6).

In JJA, the use of coarser horizontal resolutions induces a

decrease in the net radiation absorbed by soil at high lati-

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

420 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

Figure 6. Seasonal mean difference in surface energy ﬂuxes be-

tween the 100 and 50 km simulations (left) and between the 50

and 25 km simulations (right). All outputs are from the NOAH-MP-

DV simulations for the period 1980–2013. Grid cells with a non-

signiﬁcant change at the 5 % signiﬁcance level are masked in grey.

All outputs from the 25, 50 and 100 km simulations were mapped

to a common grid (CRU grid) using the nearest model grid point.

tudes, mainly caused by a decrease in net shortwave radia-

tion (Fig. 6a and c). The decrease in net radiation induced by

coarser resolutions limits the energy available for turbulent

energy ﬂuxes at high latitudes. Thus, values of latent and sen-

sible heat ﬂuxes also decrease at coarser resolutions at high

latitudes (Fig. 6d and e). The use of coarser resolutions also

induces higher sensible heat ﬂuxes at low latitudes, which is

balanced by lower latent heat ﬂux (Fig. 6d and e).

Changes in downward shortwave radiation at the surface

are mainly driven by changes in cloud cover, but changes in

atmospheric water vapour and aerosols may also affect the

shortwave radiation reaching the ground surface (Hatzianas-

tassiou et al., 2005). Changes in surface albedo also lead to

Figure 7. Seasonal mean difference in near-surface temperature

conditions between the 100 and 50 km simulations (left) and be-

tween the 50 and 25 km simulations (right). All outputs are from

the NOAH-MP-DV simulations for the period 1980–2013. Grid

cells with a non-signiﬁcant change at the 5 % signiﬁcance level are

masked in grey. All outputs from the 25, 50 and 100 km simulations

were mapped to a common grid (CRU grid) using the nearest model

grid point.

changes in the upward component of shortwave radiation,

thus affecting SNET. The resolution differences in estimat-

ing albedo and land cover in our results are smaller than those

associated with the cloud cover and the microphysics of the

model, particularly in JJA. Thus, the effect of resolution on

the downward component of shortwave radiation, which is

dependent on cloud formation and atmospheric composition,

is larger than the resolution effect on the upward component

of shortwave radiation, which is dependent on albedo val-

ues (Fig. S14 in the Supplement). The resolution changes in

winter SNET at southern latitudes of the Rocky Mountains

that are not related to cloud cover seem to be associated with

the upward component of shortwave radiation and therefore

with the resolution effect on surface albedo values (Figs. 6a,

8d and S14).

Consistent with the effect of LSM differences on near-

surface conditions (Sect. 4.1), the spatial pattern of the res-

olution impact on net shortwave radiation is similar to the

resolution-induced changes in daily maximum temperatures

(Figs. 6a and 7a). The response of minimum temperature to

changing resolution is, however, smaller than for maximum

temperatures at middle–high latitudes (Fig. 7b). Over eastern

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 421

Figure 8. Seasonal mean difference in near-surface humid condi-

tions between the 100 and 50 km simulations (left) and between the

50 and 25 km simulations (right). All outputs are from the NOAH-

MP-DV simulations for the period 1980–2013. Grid cells with a

non-signiﬁcant change at the 5 % signiﬁcance level are masked in

grey. All outputs from the 25, 50 and 100km simulations were

mapped to a common grid (CRU grid) using the nearest model grid

point.

North America, JJA minimum temperature increases with the

use of coarser resolutions, while it decreases over western

North America (Fig. 7b). The response of mean temperature

to resolution is mainly driven by the resolution impact on

maximum temperatures since both show similar spatial pat-

terns (Fig. 7c). Although the impact of resolution on air and

soil temperatures shows similar spatial patterns, in winter air

temperatures are more affected by changes in resolution than

soil temperatures, particularly at high latitudes and elevation

where snow may be present (Fig. 7c and d). Thus, soil tem-

peratures are more sensible to JJA changes in the energy bud-

get induced by reducing resolution, while in DJF soil temper-

ature remains insulated from resolution-induced changes in

surface conditions probably because of the insulating effect

of snow cover (García-García et al., 2019).

Non-convective precipitation increases with the use of

coarser horizontal resolutions over the Rocky Mountains,

particularly in winter, where the model also represents a

higher percentage of cloud cover with coarser resolutions

(Fig. 8b and d). Over the east coast of the USA, the use of

coarser resolutions leads to lower non-convective precipita-

tion rates (Fig. 8b). This behaviour is also present in JJA

over the arctic areas of our domain (Fig. 8b). Although the

response of convective precipitation to resolution in winter

is not signiﬁcant, in JJA the use of coarser horizontal res-

olutions yields a decrease in convective precipitation over

coastal areas and the Rocky Mountains, where the simulation

also reaches lower latent heat ﬂux values (Figs. 6d and 8a).

Although the spatial pattern of soil moisture is very patchy in

DJF and JJA, soil moisture tends to decrease at low latitudes

in JJA with the use of coarser resolutions (Fig. 8c). At mid-

latitudes, however, the use of coarser resolutions leads to an

increase in soil moisture during the year at most locations

(Fig. 8c).

In summary, at low NA latitudes the JJA values of the

three variables associated with the surface water balance

(LH, PRECIP and SM 1 m) decrease with the use of coarser

horizontal resolutions, although showing large spatial vari-

ability (Figs. 6d and 8a and c). At middle and high NA lati-

tudes, there are differences in the response of the water bal-

ance variables to the use of coarser resolutions. For exam-

ple, soil moisture increases with coarser resolutions over a

large area at mid-latitudes, while convective precipitation in-

creases just over a few grid cells in central NA, decreasing

over most coastal areas (Fig. 8a and c). Latent heat ﬂux de-

creases with the use of coarser resolutions over most regions

at high and middle latitudes, particularly over coastal areas

(Fig. 6d).

4.3 Comparison of temperature and precipitation

against observations

For the comparison of the effect of LSM and horizontal reso-

lution changes on climate simulations, we estimate the bias in

all WRF simulations relative to the CRU observational prod-

uct (Harris et al., 2014). As a measure of the possible uncer-

tainties in observational databases, we also estimate the bias

in the DAYMET product relative to the CRU data. The in-

consistencies between both observational data are noticeably

smaller than the biases in the WRF experiments for all vari-

ables in all regions, except for minimum temperatures in the

CAM region and for precipitation in the ALA region (Fig. 9,

see a representation of the regions in Fig. 1).

The WRF model underestimates annual and seasonal

means of daily maximum temperatures over most of North

America at all resolutions, compared grid cell by grid cell

(Fig. S15 in the Supplement) and on average over subdo-

mains (Fig. 9a). These biases are generally less pronounced

for the experiments using the CLM4 LSM at most loca-

tions and in all seasons. The impact of horizontal resolution

on these values is weaker than the LSM dependence over

each domain, showing a greater effect of resolution on max-

imum temperatures in summer than in winter over western

North America and at high latitudes (WNA, ALA and GRL

in Fig. 9a). Over these areas, ﬁner horizontal resolutions are

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

422 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

Figure 9. Regional mean annual and seasonal biases in maximum and minimum temperatures (◦C) and in accumulated precipitation

(mmd−1) for all experiments and the DAYMET data product relative to the CRU database from 1980 to 2013. Biases are estimated av-

eraging over six subregions (Fig. 1) adapted from Giorgi and Francisco (2000): Central America, CAM; western North America, WNA;

central North America, CNA; eastern North America, ENA; Alaska, ALA; and Greenland, GRL.

associated with warmer JJA maximum temperatures, reduc-

ing the bias relative to the CRU dataset at middle latitudes

(Figs. 7a and 9a). In the ALA and GRL regions, the WRF

model with the CLM4 and the NOAH-MP LSM compo-

nents overestimates JJA maximum temperatures, increasing

the bias in these simulations with the use of ﬁner resolutions

(Fig. 9a). The WRF bias in maximum temperatures in win-

ter is greatly improved over the boreal forest and the Rocky

Mountains areas by using the CLM4 as LSM (Fig. S15).

Over the same areas the CLM4 simulated very high values

of shortwave net radiation and sensible heat ﬂux in compar-

ison with the rest of the land surface models (Fig. 3a and

e), which may be related to the CLM4 albedo estimate. De-

spite the WRF underestimation of mean maximum temper-

ature, extreme maximum temperatures are overestimated by

the WRF model, particularly at high latitudes and using the

CLM4 LSM (Fig. 10a). As expected based on the litera-

ture, the resolution effect on the bias in extreme maximum

temperatures is larger than on the bias in mean maximum

temperatures; however LSM differences are still larger than

resolution-induced changes.

The performance of the WRF model in reproducing

daily minimum temperatures from the CRU observations is

slightly better than reproducing maximum temperatures at

middle and low latitudes, but it is worse at high latitudes par-

ticularly in DJF (Fig. 9b). Experiments using the CLM4 LSM

yield a warmer climatology over most areas and for all sea-

sons than the experiments with the other LSM components,

implying smaller biases in the CLM4 simulations for most of

the regions (Fig. 9b). The WRF bias in minimum temperature

is large in winter over the central and eastern areas of North

America and at middle and high latitudes (subdomains ALA,

GRL, CNA and ENA in Fig. 9b). The resolution impact on

these results is again weaker than the effect of the LSM com-

ponent. In summer, the WRF-NOAH experiments show a

large negative bias in minimum temperatures over the NA

southeastern coast (Fig. S16 in the Supplement), areas where

the same experiments also shown high values of longwave

net radiation and low values of sensible heat ﬂux (Fig. 3b and

e). The simulation of extreme minimum temperatures is also

overestimated for all experiments and regions except for the

CAM region in summer, showing particularly large biases in

winter (Fig. 10b). Thus, although the WRF model underes-

timates mean maximum and minimum temperatures, it over-

estimates the intensity of hot extremes associated with min-

imum and maximum temperatures. The effect of dynamic

vegetation (NOAH-MP vs. NOAH-MP-DV) on the biases in

extreme and mean maximum and minimum temperatures re-

mains constant using different resolutions, reaching larger bi-

ases and colder maximum and minimum temperatures with

dynamic vegetation than with prescribed vegetation for most

of the regions (Fig. 9a and b).

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 423

Figure 10. Regional mean annual and seasonal biases in extreme maximum and minimum temperatures (◦C) and in extreme precipitation

(mmd−1) for all experiments and the DAYMET data product relative to the CRU database from 1980 to 2013. Extremes are calculated as

the 95th percentile of the annual and seasonal temporal series at each grid cell. Biases are estimated averaging over six subregions (Fig. 1)

adapted from Giorgi and Francisco (2000): Central America, CAM; western North America, WNA; central North America, CNA; eastern

North America, ENA; Alaska, ALA; and Greenland, GRL.

The WRF model simulates large positive biases in daily

accumulated precipitation at the surface over most of North

America during all seasons, with larger biases in summer

(Figs. 9c and S17 in the Supplement). A negative bias is

also present in all experiments over the southeastern USA

and the eastern coast of North America in summer and win-

ter (Fig. S17). Dry biases are reduced when using ﬁner hor-

izontal resolutions, while wet biases are larger when using

smaller scales (Fig. 9c). This is due to the intensiﬁcation of

the water cycle with the use of ﬁner horizontal resolutions

discussed in Sect. 4.2 and presented in Figs. 6d and 8. For ex-

ample in winter, the dry bias shown in all experiments over

the southeastern NA is associated with an increase in non-

convective precipitation using ﬁner resolutions (Figs. S17

and 8b). In summer, the bias in precipitation is larger us-

ing ﬁner resolutions over most of coastal areas where an in-

crease in convective precipitation and latent heat ﬂux was

shown with the use of ﬁner resolutions (Figs. 6d, 8a and

S17). The impact of resolution on the accumulated precip-

itation is stronger than the effect of the LSM component,

which affects precipitation mainly in summer (Fig. 9). The

results show a larger bias in extreme precipitation than for

the mean accumulated precipitation but yielding similar con-

clusions (Fig. 10).

In summary, the LSM impact on temperatures is larger

than the resolution effect, while the opposite is true for pre-

cipitation climatologies; i.e., differences in precipitation aris-

ing from changes in resolution are larger than LSM differ-

ences (Fig. 9). The inﬂuence of both the LSM choice and

resolution intensiﬁes in summer compared with the rest of

seasons probably because of the larger energy exchanges and

the consequent intensiﬁcation of land–atmosphere coupling

in summer (Zhang et al., 2008; Mei and Wang, 2012). The

CLM4 LSM generates the smallest biases relative to the CRU

database in the WRF simulation of mean maximum and min-

imum temperatures; however it also yields the larger biases

in extreme maximum and minimum temperatures. The use

of ﬁner resolutions leads to slightly larger or smaller biases

in the simulation of maximum and minimum temperatures

depending on the LSM component and the region, while the

use of ﬁner resolutions implies larger biases in mean and ex-

treme precipitation at low and middle latitudes for all LSM

components, particularly in summer.

5 Discussion

The dependence of climate simulations on the LSM com-

ponent shown in this study agrees with conclusions drawn

from previous analyses at different temporal and spatial

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

424 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

scales (Chen et al., 2014; Van Den Broeke et al., 2018; Liu

et al., 2019; Zhuo et al., 2019; Davin and Seneviratne, 2012;

Mooney et al., 2013; Laguë et al., 2019; García-García et al.,

2020). For example, using the Consortium for Small-scale

Modeling (COSMO) and WRF RCMs, Davin and Senevi-

ratne (2012) and Mooney et al. (2013) identiﬁed an LSM

sensitivity of temperature and precipitation conditions over

Europe which intensiﬁes in summer. Additionally, our anal-

ysis has shown that the impact of the LSM choice on the

WRF simulation of precipitation is weaker than its impact

on temperature, in agreement with studies over a small re-

gion in Italy (Zhuo et al., 2019), over Europe (Mooney et al.,

2013), and over the western and central USA at seasonal

scales (Jin et al., 2010; Chen et al., 2014; Van Den Broeke

et al., 2018). The implementation and calculation of surface

properties (albedo, surface roughness and evaporative resis-

tances) by each LSM can be the cause of the LSM differences

in the simulation of energy ﬂuxes (Laguë et al., 2019). Thus,

LSM differences in SNET are probably related to albedo val-

ues within each LSM. The large LSM differences in sensible

heat ﬂux over the boreal forest are likely related to the LSM

estimates of surface roughness, while the large latent heat

ﬂux differences at low latitudes are probably associated with

different estimates of evaporative resistances and the treat-

ment of soil water in each LSM.

The great WRF sensibility to precipitation rates to resolu-

tion is also supported by the literature (e.g., Pieri et al., 2015).

Our results also show large seasonal differences, mainly

caused by the different contribution of convective precipi-

tation in summer and winter. In summer at middle and low

latitudes, the use of ﬁner grid cells leads to a change in the en-

ergy partition into sensible and latent heat ﬂux, increasing la-

tent heat, and decreasing sensible heat (Fig. 6). This increase

in latent heat ﬂux over these areas is probably the cause of

the higher values of convective precipitation (Fig. 8a). This

resolution-induced increase in precipitation through changes

in convective processes has also been suggested in the litera-

ture (Prein et al., 2016). At high latitudes both turbulent heat

ﬂuxes increase in summer with the use of ﬁner resolutions

mainly due to the increase in shortwave net radiation proba-

bly related to the decrease in cloud cover shown in Fig. 8d.

These changes in cloud cover with resolution may be caused

by the performance of the microphysical parameterizations

at different resolutions and the improvement in the represen-

tation of orography (Pieri et al., 2015; Prein et al., 2016).

Previous evaluations of RCMs using several soil models

over different domains reached the conclusion that the most

complex LSM components – that is, the LSM components

representing more physical phenomena – outperform others

(Chen et al., 2014; Van Den Broeke et al., 2018; Liu et al.,

2019). Over North America, our results indicate that the

WRF simulation of temperature conditions using the CLM4

LSM outperforms the simulation of mean maximum and

minimum temperatures generated by the NOAH and NOAH-

MP LSMs, but it yields larger biases in extreme maximum

and minimum temperatures (Figs. 9 and 10). The simulation

of precipitation in summer is, however, slightly better repre-

sented by the NOAH LSM than by the other LSM compo-

nents (Fig. 9). Nonetheless, the comparison of all WRF ex-

periments with observations shows overestimated values of

precipitation over most of North America, which is in agree-

ment with other studies using WRF over the western USA

(Jin et al., 2010; Chen et al., 2014) and over Europe (Pieri

et al., 2015). Atmospheric parameterizations were not tested

in our study; however, other WRF sensitivity experiments us-

ing several microphysics schemes over Europe found a posi-

tive bias in precipitation for all simulations, which was con-

siderably reduced in summer within a convective-permitting

simulation (Pieri et al., 2015); that is, the positive bias in pre-

cipitation has been reported in WRF simulations over dif-

ferent domains using several LSM components, horizontal

resolutions, microphysics parameterizations and reanalysis

products as initial and boundary conditions (Figs. 9 in this

paper, Pieri et al., 2015; Chen et al., 2014; Jin et al., 2010).

Therefore, the results included here, together with the results

reported in the literature, suggest that the use of ﬁner res-

olutions may raise precipitation biases in WRF simulations

over North America, but the implementation of convective-

permitting processes and other atmospheric parameteriza-

tions could reduce this bias.

6 Conclusions

This study has shown the effect of changes in horizontal res-

olution and LSM choice on the simulation of energy ﬂuxes

at the surface and temperature and water conditions near the

surface. The effect of both model choices intensiﬁes in sum-

mer due to the increase in energy and water exchanges be-

tween the lower atmosphere and the land surface. Enhancing

horizontal resolution leads to higher precipitation climatolo-

gies for all LSMs over coastal areas at low latitudes mainly

due to an increase in convective precipitation also associated

with an increase in latent heat ﬂux. Our results highlight the

important role of the LSM choice in the WRF representa-

tion of the energy partition at the surface, which mainly af-

fects the simulation of near-surface temperatures over North

America. Additionally, these results demonstrate the impact

of the LSM choice on simulated atmospheric conditions,

showing LSM-dependent differences in the simulated cloud

fraction and non-convective precipitation rates. This is prob-

ably associated with the land–atmospheric coupled character

of our simulations and the interactions between small- and

large-scale dynamical processes.

The evaluation of the WRF simulations against observa-

tions supports the use of the CLM4 LSM as the best choice

within the evaluated options for WRF simulations over North

America, although it may overestimate temperature extreme

events. The use of ﬁner resolutions yields a small improve-

ment in the representation of minimum temperatures within

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 425

WRF at middle and high latitudes. Nonetheless, the use of

ﬁner resolutions should be implemented with caution since

it may increase the WRF bias in mean and extreme pre-

cipitation. Further sensitivity experiments using other atmo-

spheric parameterizations and other resolutions ﬁne enough

for convective-permitting processes (<5 km) or for altering

the hydrostatic balance of the model (≈10 km) are necessary

to determine the best WRF conﬁguration for downscaling cli-

mate simulations over North America for paleoclimate and

climate change studies. However, the large computational re-

sources required to perform a sensitivity analysis including

convective-permitting processes would require a reduction in

the area of interest or in the length of the simulation.

Information provided by downscaling studies is used for

building climate change policies through the information col-

lected in assessment reports (IPCC, 2013; Mbow et al., 2017;

USGCRP, 2018). Thus, sensitivity studies like the one pre-

sented here are crucial to understand and ultimately restrict

uncertainties in climate simulations with direct beneﬁts to

society and environment. Particularly, these results should

be considered for downscaling studies over North America

aimed at projecting future or past conditions and informing

policymakers.

Code and data availability. As described in Sect. 2, here we

used the Research and Forecasting model (WRF, version 3.9,

http://www.wrf-model.org, last access: 12 January 2022, Ska-

marock et al., 2008). The outputs of all simulations to-

gether with the code used to estimate the presented re-

sults are available at https://doi.org/10.5281/zenodo.5106087

(García-García et al., 2021). The NARR product (Mesinger

et al., 2006) was obtained from https://www.ncei.noaa.gov/data/

north-american-regional-reanalysis/access/3-hourly/ (last access:

12 January 2022). The CRU TS4.03 database can be downloaded

from the University of East Anglia web page (https://doi.org/

10.5285/10d3e3640f004c578403419aac167d82, University of East

Anglia Climatic Research Unit et al., 2020). The DAYMET V3

database (Thornton et al., 2016) is available at https://doi.org/10.

3334/ORNLDAAC/1328.

Supplement. The supplement related to this article is available on-

line at: https://doi.org/10.5194/gmd-15-413-2022-supplement.

Author contributions. AGG designed the modelling experiment,

performed the simulations and analyzed model outputs. All authors

(AGG, FJCV, HB, JFGR and EGB) contributed to the interpretation

and discussion of results. AGG wrote the manuscript with continu-

ous feedback from FJCV, HB, JFGR and EGB.

Competing interests. The contact author has declared that neither

they nor their co-authors have any competing interests.

Disclaimer. Publisher’s note: Copernicus Publications remains

neutral with regard to jurisdictional claims in published maps and

institutional afﬁliations.

Acknowledgements. We thank the Mesoscale and Microscale Me-

teorology (MMM), the National Center for Atmospheric Re-

search (NCAR), the National Oceanic and Atmospheric Admin-

istration (NOAA), the Climatic Research Unit (CRU) at the

University of East Anglia and the Oak Ridge National Lab-

oratory Distributed Active Archive Center (ORNL DAAC) for

making the WRF code and the NARR, CRU and DAYMET

datasets available. This analysis contributes to the PALEOLINK

project (http://www.pastglobalchanges.org/science/wg/2k-network/

projects/paleolink/intro, last access: 4 May 2021), part of the

PAGES 2k network. All WRF simulations were performed in the

computational facilities provided by the Atlantic Computational Ex-

cellence Network (ACENET-Compute Canada). During the elabo-

ration of this analysis, Almudena García-García and Francisco José

Cuesta-Valero were partially funded by Hugo Beltrami’s Canada

Research Chair program, the School of Graduate Students at Memo-

rial University of Newfoundland and the Research Ofﬁce at St.

Francis Xavier University. Francisco José Cuesta-Valero is also

funded by the Alexander von Humboldt Foundation.

Financial support. This research has been supported by the Cana-

dian Network for Research and Innovation in Machining Technol-

ogy, Natural Sciences and Engineering Research Council of Canada

(grant no. DG 140576948), the Canada Excellence Research Chairs,

Government of Canada (grant no. 230687), and the Canada Foun-

dation for Innovation (CFI).

Review statement. This paper was edited by David Lawrence and

reviewed by two anonymous referees.

References

Ban, N., Schmidli, J., and Schär, C.: Evaluation of the convection-

resolving regional climate modeling approach in decade-

long simulations, J. Geophys. Res.-Atmos., 119, 7889–7907,

https://doi.org/10.1002/2014JD021478, 2014.

Barlage, M., Zeng, X., Wei, H., and Mitchell, K. E.: A global

0.05◦maximum albedo dataset of snow-covered land based

on MODIS observations, Geophys. Res. Lett., 32, L17405,

https://doi.org/10.1029/2005GL022881, 2005.

Bonan, G. B.: Ecological climatology: concepts and applications,

Cambridge University Press, Cambridge, UK, ISBN 521 80476

0, 2002.

Chen, F., Liu, C., Dudhia, J., and Chen, M.: A sensitivity study of

high-resolution regional climate simulations to three land surface

models over the western United States, J. Geophys. Res.-Atmos.,

119, 7271–7291, https://doi.org/10.1002/2014JD021827, 2014.

Collins, W. D., Rasch, P. J., Boville, B. A., Hack, J. J., Mc-

Caa, J. R., Williamson, D. L., Kiehl, J. T., Briegleb, B., Bitz, C.,

and Lin, S.-J.: Description of the NCAR community atmo-

sphere model (CAM 3.0), NCAR Tech. Note NCAR/TN-464+

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

426 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

STR, 226, University Corporation for Atmospheric Research,

https://doi.org/10.5065/D63N21CH, 2004.

Cubasch, U., Wuebbles, D., Chen, D., Facchini, M.,

Frame, D., Mahowald, N., and Winther, J.-G.: Intro-

duction, book section 1, Cambridge University Press,

Cambridge, UK and New York, NY, USA, 119–158,

https://doi.org/10.1017/CBO9781107415324.007, 2013.

Davin, E. L. and Seneviratne, S. I.: Role of land surface

processes and diffuse/direct radiation partitioning in simu-

lating the European climate, Biogeosciences, 9, 1695–1707,

https://doi.org/10.5194/bg-9-1695-2012, 2012.

Demory, M.-E., Vidale, P. L., Roberts, M. J., Berrisford, P.,

Strachan, J., Schiemann, R., and Mizielinski, M. S.: The

role of horizontal resolution in simulating drivers of the

global hydrological cycle, Clim. Dynam., 42, 2201–2225,

https://doi.org/10.1007/s00382-013-1924-4, 2014.

Di Luca, A., de Elía, R., and Laprise, R.: Challenges in

the Quest for Added Value of Regional Climate Dynami-

cal Downscaling, Current Climate Change Reports, 1, 10–21,

https://doi.org/10.1007/s40641-015-0003-9, 2015.

Dirmeyer, P. A., Jin, Y., Singh, B., and Yan, X.: Trends in Land–

Atmosphere Interactions from CMIP5 Simulations, J. Hydrome-

teorol., 14, 829–849, https://doi.org/10.1175/JHM-D-12-0107.1,

2013.

García-García, A., Cuesta-Valero, F. J., Beltrami, H., and Smer-

don, J. E.: Characterization of Air and Ground Temperature

Relationships within the CMIP5 Historical and Future Cli-

mate Simulations, J. Geophys. Res.-Atmos., 124, 3903–3929,

https://doi.org/10.1029/2018JD030117, 2019.

García-García, A., Cuesta-Valero, F. J., Beltrami, H., González-

Rouco, F., García-Bustamante, E., and Finnis, J.: Land sur-

face model inﬂuence on the simulated climatologies of tem-

perature and precipitation extremes in the WRF v3.9 model

over North America, Geosci. Model Dev., 13, 5345–5366,

https://doi.org/10.5194/gmd-13-5345-2020, 2020.

García-García, A., José Cuesta-Valero, F., Beltrami, H., González-

Rouco, F., and García-Bustamante, E.: Code and Data: WRF

v.3.9 sensitivity to land surface model and horizontal res-

olution changes over North America, Zenodo [data set],

https://doi.org/10.5281/zenodo.5106087, 2021.

Giorgi, F. and Francisco, R.: Uncertainties in regional climate

change prediction: a regional analysis of ensemble simulations

with the HADCM2 coupled AOGCM, Clim. Dynam., 16, 169–

182, https://doi.org/10.1007/PL00013733, 2000.

Giorgi, F. and Gutowski Jr., W. J.: Regional Dynamical Down-

scaling and the CORDEX Initiative, Annu. Rev. Env. Resour.,

40, 467–490, https://doi.org/10.1146/annurev-environ-102014-

021217, 2015.

Gómez-Navarro, J. J., Raible, C. C., and Dierer, S.: Sensitivity of

the WRF model to PBL parametrisations and nesting techniques:

evaluation of wind storms over complex terrain, Geosci. Model

Dev., 8, 3349–3363, https://doi.org/10.5194/gmd-8-3349-2015,

2015.

Gómez-Navarro, J. J., Ludwig, P., and Zorita, E.: The PALEOLINK

working project: The missing link in the Past-Downscaling pa-

leoclimatic Earth System Models, in: EGU General Assembly

Conference Abstracts, 14961, 2018.

Grell, G. A. and Freitas, S. R.: A scale and aerosol aware

stochastic convective parameterization for weather and air

quality modeling, Atmos. Chem. Phys., 14, 5233–5250,

https://doi.org/10.5194/acp-14-5233-2014, 2014.

Hahmann, A. N., S¯ıle, T., Witha, B., Davis, N. N., Dörenkäm-

per, M., Ezber, Y., García-Bustamante, E., González-Rouco, J. F.,

Navarro, J., Olsen, B. T., and Söderberg, S.: The making of the

New European Wind Atlas – Part 1: Model sensitivity, Geosci.

Model Dev., 13, 5053–5078, https://doi.org/10.5194/gmd-13-

5053-2020, 2020.

Harris, I., Jones, P., Osborn, T., and Lister, D.: Updated

high-resolution grids of monthly climatic observations –

the CRU TS3.10 Dataset, Int. J. Climatol., 34, 623–642,

https://doi.org/10.1002/joc.3711, 2014.

Hatzianastassiou, N., Matsoukas, C., Fotiadi, A., Pavlakis, K. G.,

Drakakis, E., Hatzidimitriou, D., and Vardavas, I.: Global dis-

tribution of Earth’s surface shortwave radiation budget, Atmos.

Chem. Phys., 5, 2847–2867, https://doi.org/10.5194/acp-5-2847-

2005, 2005.

Hauser, M., Orth, R., and Seneviratne, S. I.: Role of soil

moisture versus recent climate change for the 2010 heat

wave in western Russia, Geophys. Res. Lett., 43, 2819–2826,

https://doi.org/10.1002/2016GL068036, 2016.

Hirschi, M., Seneviratne, S. I., Alexandrov, V., Boberg, F.,

Boroneant, C., Christensen, O. B., Formayer, H., Orlowsky, B.,

and Stepanek, P.: Observational evidence for soil-moisture im-

pact on hot extremes in southeastern Europe, Nat. Geosci., 4,

17–21, https://doi.org/10.1038/ngeo1032, 2011.

Hong, S.-Y. and Lim, J.-O. J.: The WRF single-moment 6-class mi-

crophysics scheme (WSM6), J. Korean Meteor. Soc, 42, 129–

151, 2006.

Hong, S.-Y., Noh, Y., and Dudhia, J.: A New Vertical Dif-

fusion Package with an Explicit Treatment of Entrain-

ment Processes, Mon. Weather Rev., 134, 2318–2341,

https://doi.org/10.1175/MWR3199.1, 2006.

Iles, C. E., Vautard, R., Strachan, J., Joussaume, S., Eggen, B. R.,

and Hewitt, C. D.: The beneﬁts of increasing resolu-

tion in global and regional climate simulations for Euro-

pean climate extremes, Geosci. Model Dev., 13, 5583–5607,

https://doi.org/10.5194/gmd-13-5583-2020, 2020.

IPCC: Climate Change 2013: The Physical Science Basis. Con-

tribution of Working Group I to the Fifth Assessment Report

of the Intergovernmental Panel on Climate Change, Cambridge

University Press, Cambridge, UK and New York, NY, USA,

https://doi.org/10.1017/CBO9781107415324, 2013.

IPCC: Summary for Policymakers, in: Climate Change and Land:

an IPCC special report on climate change, desertiﬁcation, land

degradation, sustainable land management, food security, and

greenhouse gas ﬂuxes in terrestrial ecosystems, edited by:

Shukla, P. R., Skea, J., Calvo Buendia, E., Masson-Delmotte,

V., Pörtner, H.-O., Roberts, D. C., Zhai, P., Slade, R., Connors,

S., van Diemen, R., Ferrat, M., Haughey, E., Luz, S., Neogi, S.,

Pathak, M., Petzold, J., Portugal Pereira, J., Vyas, P., Huntley, E.,

Kissick, K., Belkacemi, M., Malley, J., in press, 2019.

Janjic, Z. I.: Advection scheme for passive substances inthe NCEP

Eta model, Research Activities in Atmospheric and Oceanic

Modelling, WMO, Genova, CAS=JSCWGNE, 3–14, 1997.

Jiménez, P. A., Dudhia, J., González-Rouco, J. F., Navarro, J., Mon-

távez, J. P., and García-Bustamante, E.: A Revised Scheme for

the WRF Surface Layer Formulation, Mon. Weather Rev., 140,

898–918, https://doi.org/10.1175/MWR-D-11-00056.1, 2012.

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes 427

Jin, J., Miller, N. L., and Schlegel, N.: Sensitivity Study of Four

Land Surface Schemes in the WRF Model, Adv. Meteorol., 2010,

167436, https://doi.org/10.1155/2010/167436, 2010.

Koster, R. D., Sud, Y. C., Guo, Z., Dirmeyer, P. A., Bonan, G.,

Oleson, K. W., Chan, E., Verseghy, D., Cox, P., Davies, H.,

Kowalczyk, E., Gordon, C. T., Kanae, S., Lawrence, D., Liu, P.,

Mocko, D., Lu, C.-H., Mitchell, K., Malyshev, S., McAvaney, B.,

Oki, T., Yamada, T., Pitman, A., Taylor, C. M., Vasic, R.,

and Xue, Y.: GLACE: The Global Land–Atmosphere Coupling

Experiment. Part I: Overview, J. Hydrometeorol., 7, 590–610,

https://doi.org/10.1175/JHM510.1, 2006.

Koven, C. D., Riley, W. J., and Stern, A.: Analysis of Per-

mafrost Thermal Dynamics and Response to Climate Change in

the CMIP5 Earth System Models, J. Climate, 26, 1877–1900,

https://doi.org/10.1175/JCLI-D-12-00228.1, 2013.

Laguë, M. M., Bonan, G. B., and Swann, A. L. S.: Separat-

ing the impact of individual land surface properties on the

terrestrial surface energy budget in both the coupled and un-

coupled land-atmosphere system, J. Climate, 32, 5725–5744,

https://doi.org/10.1175/JCLI-D-18-0812.1, 2019.

Lawrence, D. M., Fisher, R. A., Koven, C. D., Oleson, K. W.,

Swenson, S. C., Bonan, G., Collier, N., Ghimire, B., van Kam-

penhout, L., Kennedy, D., Kluzek, E., Lawrence, P. J., Li, F.,

Li, H., Lombardozzi, D., Riley, W. J., Sacks, W. J., Shi, M.,

Vertenstein, M., Wieder, W. R., Xu, C., Ali, A. A., Bad-

ger, A. M., Bisht, G., van den Broeke, M., Brunke, M. A.,

Burns, S. P., Buzan, J., Clark, M., Craig, A., Dahlin, K., Drew-

niak, B., Fisher, J. B., Flanner, M., Fox, A. M., Gentine, P., Hoff-

man, F., Keppel-Aleks, G., Knox, R., Kumar, S., Lenaerts, J., Le-

ung, L. R., Lipscomb, W. H., Lu, Y., Pandey, A., Pelletier, J. D.,

Perket, J., Randerson, J. T., Ricciuto, D. M., Sanderson, B. M.,

Slater, A., Subin, Z. M., Tang, J., Thomas, R. Q., Val Mar-

tin, M., and Zeng, X.: The Community Land Model Version

5: Description of New Features, Benchmarking, and Impact of

Forcing Uncertainty, J. Adv. Model. Earth Sy., 11, 4245–4287,

https://doi.org/10.1029/2018MS001583, 2019.

Liu, L., Ma, Y., Menenti, M., Zhang, X., and Ma, W.: Evaluation of

WRF Modeling in Relation to Different Land Surface Schemes

and Initial and Boundary Conditions: A Snow Event Simulation

Over the Tibetan Plateau, J. Geophys. Res.-Atmos., 124, 209–

226, https://doi.org/10.1029/2018JD029208, 2019.

Lorenz, R., Argüeso, D., Donat, M. G., Pitman, A. J.,

van den Hurk, B., Berg, A., Lawrence, D. M., Chéruy, F.,

Ducharne, A., Hagemann, S., Meier, A., Milly, P. C. D.,

and Seneviratne, S. I.: Inﬂuence of land-atmosphere feedbacks

on temperature and precipitation extremes in the GLACE-

CMIP5 ensemble, J. Geophys. Res.-Atmos., 121, 607–623,

https://doi.org/10.1002/2015JD024053, 2016.

Lucas-Picher, P., Laprise, R., and Winger, K.: Evidence of added

value in North American regional climate model hindcast sim-

ulations using ever-increasing horizontal resolutions, Clim. Dy-

nam., 48, 2611–2633, https://doi.org/10.1007/s00382-016-3227-

z, 2017.

MacDougall, A. H. and Beltrami, H.: Impact of deforestation

on subsurface temperature proﬁles: implications for the bore-

hole paleoclimate record, Environ. Res. Lett., 12, 074014,

https://doi.org/10.1088/1748-9326/aa7394, 2017.

Mbow, H.-O. P., Reisinger, A., Canadell, J., and O’Brien, P.: Spe-

cial Report on climate change, desertiﬁcation, land degradation,

sustainable land management, food security, and greenhouse gas

ﬂuxes in terrestrial ecosystems (SR2), IPCC, Ginevra, 2017.

Mei, R. and Wang, G.: Summer Land–Atmosphere Coupling

Strength in the United States: Comparison among Observations,

Reanalysis Data, and Numerical Models, J. Hydrometeorol., 13,

1010–1022, https://doi.org/10.1175/JHM-D-11-075.1, 2012.

Melo-Aguilar, C., González-Rouco, J. F., García-Bustamante, E.,

Navarro-Montesinos, J., and Steinert, N.: Inﬂuence of radiative

forcing factors on ground–air temperature coupling during the

last millennium: implications for borehole climatology, Clim.

Past, 14, 1583–1606, https://doi.org/10.5194/cp-14-1583-2018,

2018.

Mesinger, F., DiMego, G., Kalnay, E., Mitchell, K., Shafran, P. C.,

Ebisuzaki, W., Jovic, D., Woollen, J., Rogers, E., Berbery, E. H.,

Ek, M. B., Fan, Y., Grumbine, R., Higgins, W., Li, H.,

Lin, Y., Manikin, G., Parrish, D., and Shi, W.: North

American regional reanalysis, B. Am. Meteorol. Soc.,

87, 343–360, https://doi.org/10.1175/BAMS-87-3-343,

2006 (data available at: https://www.ncei.noaa.gov/data/

north-american-regional-reanalysis/access/3-hourly/, last

access: 12 January 2022).

Messmer, M., Gómez-Navarro, J. J., and Raible, C. C.: Sensitiv-

ity experiments on the response of Vb cyclones to sea surface

temperature and soil moisture changes, Earth Syst. Dynam., 8,

477–493, https://doi.org/10.5194/esd-8-477-2017, 2017.

Miralles, D. G., den Berg, M. J., Teuling, A. J., and

Jeu, R. A. M.: Soil moisture-temperature coupling: A multi-

scale observational analysis, Geophys. Res. Lett., 39, L21707,

https://doi.org/10.1029/2012GL053703, 2012.

Mitchell, K.: The community Noah land-surface model (LSM),

User’s Guide, 2.7.1, available at: https://ral.ucar.edu/sites/

default/ﬁles/public/product-tool/uniﬁed-noah-lsm/Noah_LSM_

USERGUIDE_2.7.1.pdf (last access: 12 January 2022), 2005.

Mooney, P. A., Mulligan, F. J., and Fealy, R.: Evaluation of the

Sensitivity of the Weather Research and Forecasting Model

to Parameterization Schemes for Regional Climates of Eu-

rope over the Period 1990–95, J. Climate, 26, 1002–1017,

https://doi.org/10.1175/JCLI-D-11-00676.1, 2013.

Niu, G.-Y., Yang, Z.-L., Mitchell, K. E., Chen, F., Ek, M. B.,

Barlage, M., Kumar, A., Manning, K., Niyogi, D., Rosero, E.,

Tewari, M., and Xia, Y.: The community Noah land

surface model with multiparameterization options (Noah-

MP): 1. Model description and evaluation with local-scale

measurements, J. Geophys. Res.-Atmos., 116, 2156–2202,

https://doi.org/10.1029/2010JD015139, 2011.

Oleson, K. W., Lawrence, D. M., Bonan, G. B., Flanner, M. G.,

Kluzek, E., Lawrence, P. J., Levis, S., Swenson, S. C., Thorn-

ton, P. E., Dai, A., Decker, M., Dickinson, R., Feddema, J.,

Heald, C. L., Hoffman, F., Lamarque, J.-F., Mahowald, N.,

Niu, G.-Y., Qian, T., Randerson, J., Running, S., Sakaguchi, K.,

Slater, A., Stockli, R., Wang, A., Yang, Z.-L., Zeng, X., and

Zeng., X.: Technical description of version 4.0 of the Commu-

nity Land Model (CLM), Tech. rep., NCAR, Boulder, 2010.

Pei, L., Moore, N., Zhong, S., Luo, L., Hyndman, D. W., Heil-

man, W. E., and Gao, Z.: WRF Model Sensitivity to Land Surface

Model and Cumulus Parameterization under Short-Term Climate

Extremes over the Southern Great Plains of the United States,

J. Climate, 27, 7703–7724, https://doi.org/10.1175/JCLI-D-14-

00015.1, 2014.

https://doi.org/10.5194/gmd-15-413-2022 Geosci. Model Dev., 15, 413–428, 2022

428 A. García-García et al.: WRF sensitivity to horizontal resolution and LSM changes

Pieri, A. B., von Hardenberg, J., Parodi, A., and Provenzale, A.:

Sensitivity of Precipitation Statistics to Resolution, Micro-

physics, and Convective Parameterization: A Case Study with

the High-Resolution WRF Climate Model over Europe, J. Hy-

drometeorol., 16, 1857–1872, https://doi.org/10.1175/JHM-D-

14-0221.1, 2015.

Prein, A., Gobiet, A., Truhetz, H., Keuler, K., Goergen, K., Te-

ichmann, C., Fox Maule, C., van Meijgaard, E., Déqué, M.,

Nikulin, G., Vautard, R., Colette, A., Kjellström, E., and Jacob,

D.: Precipitation in the EURO-CORDEX 0.11◦and 0.44◦simu-

lations: high resolution, high beneﬁts?, Clim. Dynam., 46, 383–

412, https://doi.org/10.1007/s00382-015-2589-y, 2016.

Prein, A. F., Gobiet, A., Suklitsch, M., Truhetz, H., Awan, N. K.,

Keuler, K., and Georgievski, G.: Added value of convection

permitting seasonal simulations, Clim. Dynam., 41, 2655–2677,

https://doi.org/10.1007/s00382-013-1744-6, 2013.

Rummukainen, M.: Added value in regional cli-

mate modeling, WIREs Clim. Change, 7, 145–159,

https://doi.org/10.1002/wcc.378, 2016.

Seneviratne, S. I., Lüthi, D., Litschi, M., and Schär, C.: Land–

atmosphere coupling and climate change in Europe, Nature, 443,

205–209, 2006.

Sippel, S., Zscheischler, J., Mahecha, M. D., Orth, R., Reich-

stein, M., Vogel, M., and Seneviratne, S. I.: Reﬁning multi-model

projections of temperature extremes by evaluation against land–

atmosphere coupling diagnostics, Earth Syst. Dynam., 8, 387–

403, https://doi.org/10.5194/esd-8-387-2017, 2017.

Skamarock, W. C., Klemp, J. B., Dudhia, J., Gill, D. O., Barker,

D., Duda, M. G., Huang, X.-y., Wang, W., and Powers, J. G.:

A Description of the Advanced Research WRF Version 3 (No.

NCAR/TN-475+STR), University Corporation for Atmospheric

Research, https://doi.org/10.5065/D68S4MVH, 2008.

Sun, X., Xue, M., Brotzge, J., McPherson, R. A., Hu, X.-

M., and Yang, X.-Q.: An evaluation of dynamical downscal-

ing of Central Plains summer precipitation using a WRF-

based regional climate model at a convection-permitting

4 km resolution, J. Geophys. Res.-Atmos., 121, 13801–13825,

https://doi.org/10.1002/2016JD024796, 2016.

Taylor, K. E., Stouffer, R. J., and Meehl, G. A.: An overview of

CMIP5 and the experiment design, B. Am. Meteorol. Soc., 93,

485–498, 2012.

Tewari, M., Chen, F., Wang, W., Dudhia, J., LeMone, M.,

Mitchell, K., Ek, M., Gayno, G., Wegiel, J., and Cuenca, R.: Im-

plementation and veriﬁcation of the uniﬁed NOAH land surface

model in the WRF model, 20th conference on weather analysis

and forecasting/16th conference on numerical weather predic-

tion, American Meteorological Society, Seattle, WA, US, 11–15,

2004.

Thornton, P., Thornton, M., Mayer, B., Wei, Y., De-

varakonda, R., Vose, R., and Cook, R.: Daymet: Daily

Surface Weather Data on a 1-km Grid for North America,

Version 3, ORNL DAAC, Oak Ridge, Tennessee, USA,

https://doi.org/10.3334/ORNLDAAC/1328, 2016.

University of East Anglia Climatic Research Unit, Harris, I.

C., and Jones, P. D.: CRU TS4.03: Climatic Research Unit

(CRU) Time-Series (TS) version 4.03 of high-resolution gridded

data of month-by-month variation in climate (Jan. 1901–Dec.

2018), Centre for Environmental Data Analysis [data set],

https://doi.org/10.5285/10d3e3640f004c578403419aac167d82,

2020.

U.S. Global Change Research Program (USGCRP): Impacts, Risks,

and Adaptation in the United States: Fourth National Cli-

mate Assessment, Volume II, edited by: Reidmiller, D. R.,

Avery, C. W., Easterling, D. R., Kunkel, K. E., Lewis, K.

L. M., Maycock, T. K., and Stewart, B. C., U.S. Global

Change Research Program, Washington, DC, USA, 1515 pp.,

https://doi.org/10.7930/NCA4.2018, 2018.

Van Den Broeke, M. S., Kalin, A., Alavez, J. A. T., Oglesby, R.,

and Hu, Q.: A warm-season comparison of WRF coupled to the

CLM4.0, Noah-MP, and Bucket hydrology land surface schemes

over the central USA, Theor. Appl. Climatol., 134, 801–816,

https://doi.org/10.1007/s00704-017-2301-8, 2018.

Vegas-Cañas, C., González-Rouco, J. F., Navarro-Montesinos, J.,

García-Bustamante, E., Lucio-Eceiza, E. E., García-Pereira, F.,

Rodríguez-Camino, E., Chazarra-Bernabé, A., and Álvarez-

Arévalo, I.: An Assessment of Observed and Simulated Temper-

ature Variability in Sierra de Guadarrama, Atmosphere, 11, 985,

https://doi.org/10.3390/atmos11090985, 2020.

Vogel, M. M., Orth, R., Cheruy, F., Hagemann, S., Lorenz, R.,

Hurk, B. J. J. M., and Seneviratne, S. I.: Regional ampliﬁcation

of projected changes in extreme temperatures strongly controlled

by soil moisture-temperature feedbacks, Geophys. Res. Lett., 44,

1511–1519, https://doi.org/10.1002/2016GL071235, 2017.

von Storch, H., Langenberg, H., and Feser, F.: A Spec-

tral Nudging Technique for Dynamical Downscaling, Mon.

Weather Rev., 128, 3664–3673, https://doi.org/10.1175/1520-

0493(2000)128<3664:ASNTFD>2.0.CO;2, 2000.

Wehner, M. F., Smith, R. L., Bala, G., and Duffy, P.: The effect of

horizontal resolution on simulation of very extreme US precip-

itation events in a global atmosphere model, Clim. Dynam., 34,

241–247, https://doi.org/10.1007/s00382-009-0656-y, 2010.

Wharton, S., Simpson, M., Osuna, J., Newman, J., and Biraud, S.:

Assessment of Land Surface Model Performance in WRF for

Simulating Wind at Heights Relevant to the Wind Energy Com-

munity, Tech. rep., Lawrence Livermore National Lab. (LLNL),

Livermore, CA, US, 2013.

Xue, Y., Janjic, Z., Dudhia, J., Vasic, R., and Sales, F. D.:

A review on regional dynamical downscaling in intrasea-

sonal to seasonal simulation/prediction and major factors that

affect downscaling ability, Atmos. Res., 147–148, 68–85,

https://doi.org/10.1016/j.atmosres.2014.05.001, 2014.

Zhang, J., Wang, W.-C., and Leung, L. R.: Contribution of land-

atmosphere coupling to summer climate variability over the con-

tiguous United States, J. Geophys. Res.-Atmos., 113, D2210,

https://doi.org/10.1029/2008JD010136, 2008.

Zhuo, L., Dai, Q., Han, D., Chen, N., and Zhao, B.: Assessment of

simulated soil moisture from WRF Noah, Noah-MP, and CLM

land surface schemes for landslide hazard application, Hydrol.

Earth Syst. Sci., 23, 4199–4218, https://doi.org/10.5194/hess-23-

4199-2019, 2019.

Geosci. Model Dev., 15, 413–428, 2022 https://doi.org/10.5194/gmd-15-413-2022

Performance of Seven Land Surface Schemes in the WRFv4.3 Model for Simulating Precipitation in the Record‐Breaking Meiyu Season Over the Yangtze–Huaihe River Valley in China

Article

Full-text available

Mar 2023

In 2020, the Yangtze-Huai river valley (YHRV) experienced the highest record-breaking Meiyu season since 1961, which was mainly characterized by the longest duration of precipitation lasting from early-June to mid-July, with frequent heavy rainstorms that caused severe flooding and deaths in China. Many studies have investigated the causes of this Meiyu season and its evolution, but the accuracy of precipitation simulations has received little attention. It is important to provide more accurate precipitation forecasts to help prevent and reduce flood disasters, thereby facilitating the maintenance of a healthy and sustainable earth ecosystem. In this study, we determined the optimal scheme among seven land surface model (LSMs) schemes in the Weather Research and Forecasting model for simulating the precipitation in the Meiyu season during 2020 over the YHRV region. We also investigated the mechanisms in the different LSMs that might affect precipitation simulations in terms of water and energy cycling. The results showed that the simulated amounts of precipitation were higher under all LSMs than the observations. The main differences occurred in rainstorm areas (>12 mm/day), and the differences in low rainfall areas were not significant (<8 mm/day). Among all of the LSMs, the Simplified Simple Biosphere (SSiB) model obtained the best performance, with the lowest root mean square error and the highest correlation. The SSiB model even outperformed the Bayesian model averaging result. Finally, some factors responsible for the differences modeling results were investigated to understand the related physical mechanism.

A multi-objective framework to select numerical options in air quality prediction models: A case study on dust storm modeling

Article

Full-text available

Dec 2022
SCI TOTAL ENVIRON

Numerical weather prediction models are very important tools in predicting severe weather phenomena such as dust storms. However, the prediction accuracy in these models depends on the options considered in the modeling. In this study, a multi-objective framework is presented to determine the optimal options of the weather research forecasting with chemistry (WRF-Chem) model. For this purpose, a severe dust storm that occurred in the center of Iran is considered and the effect of 10 options including grid (computational domain size, modeling start time, horizontal, vertical and temporal resolution), physical (initial conditions, boundary layer and land surface schemes) and chemical options (dust emission schemes and dust source functions) are investigated. In general, the results showed that the WRF-Chem model has a high ability to model dust storms, but its results depend on the options considered in the modeling. Evaluation of grid options showed that inappropriate selection of domain size and modeling start time can lead to the failure in dust storm forecasting. Also, the land surface scheme has the greatest impact on dust concentration among the physical options. In addition, chemical options have the greatest impact on the dust storm forecasting as well. Based on the proposed multi-objective framework, the optimal options for dust storm modeling were determined. The proposed approach is comprehensive and can be used for other atmospheric/air quality modeling.

Comprehensive evaluation of typical planetary boundary layer (PBL) parameterization schemes in China – Part 2: Influence of uncertainty factors

Article

Full-text available

Nov 2023
GMD

This study focuses on the uncertainties that influence numerical simulation results of meteorological fields (horizontal resolution: 75, 15, and 3 km; vertical resolution: 48 and 62 levels; near-surface (N-S) scheme: MM5 and Eta schemes; initial and boundary conditions: Final (FNL) and European Center for Medium-Range Weather Forecasting (ECMWF) reanalysis data; underlying surface update: model default and latest updates; model version: version 3.6.1 and 3.9.1). By further evaluating and analyzing the uncertainty factors, it is expected to provide relevance for those scholars devoted to factor analysis in order to make the results closer to the observed values. In this study, a total of 12 experiments are set up to analyze the effects of the uncertainties mentioned above, and the following conclusions are drawn: (1) horizontal resolution has a greater effect than vertical resolution; (2) the simulated effects of temperature and wind speed in the N-S scheme are smaller than those in the planetary boundary layer (PBL) scheme; (3) the initial and boundary conditions of different products have the most remarkable effect on relative humidity, while the simulation results of ECMWF data are the best; (4) the updates with urban and water bodies as the underlying surface have a more significant contribution to the meteorological fields, especially on temperature; and (5) for the PBL parameterization schemes, the update of the model version has less impact on the simulation results because each update has small changes and no major changes overall. In general, the configuration of uncertainties needs to be considered comprehensively according to what you need in order to obtain the best simulation results.

Evaluation of the performance of the latest WRF model (V4.5) in simulating surface elements and its sensitivity to the diurnal SST cycle in the South China Sea

Article

Full-text available

Oct 2023
CLIM DYNAM

The performances of six surface layer parameterizations in the latest Weather Research and Forecasting Model (WRF Version 4.5) are examined via observations at Yongxing Island, South China Sea. Three short-term simulations and two long-term simulations were carried out, and four statistical indicators were used to quantify the performance of different schemes. The statistical indicators suggested that all the surface schemes had a better accuracy in simulating surface wind than in simulating surface temperature and moisture. Meanwhile, the unresolved precipitation in the WRF model made significant increases in the model biases for all the schemes. Overall, the Global Forecast System surface scheme performed better in simulating surface variables than other schemes on nonrainy days in the study area. The simulation results also demonstrated that the diurnal SST cycle input could lead to remarkable improvements in the accuracy of the simulated surface heat fluxes and the temperature for all the schemes. The daily averaged determination coefficients between the simulation and observation results almost doubled, and the daily averaged root mean square error (RMSE), as well as the relative RMSE, were reduced by approximately half. However, the surface air specific humidity was not sensitive to the diurnal SST cycle, and only small improvements were identified in these simulations with the diurnal SST cycle. Therefore, the calculation module for the surface humidity in the surface schemes needs to be further optimized.

Examining the Role of the Land Surface on Convection Using High‐Resolution Model Forecasts Over the Southeastern United States

Article

Full-text available

Aug 2022

The influence of the Unified Noah and Noah‐MP land surface models (LSMs) on the evolution of cumulus clouds reaching convective initiation (CI) is assessed using infrared brightness temperatures (BT) from GOES‐16. Cloud properties from individual cloud objects are examined using output from high‐resolution (500 m horizontal grid spacing) model simulations. Cloud objects are tracked over time and related to observed clouds reaching CI to examine differences in cloud extent, longevity, and growth rate. The results demonstrate that differences in assumed surface properties can lead to large discrepancies in the net surface radiative budget, particularly in the sensible and latent heating components where differences exceed 40 W m⁻². These differences lead to changes in the local mesoscale circulation patterns that are more pronounced near the edges of forested and grassland boundaries where lower‐level convergence is stronger. Higher sensible heating from the Noah‐MP LSM produced growth of CI clouds earlier and with increased longevity, which was closer to the timing and growth observed from GOES‐16. The increased cloud growth in the Noah‐MP experiment results from stronger and deeper updrafts, which lofts more cloud water into the upper levels of the troposphere. The weaker updrafts from the Noah LSM experiment results in shallower convection after CI is detected due to slower growth rates. The differences in cloud properties and growth are directly related to the land surfaces they develop above and point to the importance of accurately representing land properties and radiative characteristics when simulating convection in numerical weather prediction models.

Spatial heterogeneity effects on land surface modeling of water and energy partitioning

Article

Full-text available

Jul 2022
GMD

Understanding the influence of land surface heterogeneity on surface water and energy fluxes is crucial for modeling earth system variability and change. This study investigates the effects of four dominant heterogeneity sources on land surface modeling, including atmospheric forcing (ATM), soil properties (SOIL), land use and land cover (LULC), and topography (TOPO). Our analysis focused on their impacts on the partitioning of precipitation (P) into evapotranspiration (ET) and runoff (R), partitioning of net radiation into sensible heat and latent heat, and corresponding water and energy fluxes. An initial set of 16 experiments were performed over the continental US (CONUS) using the E3SM land model (ELMv1) with different combinations of heterogeneous and homogeneous datasets. The Sobol' total and first-order sensitivity indices were utilized to quantify the relative importance of the four heterogeneity sources. Sobol' total sensitivity index measures the total heterogeneity effects induced by a given heterogeneity source, consisting of the contribution from its own heterogeneity (i.e., the first-order index) and its interactions with other heterogeneity sources. ATM and LULC are the most dominant heterogeneity sources in determining spatial variability of water and energy partitioning, mainly contributed by their own heterogeneity and slightly contributed by their interactions with other heterogeneity sources. Their heterogeneity effects are complementary, both spatially and temporally. The overall impacts of SOIL and TOPO are negligible, except TOPO dominates the spatial variability of R/P across the transitional climate zone between the arid western and humid eastern CONUS. Accounting for more heterogeneity sources improves the simulated spatial variability of water and energy fluxes when compared with ERA5-Land reanalysis dataset. An additional set of 13 experiments identified the most critical components within each heterogeneity source, which are precipitation, temperature, and longwave radiation for ATM, soil texture, and soil color for SOIL and maximum fractional saturated area parameter for TOPO.

Microphysics affect the sensitivities of rainfall to different horizontal-resolution simulations: Evidence from a case study of the Weather Research and Forecasting model runs

Article

Sep 2023
ATMOS RES

The Weather Research and Forecasting (WRF) model was used to examine the sensitivities of Typhoon Fitow (2013) to the variations in horizontal mesh spacing ranging from 9 km to 1 km and to different microphysics schemes. The minimum sea level pressure decreased by 5 hPa and the maximum wind speeds increased by 20 m s− 1 near the typhoon center as the horizontal grid spacing decreased from 9 km, 3 km, and 1 km in both Purdue-Lin and National Severe Storms Laboratory (NSSL) microphysics. The strengthening of the tropical cyclone may have been due to similar physical processes in the Purdue-Lin and NSSL simulations. Fine-resolution simulations produced localized and intense rainfall, in correspondence to localized and intense upward motions. Moreover, as the horizontal mesh spacing reduced from 9 km to 1 km, the distributions of the upward and downward motions broadened and the radar reflectivity bins with contour frequencies >10% increased. The simulated rainfall in the NSSL with a horizontal grid spacing below 3 km and in Purdue-Lin with a 1 km grid spacing reproduced spatial and temporal distributions similar to the actual observations. This suggested that the elaborate microphysics may compensate for the lack of horizontal resolution, to some extent. The precipitation budget analysis further suggested that the diminishment of rainfall was attributed to the reduced net condensation and hydrometeor convergence respectively in the Purdue-Lin and NSSL scheme as the mesh spacing reduced, which was further attributable to the decreased condensation or deposition.

Cascade sensitivity tests to model deep convective systems in complex orography with WRF

Article

Full-text available

Aug 2023
ATMOS RES

WRF v.3.9 sensitivity to land surface model and horizontal resolution changes over North America

Preprint

Full-text available

Aug 2021

Understanding the differences between regional simulations of land-atmosphere interactions and near-surface conditions is crucial for a more reliable representation of past and future climate. Here, we explore the effect of changes in the model's horizontal resolution on the simulated energy balance at the surface and near-surface conditions using the Weather Research and Forecasting (WRF) model. To this aim, an ensemble of twelve simulations using three different horizontal resolutions (25 km, 50 km and 100 km) and four different Land Surface Model (LSM) configurations over North America from 1980 to 2013 is developed. Our results show that finer resolutions lead to higher surface net shortwave radiation and maximum temperatures at mid- and high latitudes. At low latitudes over coastal areas, an increase in resolution leads to lower values of sensible heat flux and higher values of latent heat flux, as well as lower values of surface temperatures and higher values of precipitation and soil moisture in summer. The use of finer resolutions leads then to an increase in summer values of latent heat flux, convective and non-convective precipitation and soil moisture at low latitudes. The effect of the LSM choice is larger than the effect of horizontal resolution on the near-surface temperature conditions. By contrast, the effect of the LSM choice on the simulation of precipitation is weaker than the effect of horizontal resolution, showing larger differences among LSM simulations in summer and over regions with high latent heat flux. Comparison between observations and the simulation of daily maximum and minimum temperatures and accumulated precipitation indicates that the CLM4 LSM yields the lowest biases in maximum and minimum mean temperatures, but the highest biases in extreme temperatures. Increasing horizontal resolution leads to larger biases in accumulated precipitation over all regions particularly in summer. The reasons behind relate the partition between convective and non-convective precipitation, specially noticeable over western US.

The benefits of increasing resolution in global and regional climate simulations for European climate extremes

Article

Full-text available

Nov 2020
GMD

Many climate extremes, including heatwaves and heavy precipitation events, are projected to worsen under climate change, with important impacts for society. Future projections required for adaptation are often based on climate model simulations. Given finite resources, trade-offs must be made concerning model resolution, ensemble size, and level of model complexity. Here we focus on the resolution component. A given resolution can be achieved over a region using either global climate models (GCMs) or at lower cost using regional climate models (RCMs) that dynamically downscale coarser GCMs. Both approaches to increasing resolution may better capture small-scale processes and features (downscaling effect), but increased GCM resolution may also improve the representation of the large-scale atmospheric circulation (upscaling effect). The size of this upscaling effect is therefore important for deciding modelling strategies. Here we evaluate the benefits of increased model resolution for both global and regional climate models for simulating temperature, precipitation, and wind extremes over Europe at resolutions that could currently be realistically used for coordinated sets of climate projections at the pan-European scale. First we examine the benefits of regional downscaling by comparing EURO-CORDEX simulations at 12.5 and 50 km resolution to their coarser CMIP5 driving simulations. Secondly, we compare global-scale HadGEM3-A simulations at three resolutions (130, 60, and 25 km). Finally, we separate out resolution-dependent differences for HadGEM3-A into downscaling and upscaling components using a circulation analogue technique. Results suggest limited benefits of increased resolution for heatwaves, except in reducing hot biases over mountainous regions. Precipitation extremes are sensitive to resolution, particularly over complex orography, with larger totals and heavier tails of the distribution at higher resolution, particularly in the CORDEX vs. CMIP5 analysis. CMIP5 models underestimate precipitation extremes, whilst CORDEX simulations overestimate compared to E-OBS, particularly at 12.5 km, but results are sensitive to the observational dataset used, with the MESAN reanalysis giving higher totals and heavier tails than E-OBS. Wind extremes are somewhat stronger and heavier tailed at higher resolution, except in coastal regions where large coastal grid boxes spread strong ocean winds further over land. The circulation analogue analysis suggests that differences with resolution for the HadGEM3-A GCM are primarily due to downscaling effects.

Land surface model influence on the simulated climatologies of temperature and precipitation extremes in the WRF v3.9 model over North America

Article

Full-text available

Nov 2020
GMD

The representation and projection of extreme temperature and precipitation events in regional and global climate models are of major importance for the study of climate change impacts. However, state-of-the-art global and regional climate model simulations yield a broad inter-model range of intensity, duration and frequency of these extremes. Here, we present a modeling experiment using the Weather Research and Forecasting (WRF) model to determine the influence of the land surface model (LSM) component on uncertainties associated with extreme events. First, we analyze land–atmosphere interactions within four simulations performed by the WRF model from 1980 to 2012 over North America, using three different LSMs. Results show LSM-dependent differences at regional scales in the frequency of occurrence of events when surface conditions are altered by atmospheric forcing or land processes. The inter-model range of extreme statistics across the WRF simulations is large, particularly for indices related to the intensity and duration of temperature and precipitation extremes. Our results show that the WRF simulation of the climatology of heat extremes can be 5 ∘C warmer and 6 d longer depending on the employed LSM component, and similarly for cold extremes and heavy precipitation events. Areas showing large uncertainty in WRF-simulated extreme events are also identified in a model ensemble from three different regional climate model (RCM) simulations participating in the Coordinated Regional Climate Downscaling Experiment (CORDEX) project, revealing the implications of these results for other model ensembles. Thus, studies based on multi-model ensembles and reanalyses should include a variety of LSM configurations to account for the uncertainty arising from this model component or to test the performance of the selected LSM component before running the whole simulation. This study illustrates the importance of the LSM choice in climate simulations, supporting the development of new modeling studies using different LSM components to understand inter-model differences in simulating extreme temperature and precipitation events, which in turn will help to reduce uncertainties in climate model projections.

The making of the New European Wind Atlas – Part 1: Model sensitivity

Article

Full-text available

Oct 2020
GMD

This is the first of two papers that document the creation of the New European Wind Atlas (NEWA). It describes the sensitivity analysis and evaluation procedures that formed the basis for choosing the final setup of the mesoscale model simulations of the wind atlas. The suitable combination of model setup and parameterizations, bound by practical constraints, was found for simulating the climatology of the wind field at turbine-relevant heights with the Weather Research and Forecasting (WRF) model. Initial WRF model sensitivity experiments compared the wind climate generated by using two commonly used planetary boundary layer schemes and were carried out over several regions in Europe. They confirmed that the most significant differences in annual mean wind speed at 100 m a.g.l. (above ground level) mostly coincide with areas of high surface roughness length and not with the location of the domains or maximum wind speed. Then an ensemble of more than 50 simulations with different setups for a single year was carried out for one domain covering northern Europe for which tall mast observations were available. We varied many different parameters across the simulations, e.g. model version, forcing data, various physical parameterizations, and the size of the model domain. These simulations showed that although virtually every parameter change affects the results in some way, significant changes in the wind climate in the boundary layer are mostly due to using different physical parameterizations, especially the planetary boundary layer scheme, the representation of the land surface, and the prescribed surface roughness length. Also, the setup of the simulations, such as the integration length and the domain size, can considerably influence the results. We assessed the degree of similarity between winds simulated by the WRF ensemble members and the observations using a suite of metrics, including the Earth Mover's Distance (EMD), a statistic that measures the distance between two probability distributions. The EMD was used to diagnose the performance of each ensemble member using the full wind speed and direction distribution, which is essential for wind resource assessment. We identified the most realistic ensemble members to determine the most suitable configuration to be used in the final production run, which is fully described and evaluated in the second part of this study (Dörenkämper et al., 2020).

An Assessment of Observed and Simulated Temperature Variability in Sierra de Guadarrama

Article

Full-text available

Sep 2020

This work provides a first assessment of temperature variability at interannual and decadal timescales in Sierra de Guadarrama, a high mountain protected area of the Central System in the Iberian Peninsula. Observational data from stations located in the area and simulated data from a high-resolution simulation (1 km) with the Weather Research and Forecasting (WRF) model, fed from ERA Interim reanalysis, are used in order to analyse the temperature variability in the period 2000–2018. Comparison among all datasets allows evaluation of the realism of the model simulations. The results show that the model tends to underestimate the observational mean temperatures and anomalies at high-altitude stations. A linear mean temperature vertical gradient of −5.81 °C/km is observed, but it is overestimated by the model (−6.56 °C/km). The variability of the daily temperature anomalies for both observations and, to a lesser extent, simulations increases with height. The added value that the WRF offers against the use of the ERA Interim is evaluated. The results show that the WRF provides a better performance than the reanalysis, as it shows smaller biases with respect to observational temperature anomalies. Finally, the study of temperature trends over the Sierra de Guadarrama and its surroundings for the period 2000–2018 shows a warming in the area, significantly pronounced in autumn. When extended to the last decades, observations show that this warming has been happening since the first half of the 20th century, especially during the period 1970–2018, but not as much as during 2000–2018.

The Community Land Model Version 5: Description of New Features, Benchmarking, and Impact of Forcing Uncertainty

Article

Full-text available

Dec 2019

The Community Land Model (CLM) is the land component of the Community Earth System Model (CESM) and is used in several global and regional modeling systems. In this paper, we introduce model developments included in CLM version 5 (CLM5), which is the default land component for CESM2. We assess an ensemble of simulations, including prescribed and prognostic vegetation state, multiple forcing data sets, and CLM4, CLM4.5, and CLM5, against a range of metrics including from the International Land Model Benchmarking (ILAMBv2) package. CLM5 includes new and updated processes and parameterizations: (1) dynamic land units, (2) updated parameterizations and structure for hydrology and snow (spatially explicit soil depth, dry surface layer, revised groundwater scheme, revised canopy interception and canopy snow processes, updated fresh snow density, simple firn model, and Model for Scale Adaptive River Transport), (3) plant hydraulics and hydraulic redistribution, (4) revised nitrogen cycling (flexible leaf stoichiometry, leaf N optimization for photosynthesis, and carbon costs for plant nitrogen uptake), (5) global crop model with six crop types and time‐evolving irrigated areas and fertilization rates, (6) updated urban building energy, (7) carbon isotopes, and (8) updated stomatal physiology. New optional features include demographically structured dynamic vegetation model (Functionally Assembled Terrestrial Ecosystem Simulator), ozone damage to plants, and fire trace gas emissions coupling to the atmosphere. Conclusive establishment of improvement or degradation of individual variables or metrics is challenged by forcing uncertainty, parametric uncertainty, and model structural complexity, but the multivariate metrics presented here suggest a general broad improvement from CLM4 to CLM5.

Assessment of simulated soil moisture from WRF Noah, Noah-MP, and CLM land surface schemes for landslide hazard application

Article

Full-text available

Oct 2019
HESS

This study assesses the usability of Weather Research and Forecasting (WRF) model simulated soil moisture for landslide monitoring in the Emilia Romagna region, northern Italy, during the 10-year period between 2006 and 2015. In particular, three advanced land surface model (LSM) schemes (i.e. Noah, Noah-MP, and CLM4) integrated with the WRF are used to provide detailed multi-layer soil moisture information. Through the temporal evaluation with the single-point in situ soil moisture observations, Noah-MP is the only scheme that is able to simulate the large soil drying phenomenon close to the observations during the dry season, and it also has the highest correlation coefficient and the lowest RMSE at most soil layers. It is also demonstrated that a single soil moisture sensor located in a plain area has a high correlation with a significant proportion of the study area (even in the mountainous region 141 km away, based on the WRF-simulated spatial soil moisture information). The evaluation of the WRF rainfall estimation shows there is no distinct difference among the three LSMs, and their performances are in line with a published study for the central USA. Each simulated soil moisture product from the three LSM schemes is then used to build a landslide prediction model, and within each model, 17 different exceedance probability levels from 1 % to 50 % are adopted to determine the optimal threshold scenario (in total there are 612 scenarios). Slope degree information is also used to separate the study region into different groups. The threshold evaluation performance is based on the landslide forecasting accuracy using 45 selected rainfall events between 2014 and 2015. Contingency tables, statistical indicators, and receiver operating characteristic analysis for different threshold scenarios are explored. The results have shown that, for landslide monitoring, Noah-MP at the surface soil layer with 30 % exceedance probability provides the best landslide monitoring performance, with its hit rate at 0.769 and its false alarm rate at 0.289.

Characterization of Air and Ground Temperature Relationships within the CMIP5 Historical and Future Climate Simulations

Article

Full-text available

Apr 2019

The relationships between air and ground surface temperatures across North America are examined in the historical and future projection simulations from 32 general circulation models (GCMs) included in the fifth phase of the Coupled Model Intercomparison Project (CMIP5). The difference between surface air (2 m) and ground surface (10 cm) temperatures is affected by simulated snow cover, vegetation cover, and precipitation by means of changes in soil moisture and soil properties. In winter, the differences between air and ground surface temperatures, for all CMIP5 simulations, are related to the insulating effect of snow cover and soil freezing phenomena. In summer, large leaf area index and large precipitation rates correspond to smaller differences between air and ground temperatures for the majority of simulations, likely due to induced changes in latent and sensible heat fluxes at the ground surface. Our results show that the representation of air‐ground coupling, analyzed using the difference between ground and air surface temperatures as metric, differs from observations, the North American Regional Reanalysis product and among the CMIP5 GCM simulations, by amounts that depend on the employed land surface model. The large variability among GCMs and the marked dependence of the results on the choice of the land surface model illustrate the need for improving the representation of processes controlling the coupling of the lower atmosphere and the land surface in GCMs as a mean of reducing the variability in their representation of weather and climate phenomena.

Evaluation of WRF Modeling in Relation to Different Land Surface Schemes and Initial and Boundary Conditions: A Snow Event Simulation Over the Tibetan Plateau

Article

Full-text available

Jan 2019

Snowfall and the subsequent evolution of the snowpack play important roles in the cryospheric and hydrospheric processes that occur on the Tibetan Plateau (TP). Current literature provides scarce evidence covering the sensitivity of solid precipitation to land surface physics schemes and initial and boundary conditions on the TP. Six numerical experiments using the Weather Research and Forecasting (WRF) model were conducted to simulate a snow event over the TP in March 2017. Different land surface physics schemes, that is, Community Land Model (CLM), Noah, and Noah‐MP, and initial and boundary conditions provided by atmospheric reanalysis data sets, that is, the National Centers for Environmental Prediction‐FNL and ERA‐Interim data sets, were applied in sensitivity analyses. The observed near‐surface air temperature, snow depth, and snow water equivalent (SWE) values were used to evaluate each model's performance. The results demonstrate that (1) the sensitivity of the near‐surface air temperature to land surface physics schemes is greater than it is to both the initial and boundary conditions; (2) the best performance is achieved when applying WRF + CLM with a root‐mean‐square error of 8.4 °C, a mean absolute deviation of 7.3 °C, a correlation coefficient of 0.75, and a spatial correlation coefficient of ~0.5 to air temperature estimates. A potentially important factor appears to be the advanced parametrization of albedo in the CLM scheme; (3) the advanced land surface schemes in the WRF model describes the physics of cryospheric and hydrospheric processes in detail, and the land surface response is determined by multiple variables and parameters in such schemes. The spatial patterns in such variables and parameters determined the detailed spatial variabilities observed in snow cover and amount and its temporal evolution. The WRF model overestimates, however, the intensity and extent of snow depth and SWE; (4) simulations of solid precipitation are more accurate when applying CLM or Noah‐MP + ERA‐Interim in WRF; and (5) WRF performance with regard to SWE estimates clearly depends upon the discrimination of lighter from heavier snowfall.

Separating the Impact of Individual Land Surface Properties on the Terrestrial Surface Energy Budget in both the Coupled and Uncoupled Land–Atmosphere System

Article

Jun 2019

Changes in the land surface can drive large responses in the atmosphere on local, regional, and global scales. Surface properties control the partitioning of energy within the surface energy budget to fluxes of shortwave and longwave radiation, sensible and latent heat, and ground heat storage. Changes in surface energy fluxes can impact the atmosphere across scales through changes in temperature, cloud cover, and large-scale atmospheric circulation. We test the sensitivity of the atmosphere to global changes in three land surface properties: albedo, evaporative resistance, and surface roughness. We show the impact of changing these surface properties differs drastically between simulations run with an offline land model, compared to coupled land–atmosphere simulations that allow for atmospheric feedbacks associated with land–atmosphere coupling. Atmospheric feedbacks play a critical role in defining the temperature response to changes in albedo and evaporative resistance, particularly in the extratropics. More than 50% of the surface temperature response to changing albedo comes from atmospheric feedbacks in over 80% of land areas. In some regions, cloud feedbacks in response to increased evaporative resistance result in nearly 1 K of additional surface warming. In contrast, the magnitude of surface temperature responses to changes in vegetation height are comparable between offline and coupled simulations. We improve our fundamental understanding of how and why changes in vegetation cover drive responses in the atmosphere, and develop understanding of the role of individual land surface properties in controlling climate across spatial scales—critical to understanding the effects of land-use change on Earth’s climate.

WRF v.3.9 sensitivity to land surface model and horizontal resolution changes over North America

Abstract and Figures

Recommended publications

WRF v.3.9 sensitivity to land surface model and horizontal resolution changes over North America

Land surface model influence on the simulated climatologies of temperature and precipitation extreme...

Land Surface Model influence on the simulated climatologies of temperature and precipitation extreme...

Characterization of Air and Ground Temperature Relationships within the CMIP5 Historical and Future...