ArticlePDF Available

Simulation of the present-day climate with the climate model INMCM5


Abstract and Figures

In this paper we present the fifth generation of the INMCM climate model that is being developed at the Institute of Numerical Mathematics of the Russian Academy of Sciences (INMCM5). The most important changes with respect to the previous version (INMCM4) were made in the atmospheric component of the model. Its vertical resolution was increased to resolve the upper stratosphere and the lower mesosphere. A more sophisticated parameterization of condensation and cloudiness formation was introduced as well. An aerosol module was incorporated into the model. The upgraded oceanic component has a modified dynamical core optimized for better implementation on parallel computers and has two times higher resolution in both horizontal directions. Analysis of the present-day climatology of the INMCM5 (based on the data of historical run for 1979-2005) shows moderate improvements in reproduction of basic circulation characteristics with respect to the previous version. Biases in the near-surface temperature and precipitation are slightly reduced compared with INMCM4 as well as biases in oceanic temperature, salinity and sea surface height. The most notable improvement over INMCM4 is the capability of the new model to reproduce the equatorial stratospheric quasi-biannual oscillation and statistics of sudden stratospheric warmings.
This content is subject to copyright. Terms and conditions apply.
1 23
Climate Dynamics
Observational, Theoretical and
Computational Research on the Climate
ISSN 0930-7575
Clim Dyn
DOI 10.1007/s00382-017-3539-7
Simulation of the present-day climate with
the climate model INMCM5
E.M.Volodin, E.V.Mortikov,
S.V.Kostrykin, V.Ya.Galin,
V.N.Lykossov, A.S.Gritsun,
N.A.Diansky, A.V.Gusev, et al.
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer-
Verlag Berlin Heidelberg. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at”.
1 3
Clim Dyn
DOI 10.1007/s00382-017-3539-7
Simulation ofthepresent-day climate withtheclimate model
E.M.Volodin1,2 · E.V.Mortikov1,2· S.V.Kostrykin1· V.Ya.Galin1·
V.N.Lykossov1,2· A.S.Gritsun1· N.A.Diansky1· A.V.Gusev1· N.G.Iakovlev1
Received: 30 June 2016 / Accepted: 14 January 2017
© Springer-Verlag Berlin Heidelberg 2017
1 Introduction
This paper is devoted to the new version of the INMCM
climate model that is being developed at the Institute of
Numerical Mathematics of the Russian Academy of Sci-
ences. This version (INMCM5) is an evolutionary upgrade
of the previous version INMCM4, which was part of the
CMIP5 (Coupled Model Intercomparison Project, Phase 5,
Taylor etal. 2012). With the new version we are trying to
reduce systematic model biases indicated in several papers
devoted to the analysis of CMIP5 data and improve repro-
duction of some key processes responsible for the seasonal
and interannual predictability.
The paper is structured as follows. First, we briefly
describe the main features of the INMCM4’s atmospheric
and oceanic components as well as correspondent changes
made in the INMCM5 [a detailed description of INMCM4
can be found in Volodin etal. (2010)]. Key improvements
include the increase of the vertical resolution in the atmos-
pheric module, revision of the large-scale condensation and
cloud formation parameterizations, and the newly devel-
oped aerosol block. In the block of the ocean dynamics, the
integration scheme for advection was changed (from coor-
dinates splitting to an explicit one) and an iterative method
for solving linear shallow water equation systems replaced
the direct method used in INMCM4 (these changes were
necessary to improve model scalability on parallel com-
puters). In addition, the horizontal resolution of the oce-
anic model was doubled. The INMCM5 program code was
reworked for better performance on parallel computers with
distributed memory.
The second half of the paper describes the INMCM5
biases in the reproduction of the present-day climate with
respect to the INMCM4 climate characteristics [Volodin
Abstract In this paper we present the fifth generation of
the INMCM climate model that is being developed at the
Institute of Numerical Mathematics of the Russian Acad-
emy of Sciences (INMCM5). The most important changes
with respect to the previous version (INMCM4) were made
in the atmospheric component of the model. Its vertical res-
olution was increased to resolve the upper stratosphere and
the lower mesosphere. A more sophisticated parameteriza-
tion of condensation and cloudiness formation was intro-
duced as well. An aerosol module was incorporated into the
model. The upgraded oceanic component has a modified
dynamical core optimized for better implementation on par-
allel computers and has two times higher resolution in both
horizontal directions. Analysis of the present-day climatol-
ogy of the INMCM5 (based on the data of historical run for
1979–2005) shows moderate improvements in reproduction
of basic circulation characteristics with respect to the previ-
ous version. Biases in the near-surface temperature and pre-
cipitation are slightly reduced compared with INMCM4 as
well as biases in oceanic temperature, salinity and sea sur-
face height. The most notable improvement over INMCM4
is the capability of the new model to reproduce the equato-
rial stratospheric quasi-biannual oscillation and statistics of
sudden stratospheric warmings.
Keywords Climate· Model· Atmosphere· Ocean·
Parameterization· Simulation· Temperature·
Precipitation· Bias
* E. M. Volodin
1 Institute ofNumerical Mathematic, Gubkina 8,
Moscow119333, Russia
2 Moscow State University, Leninskie Gory 1, Moscow, Russia
Author's personal copy
E.M.Volodin et al.
1 3
etal. (2010, 2013), Volodin (2013)] and that of the other
CMIP5 models.
2 Model description
The family of INMCM climate models, as most climate
system models, consists of two main blocks: the atmos-
phere general circulation model, and the ocean general
circulation model. The atmospheric part is based on the
standard set of hydrothermodynamic equations with hydro-
static approximation written in advective form. The model
prognostic variables are wind horizontal components, tem-
perature, specific humidity and surface pressure. The sec-
ond order finite difference approximation uses the Arakawa
C-grid (the spatial coordinates are geographical latitude,
longitude and vertical sigma-coordinate). The leapfrog
scheme with Asselin (1972) filtering is used for time step-
ping. The gravity waves are treated by implicit time scheme
to improve numerical stability. Near the poles, the Fou-
rier filter is applied along longitudinal direction to avoid
numerical instability. Detailed description of the model’s
finite difference schemes as well as the set of the equations
could be found in Alekseev etal. (1998) and Galin et al.
(2003). The model version INMCM5 has a spatial resolu-
tion of 2× 1.5° in longitude and latitude, and 73 levels in
vertical. Lowermost and uppermost levels are placed at
σ = 0.993 and σ = 0.0002, respectively. In the stratosphere,
the model’s vertical resolution is about 500m [this reso-
lution is important for correct description of interactions
between the gravity wave drag parameterization and large
scale flow to reproduce the Equatorial quasi-biannual oscil-
lation (QBO)]. The time step in the dynamical block is
5min for this particular spatial resolution. Previous model
version (INMCM4) had similar atmospheric dynamical
core, but with the uppermost level located at σ = 0.01 and
with 21 vertical levels.
The INMCM5 borrows most of the atmospheric parame-
terizations from its previous version. One of the few notable
changes is the new parameterization of clouds and large-
scale condensation. In the INMCM5 cloud area and cloud
water are computed prognostically according to Tiedtke
(1993). That includes the formation of large-scale cloudi-
ness as well as the formation of clouds in the atmospheric
boundary layer and clouds of deep convection. Decrease of
cloudiness due to mixing with unsaturated environment and
precipitation formation are also taken into account. Evapo-
ration of precipitation is implemented according to Kessler
(1969). The INMCM4 in turn, determined cloud amount
diagnostically. Similarly, cloud water is a diagnostically
calculated function of temperature and pressure. Large
scale condensation is obtained assuming that all specific
humidity exceeding the saturation threshold value instantly
falls as precipitation [see Alekseev etal. (1998) for details].
Other atmospheric parameterizations in the INMCM5
are identical to the ones in the INMCM4 (some coefficients
were re-adjusted to account for the changes in the vertical
resolution). Deep and shallow convection parameteriza-
tions are analogous to Betts (1986), but with an additional
mixing of momentum and with the penetration of deep
convection a little higher than the level of zero buoyancy.
Orographical and nonorographical gravity wave drags are
implemented according to Palmer etal. (1986) and Hines
(1997). In addition, nonorographical wave drag parameteri-
zation include vertical diffusion induced by the breaking of
gravity waves.
The land surface and soil are represented according to
Volodin and Lykossov (1998). Prognostic equations for
soil temperature and soil specific humidity are solved at 23
levels from the surface to 10m of depth, including freez-
ing/melting of soil water. Spatial distribution of potential
vegetation is prescribed, and actual vegetation is calcu-
lated using soil moisture in root zone. Maximum leaf area
index for each vegetation type is prescribed, and actual
leaf area index is determined using soil moisture and soil
Atmospheric radiation is calculated the same way as in
Galin (1998). Solar spectrum is divided by four intervals
and we use ten intervals to approximate the longwave part
of the spectrum.
In the atmospheric boundary layer, vertical diffusion is
applied to the prognostic variables (Alekseev etal. 1998).
Entrainment of potential temperature, specific humidity,
cloud area and cloud water at the top of the boundary layer
are considered as in Tiedtke (1993). Calculation of cloud
formation and condensation occurs at each time step of
atmospheric dynamics. Atmospheric radiation is calculated
once per 3 h, while the other atmospheric parameteriza-
tions are called every hour.
In the INMCM5 the atmospheric model is comple-
mented by the interactive aerosol block, which is absent in
the INMCM4. Concentrations of coarse and fine sea salt,
coarse and fine mineral dust, SO2, sulfate aerosol, hydro-
philic and hydrophobic black and organic carbon are all
calculated prognostically. Dynamic processes described by
the aerosol model include prescribed and simulated aerosol
sources, advection, gravitational falling, wet and dry depo-
sition and removal by precipitation. The model also takes
into account the direct and indirect aerosol radiation effect
that is connected with cloud drop radius and cloud water
life time. The detailed description of the aerosol block used
in the INMCM5 can be found in Volodin and Kostrykin
The oceanic module of the INMCM5 uses generalized
spherical coordinates. The model “South Pole” coincides
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
with the geographical one, while the model “North Pole” is
located in Siberia beyond the ocean area to avoid numeri-
cal problems near the pole. Vertical sigma-coordinate is
used. The finite-difference equations are written using the
Arakawa C-grid. The differential and finite-difference equa-
tions, as well as methods of solving them can be found in
Zalesny etal. (2010). The INMCM5 uses explicit schemes
for advection, while the INMCM4 used schemes based on
splitting upon coordinates. Also, the iterative method for
solving linear shallow water equation systems is used in the
INMCM5 rather than direct method used in the INMCM4.
The two previous changes were made to improve model
parallel scalability. The horizontal resolution of the ocean
part of the INMCM5 is 0.5 × 0.25° in longitude and latitude
(compared to the INMCM4’s 1 × 0.5°). Both the INMCM4
and the INMCM5 have 40 levels in vertical. The parallel
implementation of the ocean model can be found in (Terek-
hov etal. 2011). The oceanic block includes vertical mix-
ing and isopycnal diffusion parameterizations (Zalesny
et al. 2010). Sea ice dynamics and thermodynamics are
parameterized according to Iakovlev (2009). Assumptions
of elastic-viscous-plastic rheology and single ice thickness
gradation are used. The time step in the oceanic block of
the INMCM5 is 15min.
The climate model INMCM5 has а carbon cycle mod-
ule (Volodin 2007), where atmospheric CO2 concentration,
Fig. 1 Near-surface air
temperature for annual (top),
December-February (middle)
and June–August (bottom)
means. Shading (C) represents
model bias with respect to
ERA-Interim reanalysis data
while isolines (at −50, −30,
−10, 0, 5, 10, 15, 20 and 25C)
show model climatology. The
1979–2005 data interval was
used to calculate averages
Author's personal copy
E.M.Volodin et al.
1 3
carbon in vegetation, soil and ocean are calculated. In soil,
а single carbon pool is considered. In the ocean, the only
prognostic variable in the carbon cycle is total inorganic
carbon. Biological pump is prescribed. The model calcu-
lates methane emission from wetlands and has a simplified
methane cycle (Volodin 2008). Parameterizations of some
electrical phenomena, including calculation of ionospheric
potential and flash intensity (Mareev and Volodin 2014),
are also included in the model.
The codes of the atmospheric block, aerosol block and
oceanic block are adopted for parallel computers by two-
dimensional decomposition. The program’s realization of
the climate model allows for distributing calculations of
atmospheric dynamics, atmospheric aerosol and oceanic
dynamics on different groups of processors using the MPI
(Message Passing Interface) library. This possibility is also
provided for the advection of oceanic tracers. The sea ice
module is included in the oceanic block. The soil, surface
and vegetation modules are included in the atmospheric
block. The atmospheric and oceanic blocks exchange data
once per 2 h. The atmospheric dynamics data are sent to
the aerosol block at each dynamical time step. The aerosol
concentration data are sent to the atmospheric block once
per 3h.
To reduce the number of synchronization points, the
exchanges between the atmosphere and aerosol blocks
Fig. 2 Precipitation (mm
day−1) for annual (top), Decem-
ber-February (middle) and
June–August (bottom) means.
Shading represents model bias
with respect to GPCP 2.2 data,
while isolines (at 1, 2, 3, 6, 10
and 14mm day−1) show model
climatology. The 1979–2005
data interval was used to calcu-
late averages
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
are carried out asynchronously with additional buffer-
ing of messages. Numerical tests show that for the cur-
rent model version, which has 10 prognostic variables
in the aerosol block, an equal number of processors for
atmospheric dynamics and aerosol block is optimal (giv-
ing acceleration by the factor of 2). For the INMCM5 the
optimum number of cores at the supercomputer Lomon-
osov located in the Moscow State University is 96 for
the atmospheric block, 96 for the aerosol block and 192
for the oceanic block (384 cores in total). Model perfor-
mance in this case is about 6 model years per day.
Several INMCM5 model versions exist and share simi-
lar dynamical cores, parameterizations and parallel archi-
tecture, but have different spatial resolutions. The aim is to
simulate the climate and climate changes at different time
scales: from seasons to millennia. In addition to the basic
model INMCM5 we have a model version for paleoclimate
modeling (with a resolution of 5 × 4° and 21 levels in the
atmosphere, and 2.5 × 2° and 40 levels in the ocean). The
model INMCM5H with finer resolution (0.67 × 0.5° and 73
levels in the atmosphere and 0.167 × 0.125° and 40 levels in
the ocean) is capable of running for several decades and is
aimed at HighResMIP experiments.
Fig. 3 Sea level pressure (hPa)
for annual (top), December-
February (middle) and
June–August (bottom) means.
Shading represents model bias
with respect to the ERA-Interim
reanalysis, while isolines (at
990, 1000, 1010 and 1020hPa)
show model climatology
Author's personal copy
E.M.Volodin et al.
1 3
3 Simulation ofthepresent day climate
To estimate the quality of the new climate system and com-
pare it with the previous one we performed a historical run
for 1850–2005, where all forcings were specified according
to the CMIP6 historical run protocol (https://www.wcrp- The initial condi-
tions were taken from the model preindustrial run (where
all the forcings were prescribed at the year 1850 conditions)
after 300 years of integration. We mostly analyzed averages
(seasonal or annual) over years 1979–2005 and restricted
our attention to consideration of some basic features of
atmospheric and oceanic dynamics and thermodynamics.
3.1 Atmosphere
Figure 1 presents the INMCM5 model bias in the near
surface temperature with respect to the ERA-Interim
(Dee et al. 2011) for annual, December-February and
June–August means for 1979–2005. For the annual mean
conditions, one can see cold bias over Arctic with the
maximum in Greenland Sea and cold bias over Antarctica.
Possible source of these errors is, probably, an insufficient
amount of cloud ice in upper polar cloudiness. Positive
temperature bias over the Southern ocean near Antarc-
tica (mostly evident in the summer) can be attributed to
underestimation of cloud radiation forcing (CRF) and
Fig. 4 Shortwave (top), long-
wave (middle) and net (bottom)
annual mean cloud radiative
forcing (Wm−2). Shading repre-
sents model bias with respect to
the CERES v2.8 for 2000–2005,
while isolines (from −60 to 40
Wm−2) show model climatology
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
underestimation of sea ice compactness in the regions cov-
ered by ice.
When compared to the INMCM4 surface temperature
climatology, the INMCM5 shows several improvements.
Negative bias over continents is reduced mainly because
of the increase in daily minimum temperature over land,
which is achieved by tuning the surface flux parameteriza-
tion. In addition, positive bias over southern Europe and
eastern USA in summer typical for many climate models
(Mueller and Seneviratne 2014) is almost absent in the
INMCM5. A possible reason for this bias in many models
is the shortage of soil water and suppressed evaporation
leading to overestimation of the surface temperature. In
the INMCM5 this problem was addressed by the increase
of the minimum leaf resistance for some vegetation types.
Nevertheless, some problems migrate from one model ver-
sion to the other: negative bias over most of the subtropical
and tropical oceans, and positive bias over the Atlantic to
the east of the USA and Canada. Root mean square (RMS)
error of annual mean near surface temperature was reduced
from 2.48K in the INMCM4 to 1.85K in the INMCM5.
Model bias in precipitation with respect to GPCP v.2.2
(Adler etal. 2003) can be seen in Fig.2. Over the tropics,
one can see the typical precipitation error for contempo-
rary climate models: overestimation of precipitation in the
western Indian Ocean, Indonesia, Atlantic Ocean south of
Fig. 5 Annual mean zonal
mean air temperature (K,
top) and zonal mean zonal
wind (ms−1, bottom). Shad-
ing represents model bias with
respect to the ERA-Interim data
for 1979–2005, isolines show
model climatology
Author's personal copy
E.M.Volodin et al.
1 3
the Equator, and in the tropical Pacific outside of Equa-
tor. Model precipitation is underestimated over the equa-
torial Pacific, Central America and Brazil. All the listed
shortcomings are seen in the figure representing model
bias averaged over all the CMIP5 models (Flato et al.
2013, Fig. 9.4). In midlatitudes, the positive precipitation
bias over the ocean prevails in winter while negative bias
occurs in summer. Compared to the INMCM4, the biases
over the western Indian Ocean, Indonesia, the eastern tropi-
cal Pacific and the tropical Atlantic are reduced. A possi-
ble reason for this is the better reproduction of the tropi-
cal sea surface temperature (SST) in the INMCM5 due to
the increase of the spatial resolution in the oceanic block,
as well as the new condensation scheme. RMS annual
mean model bias for precipitation is 1.35mm day−1 for the
INMCM5 compared to 1.60mm day−1 for the INMCM4.
Figure3 presents model bias in sea level pressure (SLP)
with respect to the 1979–2005 ERA-Interim data. The
model underestimates annual mean SLP over most parts
of Africa and southern Eurasia, and overestimates it in the
North Pacific and Atlantic sector of the Arctic. These short-
comings do not change significantly from season to season.
In the southern midlatitudes, positive bias in summer tends
to compensate for negative bias in winter. Large SLP bias
in the Antarctic in Austral winter could be attributed to the
discrepancy between model and ERA-Interim reanalysis
methods for calculation of SLP in regions with high eleva-
tion. Large bias in the Tibet has probably the same origin.
Fig. 6 Annual cycle of zonal
mean zonal wind (m s−1) at
10hPa, 60N (top) and 60S (bot-
tom). Red line model data, black
line ERA Interim data. The
1979–2005 period was used for
calculation of averages
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
The RMS bias of annual mean SLP is equal to 1.86hPa for
the INMCM5, while it is 2.15hPa for INMCM4.
Cloud radiation forcing (CRF) at the top of the atmos-
phere is one of the most important climate model character-
istics, as errors in CRF frequently lead to an incorrect sur-
face temperature. Annual mean shortwave, longwave and
net CRF for the model and CERES v.2.8 (Loeb etal. 2009)
are shown in Fig.4 (time mean over 2000–2005 is used for
model and CERES data). The model underestimates the
absolute value of shortwave forcing in the subtropics and
midlatitudes by approximately 10–20W m−2. In the sub-
tropical regions of marine stratocumulus model error is
even higher. Total shortwave forcing in the tropics is not far
from the observed one, but one can see regional errors due
to the model shift of the intertropical convergence zone.
In the high latitudes model errors in shortwave CRF are
small. The model underestimates longwave CRF in the sub-
tropics but overestimates it in the high latitudes. Errors in
longwave CRF in the tropics tend to partially compensate
errors in shortwave CRF. Both errors have positive sign
near 60S leading to warm bias in the surface temperature
here. As a result, we have some underestimation of the net
CRF absolute value at almost all latitudes except the trop-
ics. Additional experiments with tuned conversion of cloud
water (ice) to precipitation (for upper cloudiness) showed
that model bias in the net CRF could be reduced, but that
the RMS bias for the surface temperature will increase in
this case.
Fig. 7 RMS of monthly mean
zonal mean zonal wind, ms−1,
in December–February in the
model (top) and ERA-Interim
Author's personal copy
E.M.Volodin et al.
1 3
The model zonal and annual mean temperature (Fig.5)
exhibits 4–6 K negative bias near the polar tropopause,
which is usual for many climate models. Positive bias
near the tropical tropopause is about 1–2K. It is smaller
than the one in the INMCM4, where the positive bias was
equal to 4–5 K. The reason for this improvement is the
adjustment of the deep convection parameterization. In
the troposphere, the magnitude of the temperature bias
is about or below 2°, except mountain areas where tem-
perature extrapolation at low pressure levels is required.
The model bias for zonal and annual mean zonal wind is
about or less than 4 m s−1 in the troposphere and never
exceeds 8m s−1 in the stratosphere, which seems satisfac-
tory. Analysis of the annual cycle of zonal wind at 10hPa
and 60N (Fig.6) shows that model tends to overestimate
zonal wind by 5–10m/s in February–August and underes-
timate it in October-December, that leads to small annual
mean error. In the Southern stratosphere at 60S positive
model bias in zonal wind is mostly pronounced in Octo-
ber–December, when model wind is 10–15 m/s higher
than in the reanalysis. The reason for this error is underes-
timated meridional heat flux induced by planetary waves
propagating upward.
Several important features of the stratosphere dynamics
could be described in terms of monthly mean zonal wind
RMS (see Fig.7). In the Northern Hemisphere the strong-
est variability of the wind takes place in December–Febru-
ary so the RMS for the winter season is shown. The RMS
maximum near the Equator is about 18m s−1 in the ERA-
Interim, and about 18m s−1 in the model. This maximum is
a manifestation of the equatorial QBO. To consider model
QBO in details, on Fig.8 we represent time series of zonal
equatorial wind in the stratosphere for years 1979–1988.
The QBO amplitude in the model is not far from the obser-
vations. One can see downward propagation of the QBO
signal from level of 5–100hPa both in the model and ERA
data. At levels of 1–5 hPa semiannual oscillation can be
seen, some of them initiate phase change of the QBO. The
QBO period is about 28 months in the observational data,
and about 29–30months in the model. The procedure for
the adjustment of QBO amplitude and period by tuning
parameters for nonorographic gravity wave drag, vertical
and horizontal diffusion can be found in Kulyamin etal.
Interaction of zonal mean flow and long planetary waves
induce high variance of zonal wind in winter in the strato-
sphere mid- and highlatitudes. Maximum RMS in the mid-
latidudes is about 14–16 m s−1 in ERA-Interim data, and
12–14m s−1 in the model data (Fig.7).
An important indicator of the winter stratospheric vari-
ability is the number of sudden stratospheric warmings
(SSW). Here we define the SSW event as one when the
zonal mean of zonal wind at 60N and 10 hPa is nega-
tive. In the model, we have about 20 SSWs per 30 years,
while in the observations we have 18 SSWs per 30 years
(Butchart et al. 2011). Figure 9 presents amplitudes of
stationary waves in geopotential height in the model and
ERA-Interim for the northern hemisphere winter. The
amplitude of wave number 1 is underestimated in the
model by 10–15%, while the amplitude of wave number
Fig. 8 Time series of zonal
mean zonal wind (ms−1) in
the stratosphere at Equator for
1979–1988 in the model (top)
and Era Interim (bottom)
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
2 in the model is close to the reanalysis data. In gen-
eral, in the INMCM5 the winter stratospheric variabil-
ity is reproduced reasonably well. The detailed study of
the INMCM5 stratospheric dynamics including analysis
of the SSWs, its influence on the lower stratosphere and
troposphere can be seen in Vargin and Volodin (2016).
Fig. 9 Amplitude of the wave number 1 (left) and the wave number 2 (right) in geopotential height (m) in December–February in the model
data (top) and ERA-Interim (bottom)
Author's personal copy
E.M.Volodin et al.
1 3
3.2 Ocean
Let us first consider the basic features of oceanic mean
state. Figure10 presents model bias in the surface salinity
and the mean dynamical sea surface height with respect to
the World Ocean Atlas 2009 (WOA09, Antonov etal. 2010)
and Rio and Hernandez (2004). On average, the model pro-
duces negative bias in surface salinity, but it is not as high
as in the INMCM4. A possible reason for this improvement
is the increase in the oceanic model resolution by the fac-
tor of two in both horizontal directions and new represen-
tation of advection by large oceanic eddies. In the Arctic,
the INMCM5 has a positive bias in salinity up to 1–5 PSU.
It seems that polar river runoff is underestimated in the
model, and positive bias can be attributed to strong vertical
mixing. RMS error of the surface salinity in the INMCM5
is much better than that in the INMCM4 (0.78 PSU com-
pared to 1.20 PSU). A study by Landerer et al. (2014)
shows large bias in sea surface height in the INMCM4
(RMS of the model bias is 0.36m). In the INMCM5, RMS
model bias is reduced to 0.19m. Nevertheless, one can see
some underestimation of sea surface height in the Southern
Ocean and its overestimation by 0.1–0.3 m in the Indian
Ocean and partially in the Pacific. These biases cannot be
attributed to temperature bias because surface temperature
and temperature in upper 700m in the Southern Ocean (see
Fig. 10 Annual mean model
bias in sea surface salinity
(PSU) with respect to WOA09
(Antonov etal. 2010) data (top)
and sea surface height (m) with
respect to (Rio and Hernandez
2004) (bottom)
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
below) in the model are not far from the observed (with
positive biases), while in the Indian Ocean they are lower
than the observed ones. Further analysis of the sea surface
height characteristics (including its natural variability) for
different versions of the INM climate models with differ-
ent spatial resolutions in the ocean can be found in Iakovlev
etal. (2016).
The model annual mean temperature and salinity bias
[with respect to WOA09 (Antonov et al. 2010)] in the
upper 700m layer are presented in Fig.11. Generally, one
can see cold bias in the tropical regions, and warm bias in
the southern midlatitudes, the north-west Atlantic and the
north-west Pacific. In the regions with positive tempera-
ture bias, salinity bias is also positive, and vice versa. The
exception is the Arctic Ocean, where one can see positive
salinity and negative temperature errors. At least part of the
temperature biases can be attributed to the shortcomings
of the atmospheric model. Small values of CRF over the
Southern ocean lead to a positive bias in the SST and tem-
perature of the upper oceanic layers. We suppose that pos-
sible reason of positive bias in the Arctic salinity is a strong
vertical mixing in the upper layer, but additional model
runs are required to prove this hypothesis.
The model biases in potential temperature and salinity
averaged over longitude with respect to WOA09 (Antonov
etal. 2010) are shown in Fig.12. Positive bias in the South-
ern Ocean penetrates from the surface downward for up to
300m, while negative bias in the tropics can be seen even
in the 100–1000 m layer. Nevertheless, zonal mean tem-
perature error at any level from the surface to the bottom
Fig. 11 Annual mean model
bias in mean temperature, C
(top), and salinity, PSU (bot-
tom), in the 0–700m ocean
layer with respect to WOA09
(Antonov etal. 2010). Shading
represents bias, contours show
observed state
Author's personal copy
E.M.Volodin et al.
1 3
is small. This was not the case for the INMCM4, where
one could see negative temperature bias up to 2–3K from
1.5 km to the bottom nearly al all latitudes, and 2–3 K
positive bias at levels of 700–1000m. The reason for this
improvement is the introduction of a higher background
coefficient for vertical diffusion at high depth (3000 m
and higher) than at intermediate depth (300–500m). Posi-
tive temperature bias at 45–65N at all depths could prob-
ably be explained by shortcomings in the representation of
deep convection [similar errors can be seen for most of the
CMIP5 models (Flato etal. 2013, their Fig.9.13)]. Another
feature common for many present day climate models
(and for the INMCM5 as well) is negative bias in southern
tropical ocean salinity from the surface to 500m. It can be
explained by overestimation of precipitation at the southern
branch of the Inter Tropical Convergence zone.
Meridional heat flux in the ocean (Fig. 13) is not far
from available estimates (Trenberth and Caron 2001). It
looks similar to the one for the INMCM4, but maximum
of northward transport in the Atlantic in the INMCM5
is about 0.1–0.2 × 1015 W higher than the one in the
INMCM4, probably, because of the increased horizontal
resolution in the oceanic block.
Sea ice area is an important parameter of the model cry-
osphere. Figure14 shows the annual cycle of sea ice area
in the Arctic and Antarctic. Data by Hurrell et al. (2008)
for 1979–2005 were chosen as the observations. In the Arc-
tic, the model sea ice area is just slightly overestimated.
Fig. 12 Zonal and annual mean
model bias in potential tem-
perature, C (top), and salinity,
PSU (bottom), with respect to
WOA09 (Antonov etal. 2010).
Shading represents bias, con-
tours represent observed state
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
Overestimation of the Arctic sea ice area is connected
with negative bias in the surface temperature. In the same
time, connection of the sea ice area error with the positive
salinity bias is not evident because ice formation is almost
compensated by ice melting, and the total salinity source
for these pair of processes is not large. The amplitude and
phase of the sea ice annual cycle are reproduced correctly
by the model. In the Antarctic, sea ice area is underesti-
mated by a factor of 1.5 in all seasons, apparently due to
the positive temperature bias. Note that the correct simula-
tion of sea ice area dynamics in both hemispheres simulta-
neously is a difficult task for climate modeling.
The El Niño is one of the most important phenomena of
interannual variability in atmosphere and ocean. The RMS
deviation for the monthly mean surface temperature in the
tropical Pacific for the model and the ERSST v4 (Huang
etal. 2015) data are shown in Fig. 15. RMS values in the
model are underestimated by a factor of 1.2–1.5, but the
location of variance maximum is reproduced correctly. In
the INMCM4, the surface temperature variability was also
underestimated but the El Niño tended to propagate too
westward in the west Pacific while the RMS maximum near
America was absent. The improvements could probably be
attributed to the horizontal resolution increase in the oce-
anic model. However, the reason for the underestimated
El Niño amplitude is unclear and requires further studies.
The analysis of the model time series of the SST anomalies
shows that the El Niño event frequency is approximately
the same in the model and data, but the model El Niños
happen too regularly. Atmospheric response to the El Niño
events is also underestimated in the model by a factor of
1.5 with respect to the reanalysis data. Time spectra of the
model and observed SST (ERSST v4 data) in NINO3,4
region (Fig.16) show that in the observations there is spec-
tral peak at 50 months attributed to the El-Nino. In the
model, two peaks associated with El-Nino can be seen at
50 and 80 months but they are weaker by the factor of 1.5-2
compared with the data.
3.3 Carbon cycle
In this section we present some integral characteristics of
model carbon cycle module. Gross primary production
(GPP) of the land vegetation in the model for 1986–2005
is 155 GtC/yr, while GPP for CMIP5 models lies in the
interval from 105 to 178 GtC/yr (Anav etal. 2013). Estima-
tions of this value from the observations gives values from
114 GtC/yr (Mao et al. 2012) to 150–175 GtC/yr (Welp
etal. 2011). Annual cycle of GPP for the global domain,
the tropics, the northern and southern extratropics can be
seen at Fig. 17. The estimate of observed annual cycle is
given by Jung etal. (2009), where global mean value is 120
GtC/yr. One can see that model GPP is higher than that of
Jung etal. (2009) in all subdomains however the seasonal
cycle is reproduced reasonably well for all domains except
the tropics. Note that the GPP seasonal cycle in the tropics
is poorly reproduced by many CMIP5 models (see Fig.9 in
Anav etal. 2013).
Global vegetation carbon amount in the model for
1986–2005 is 629 GtC while it is 522 ± 162 GtC for CMIP5
models (Anav et al. 2013). Global soil carbon amount
is 1781 GtC in the model and 1502 ± 798 GtC in CMIP5
models. Estimation of observed values of vegetation and
soil carbon are 550 GtC and 1340 GtC (Todd-Brown etal.
Annual cycle of the net carbon flux from the atmos-
phere (to vegetation, soil and ocean) can be seen in Fig.18.
Estimation of observed values is from Jung etal. (2009).
Model underestimates the amplitude of the annual cycle of
net carbon flux in global domain and Northern extratropics
by the factor of 1.5–2. Also, the maximum uptake in the
model is 1month earlier than in the Jung etal. (2009) data.
Probable reason for this discrepancy is different behavior of
plant and soil respiration in the model and reality. Similar
large deviations from the observed data are also common
for many CMIP5 models (see Fig.7 in Anav etal. 2013).
In general, model reproduces basic features of carbon
cycle reasonably well. Other details of the model carbon
cycle including trend of carbon uptake in historical run will
be a subject of specific paper.
Fig. 13 Meridional heat transport, 1015 W, in the Global Ocean
(solid line), Atlantic (dashed line) and Indo-Pacific (dotted line) in the
Author's personal copy
E.M.Volodin et al.
1 3
4 Conclusion
Based on the CMIP5 model INMCM4 the next version of
the Institute of Numerical Mathematics RAS climate model
was developed (INMCM5). The most important changes
include new parameterizations of large scale condensa-
tion (cloud fraction and cloud water are now the prognostic
variables), and increased vertical resolution in the atmos-
phere (73 vertical levels instead of 21, top model level
raised from 30 to 60km). In the oceanic block, horizontal
resolution was increased by a factor of 2 in both directions.
The climate model was supplemented by the aerosol block.
The model got a new parallel code with improved computa-
tional efficiency and scalability.
With the new version of climate model we performed a
test model run (80years) to simulate the present-day Earth
climate. The model mean state was compared with the
available datasets. The structures of the surface tempera-
ture and precipitation biases in the INMCM5 are typical for
the present climate models. Nevertheless, the RMS error
in surface temperature, precipitation as well as zonal mean
temperature and zonal wind are reduced in the INMCM5
Fig. 14 Annual cycle of sea ice
area, 106 km2, in the Northern
(top) and Southern (bottom)
Hemisphere in the model (red)
and Hurrell etal. (2008) (black)
for 1979–2005
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
Fig. 15 RMS of monthly
mean surface temperature (K)
in tropical Pacific for ERSST
v4 (Huang etal. 2015) data
(top) and model (bottom) for
Fig. 16 Time spectrum of
the sea surface temperature
(K) in Nino 3, 4 region for
1865–2014 in the model (red)
and ERSSTv4 (black)
Author's personal copy
E.M.Volodin et al.
1 3
with respect to its previous version, the INMCM4. The
model is capable of reproducing equatorial stratospheric
QBO and SSWs.
The model biases for the sea surface height and surface
salinity are reduced in the new version as well, probably
due to increasing spatial resolution in the oceanic block.
Bias in ocean potential temperature at depths below 700m
in the INMCM5 is also reduced with respect to the one in
the INMCM4. This is likely because of the tuning back-
ground vertical diffusion coefficient. Model sea ice area is
reproduced well enough in the Arctic, but is underestimated
in the Antarctic (as a result of the overestimated surface
temperature). RMS error in the surface salinity is reduced
almost everywhere compared to the previous model except
the Arctic (where the positive bias becomes larger).
As a final remark one can conclude that the INMCM5 is
substantially better in almost all aspects than its previous
version and we plan to use this model as a core component
for the coming CMIP6 experiment.
Fig. 17 Annual cycle of GPP
(GtC/month) for global domain
(top, left), Northern hemisphere
20N–90N (top, right), Southern
Hemisphere 90S–20S (bot-
tom, left) and tropics 20S–20N
(bottom, right) in the model
(red) and Stephens etal. (2007)
(black) for 1986–2005
Author's personal copy
Simulation ofthepresent-day climate withtheclimate model INMCM5
1 3
Acknowledgements The study was performed at the Institute of
Numerical Mathematics of the Russian Academy of Sciences and
supported by the Russian Science Foundation, grant 14-17-00126
(model development) and Russian Foundation for Basic Research,
grant 16-55-76004 ERA.NET RUS (numerical experiments). Climate
model runs were produced at the supercomputer MVS10P of the Joint
Supercomputer Center of the Russian Academy of Sciences.
Adler RF, Huffman GJ, Chang A, Ferraro R, Xie P, Janowiak J,
Rudolf B, Schneider U, Curtis S, Bolvin D, Gruber A, Susskind
J, Arkin P (2003) The Version 2 Global Precipitation Climatol-
ogy Project (GPCP) Monthly Precipitation Analysis (1979-Pre-
sent). J Hydrometeor 4:1147–1167
Alekseev VA, Volodin EM, Galin VYa, Dymnikov VP, Lykossov
VN (1998) Simulation of present day climate with atmospheric
model of INM RAS. INM Reprint, p 198, (available by request)
Anav A, Friedlingstein P, Kidston M, Bopp L, Ciais P, Cox P, Jones
C, Jung M, Myneni R, Zhu Z (2013) Evaluating the land and
ocean components of the global carbon cycle in the CMIP5
earth system models. J Climate 26:6801–6843. doi:10.1175/
Antonov JI etal (2010) World ocean atlas 2009, Vol. 2: Salinity. [S.
Levitus (eds.)]. NOAA Atlas NESDIS 69, U.S. Gov. Printing
Office, Washington, D.C., pp 184
Asselin R (1972) Frequency filter for time integrations. Mon Wea Rev
Fig. 18 Annual cycle of net
CO2 flux (GtC/month) for
global domain (top, left),
Northern hemisphere 20N–90N
(top, right), Southern Hemi-
sphere 90S–20S (bottom, left)
and tropics 20S–20N (bottom,
right) in the model (red) and
Stephens etal. (2007) (black)
for 1986–2005
Author's personal copy
E.M.Volodin et al.
1 3
Betts AK (1986) A new convective adjustment scheme. Part 1. Obser-
vational and theoretical basis. Quart J Roy Met Soc 112:677–691
Butchart N et al (2011) Multimodel climate and variability of the
stratosphere. J Geophys Res 116:D05102. doi:10.1029/201
Dee DP etal (2011) The ERA-Interim reanalysis: configuration and
performance of the data assimilation system. Q J R Meteorol Soc
Flato G, Marotzke J, Abiodun B, Braconnot P, Chou SC, Collins W,
Cox P, Driouech F, Emori S, Eyring V, Forest C, Gleckler P,
Guilyardi E, Jakob C, Kattsov V, Reason C, Rummukainen M
(2013) Evaluation of climate models. In: Climate change 2013:
the physical science basis. contribution of working group i to the
fifth assessment report of the intergovernmental panel on climate
change [Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK,
Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds)]. Cam-
bridge University Press, Cambridge
Galin VYa (1998) Parametrization of radiative processes in the DNM
atmospheric model. Izv Atmos Ocean Phy 34:339–347
Galin VYa, Volodin EM, Smyshlyaev SP (2003) Atmospheric general
circulation model with ozone dynamics. Rus Meteorol Hydrol
Hines CO (1997) Doppler spread parameterization of gravity wave
momentum deposition in the middle atmosphere 2. Broad and
quasimonochromatic spectra, and implementation. J Atm Sol
Terr Phys 59:387–400
Huang B, Banzon VF, Freeman E, Lawrimore J, Liu W, Peterson
TC, Smith TM, Thorne PW, Woodruff SD, Zhang H-M (2015)
Extended reconstructed sea surface temperature version 4
(ERSST.v4): part I. upgrades and intercomparisons. J Climate
Hurrell JW, Hack JJ, Shea D, Caron JM, Rosinski J (2008) A new sea
surface temperature and sea ice boundary dataset for the commu-
nity atmosphere model. J Climate 21:5145–5153
Iakovlev NG (2009) Reproduction of the large scale state of water
and sea ice in the Arctic Ocean in 1948–2002. Part 1. Numeri-
cal model. Izv Atmos Ocean Phy 45:357–371. doi:10.1134/
Iakovlev NG, Volodin EM, Gritsun AS (2016) Simulation of the spa-
tiotemporal variability of the World Ocean sea surface height by
theINM climate models. Izv Atmos Ocean Phy 52(4):376–385.
Jung M, Reichstein M, Bondeau A (2009) Towards global empirical
upscaling of FLUXNET eddy covariance observations: valida-
tion of a model tree ensemble approach using a biosphere model.
Biogeosciences 6:2001
Kessler E (1969) On the distribution and continuity of water sub-
stance in atmospheric circulations. Meteor. Monogr. 10. N32,
Amer Meteor Soc p 84
Kulyamin DV, Volodin EM, Dymnikov VP (2009) Simulation of the
quasi-biannual oscillation in the zonal wind in the equatorial
stratosphere: Part II. Atmospheric general circulation models.
Izv Atmos Ocean Phy 45:37–54
Landerer FW, Gleckler PJ, Lee T (2014) Evaluation of CMIP5
dynamic sea surface height multi-model simulations against
satellite observations. Clim Dyn 43:1271–1283. doi:10.1007/
Loeb NG etal (2009) Toward optimal closure of the Earth’s top-of-
atmosphere radiation budget. J Climate 22:748–766
Mao J, Thornton P, Shi X, Zhao M, Post W (2012) Remote sensing
evaluation of CLM4 GPP for the period 2000 to 2009. J Climate
Mareev EA, Volodin EM (2014) Variations of the global electric cir-
cuit and the ionospheric potential in a general circulation model.
Geophys Res Lett 41:9009–9016. doi:10.1002/2014GL062352
Mueller B, Seneviratne SI (2014) Systematic land climate and
evapotranspiration biases in CMIP5 simulations. Geophys Res
Lett 41:128–134. doi:10.1002/2013GL058055
Palmer TN, Shutts GJ, Swinbank R (1986) Alleviation of a systematic
westerly bias in general circulation and numerical weather pre-
diction models through an orographic gravity wave drag param-
eterization. Quart J Roy Met Soc 112:1001–1031
Rio M-H, Hernandez F (2004) A mean dynamical topography com-
puted over the world ocean from altimetry, in-situ measurements
and a geoid model. J Geophys Res 109:C12032. doi:10.1029/20
Stephens BB etal (2007) Weak northern and strong tropical land car-
bon uptake from vertical profiles of atmospheric CO2. Science
Taylor KE, Stouffer RJ, Meehl GA (2012) An overview of CMIP5
and the experiment design. Bull Am Met Soc 93:485–498
Terekhov KM, Volodin EM, Gusev AV (2011) Methods and efficiency
estimation of parallel implementation of the σ-model of general
ocean circulation. Russ J Num Anal Math Model 26(2):189–208
Tiedtke M (1993) Representation of clouds in large-scale models.
Mon Weather Rev 121:3040–3061
Todd-Brown KEO, Randerson JT, Post WM, Hoffman FM, Tarnocai
C, Schuur EA, Allison SD (2012) Causes of variation in soil car-
bon predictions from CMIP5 Earth system models and compari-
son with observations. Biogeosci Discuss 9:14,437–14,473
Trenberth KE, Caron JM (2001) Estimates of meridional atmosphere
and ocean heat transports. J Clim 14:3433–3443
Vargin PN, Volodin EM (2016) Analysis of the reproduction of
dynamic processes in the stratosphere using the climate model
of the Institute of Numerical Mathematics, Russian Academy of
Sciences. Izv Atmos Ocean Phy 52:1–15
Volodin EM (2007) Atmosphere—ocean general circulation model
with carbon cycle. Izv Atmos Ocean Phy 43:266–280
Volodin EM (2008) Methane cycle in the INM RAS climate model.
Izv Atmos Ocean Phy 44:153–159
Volodin EM (2013) The mechanism of multidecadal variability in the
Arctic and North Atlantic in climate model INMCM4. Environ
Res Lett8:035038. doi:10.1088/1748-9326/8/3/035038
Volodin EM, Kostrykin SV (2016) The Aerosol module in the INM
RAS climate model. Rus Meteorol Hydrol 41(8):519–528
Volodin EM, Lykosov VN (1998) Parametrization of neat and mois-
ture transfer in the soil-vegetation system for use in atmospheric
general circulation models: 1. Formulation and simulations based
on local observational data. Izv Atmos Ocean Phy 34:405–416
Volodin EM, Dianskii NA, Gusev AV (2010) Simulating present
day climate with the INMCM4.0 coupled model of the atmos-
pheric and oceanic general circulations. Izv Atmos Ocean Phy
Volodin EM, Diansky NA, Gusev AV (2013) Simulation and predic-
tion of climate changes in the 19th to 21st centuries with the
Institute of Numerical Mathematics, Russian Academy of Sci-
ences, model of Earth climate system. Izv Atmos Ocean Phy
Welp LR etal (2011) Interannual variability in the oxygen isotopes
of atmospheric CO2 driven by El Ni ~ no. Nature 477:579–582
Zalesny VB, Marchuk GI, Agoshkov VI, Bagno AV, Gusev AV,
Diansky NA, Moshonkin SN, Tamsalu R, Volodin EM (2010)
Numerical simulation of large-scale ocean circulation based on
the multicomponent splitting method. Russ J Num Anal Math
Model 25(6) 581–609
Author's personal copy
... Out of the thirteen models, two (INM-CM4-8, INM-CM5-0) are noticeably poorer than the rest. This is probably due to their outdated convection scheme, which appears to have not been materially updated since the 1980s (Volodin et al. 2017). Seven other models can be compared to their earlier versions contributed to CMIP3, as analyzed by Lin et al. (2006) (CESM2, CNRM-CM6-1, CNRM-ESM2-1, GFDL-CM4, MIROC6, MPI-ESM1-2-HR and MPI-ESM1-2-LR). ...
Full-text available
The simulation of the Madden–Julian Oscillation (MJO) and convectively coupled equatorial waves (CCEWs) is considered in 13 state-of-the-art models from phase 6 of the Coupled Model Intercomparison Project (CMIP6). We use frequency–wavenumber power spectra of the models and observations for Outgoing Longwave Radiation (OLR) and zonal winds at 250 hPa (U250), and consider the historical simulations and end of twenty-first century projections for the SSP245 and SSP585 scenarios. The models simulate a spectrum quantitatively resembling that observed, though systematic biases exist. MJO and Kelvin waves (KW) are mostly underestimated, while equatorial Rossby waves (ER) are overestimated. Most models project a future increase in power spectra for the MJO, while nearly all project a robust increase for KW and weaker power values for most other wavenumber–frequency combinations, including higher wavenumber ER. In addition to strengthening, KW also shift toward higher phase speeds (or equivalent depths). Models with a more realistic MJO in their control climate tend to simulate a stronger future intensification.
... Overall, our results indicate that the occurrence of higher plants mark a tipping point for mineral weathering and soil development of Maritime Antarctica. (Volodin et al., 2017), participating in the CMIP (Coupled Model Intercomparison Project) series of climate simulations. The basis of INM RAS-MSU LSM is a soil model computing vertical distribution of temperature, liquid water, ice and water vapor content in each land grid cell of the ESM (Volodin and Lykosov, 1998). ...
Full-text available
Focus Siberian Permafrost – Terrestrial Cryosphere and Climate Change International Online Symposium Institute of Soil Science, Universität Hamburg, 24 – 25 March 2021
Full-text available
The “warm-humid” climate change across the Tibetan Plateau (TP) has promoted grassland growth and an overall greening trend has been observed by remote sensing products. Many of the current generations of Earth System Models (ESMs) incorporate advanced process-based vegetation growth in the land surface module that can simulate vegetation growth, but the evaluation of their performance has not received much attention, especially over hot spots where projections of the future climate and vegetation growth are greatly needed. In this study, we compare the leaf area index (LAI) simulations of 35 ESMs that participated in CMIP6 to a remote-sensing-derived LAI product (GLASS LAI). The results show that about 40% of the models overestimated the Tibetan Plateau’s greening, 48% of the models underestimated the greening, and 11% of the models showed a declining LAI trend. The CMIP6 models generally produced poor simulations of the spatial distribution of LAI trend, and overestimated the LAI trend of alpine vegetation, grassland, and forest, but underestimated meadow and shrub. Compared with other vegetation types, simulations of the forest LAI trend were the worst, the declining trend in forest pixels on the TP was generally underestimated, and the greening of the meadow was underestimated as well. However, the greening of the grassland, was greatly overestimated. For the Tibetan Plateau’s averaged LAI, more than 70% of the models overestimated this during the growing seasons of 1981–2014. Similar to the forest LAI trend, the performance of the forest LAI simulation was the worst among the different vegetation types, and the forest LAI was underestimated as well.
Full-text available
Future climate projections and their uncertainties affect many aspects of the world, so reliable assessments are essential for policymakers who need to prepare mitigation measures in the context of climate change. In this study, we examined the projected future climate and estimated uncertainty for South Korea using results from the global climate model (GCM), updated from the sixth phase of the coupled model intercomparison project (CMIP6); we then compared the differences in outcome between the fifth and sixth phases of the CMIP (CMIP5 and CMIP6). Future projections were estimated as the averaged climatological mean (denoted as) for the four proposed hydrological indicators. Model uncertainty (UEMI) and stochastic uncertainty (USTO) were quantified as the range of ensembles of the climatological mean, while the emission uncertainty (UEMI) was estimated as the difference between the values of two emission scenarios. The following are the key findings of our study: (1) using an ensemble of multiple GCMs is recommended over using individual GCMs, and models in CMIP6 performed better for reproducing climate during the control period than models in the CMIP5; (2) the values in the CMIP6 increased for future periods, especially toward the end of this century, increasing mean temperature (meanTa) by approximately 5 °C, total precipitation (totPr), and daily maximum precipitation (maxDa) by about 20%, and these values were higher than those of the CMIP5; (3) the UGCM, USTO, and UEMI values increased for future periods in most of the indices; (4) the UGCM (for meanTa, totPr, and maxDa) and USTO (for totPr and maxDa) magnitudes in the CMIP6 were higher than those in the CMIP5, while the UEMI values between the two CMIPs were similar for all of the indices; (5) the UGCM was the major source of the largest uncertainty for meanTa, the USTO had a significant impact on future projections of totPr and maxDa, especially in the summer, and the UEMI became the dominant source of uncertainty for projecting the future meanTa, especially in the period farthest from the present. These results should provide useful information for studies that quantify future climate-induced hydrological impacts.
Full-text available
Regional hydroclimates constitute the interplay between climate, weather, and water resources at sub-continental scales. They continually evolve by responding to changes and perturbations in global climate, water cycle, and terrestrial surface and subsurface processes. This dissertation is on region-specific and process-based assessments of the atmosphere, cryosphere, and hydrosphere interactions to understand the mechanisms driving variability in hydroclimates. The overarching motivation is to quantify the availability of water within the Earth’s components and understand how water systems evolve in space and time. This research combines data analysis and modeling tools with fundamental physical laws to study two distinct systems – large inland lakes and mountain glaciers – both critical water resources for societies, economies, and ecosystems. In the first study, we discuss how large inland water bodies regulate the water cycle for three geographical regions: the African Great Lakes, the Laurentian Great Lakes, and Lake Baikal. We found that these lakes control regional micro- and meso-scale weather and climate, and different modeled lake representations can simulate markedly different coupled lake-atmosphere processes. The second study analyses the atmospheric moisture budget in the Laurentian Great Lakes region. We developed process-level understanding of the precipitation seasonality and established the role of lakes in inducing differences in the water cycle seasonality. These lakes are a source of moisture through evaporation and generate localized moisture flux convergence/divergence patterns. We further quantified the future changes in the budget using Coupled Model Intercomparison Project (CMIP6) data. There are common patterns of change in the mid-century (2041 – 2070) projections of climate variables, specifically, an increase in evapotranspiration throughout the year and intensification of winter/spring precipitation, indicating a shift in the precipitation seasonal cycle towards the colder months. The next chapter delves into land surface hydrology, specifically looking at the controls of variability in the terrestrial water budget using a high-resolution model for the Laurentian Great Lakes domain. This is a hydrologically heterogeneous region, with various regulators of the water budget at different timescales. At higher latitudes, snowpack and soil moisture are the principal drivers of variability, while at lower latitudes precipitation, evapotranspiration, and runoff are the dominant controls of budget variability. The subsequent work centers on mountain cryosphere to study the evolution of Karakoram glaciers using a numerical model. We explored the relation between climate, mass balance, and ice dynamics in driving glacier geometry (length, thickness, area, and volume) changes. We found that similar climate forcing can trigger radically different responses, even in neighboring glaciers, and changes in area and length do not always correspond to a similar change in the glacial volume. We also applied a new approach to calibrate ice dynamics parameters and introduced a novel scheme for dynamic spin-up to match geodetic mass balance observations while accounting for changing glacier area. In the final chapter, the dissertation lays the groundwork to simulate three-dimensional dynamics of mountain glaciers using the Community Ice Sheet Model to advance the field of glaciology as a coupled process within the larger Earth system modeling framework. Throughout this dissertation, we highlight the significance of understanding the governing processes modulating regional hydroclimates before assessing their future evolution, and the importance of effectively representing water/ice reservoirs in numerical models. This is critical to improve accuracy and reliability of future climate assessments using models which currently have limitations in simulating regional-scale climate.
Full-text available
We analyze the relationship between fine particulate matter (PM2.5) and meteorology in winter in the Indo‐Gangetic Plain (IGP). We find that the concentration of PM2.5 exhibits similar increase with decreasing surface wind speed in 15 out of 18 cities considered. Using this observed relationship, we estimate that the reduction of surface wind speed with increasing CO2 simulated by models participating in the Coupled Model Intercomparison Project Phase 6 will result in higher average wintertime PM2.5 concentrations (1% per degree K of global warming) and more frequent high‐pollution events. This observation‐based estimate is qualitatively consistent with the simulated response of black carbon to global warming inferred from the AerChemMIP ssp370SST and ssp370pdSST experiments. We hypothesize that a reduction in the frequency and intensity of western disturbances with increasing CO2 may contribute to the reduction in the surface wind in the IGP.
Full-text available
The May and June precipitation (known as early summer precipitation; ESP) is an important water resource for South China and Taiwan (SCTW). This study explores the similarities and differences between the Community Earth System Model Version 2‐Large Ensemble (CESM2‐LE) and the Coupled Model Intercomparison Project Phase 6 (CMIP6) multi‐model ensemble in simulating the characteristics of ESP over SCTW in the present day, and in projecting the related changes at the end of 21st century. For the present‐day simulation, our results showed that CESM2‐LE and CMIP6 outperformed each other for different examined features. CESM2‐LE was slightly better than CMIP6 in capturing the magnitude of ESP over SCTW, while CMIP6 was more capable of representing the interdecadal shifts in the occurrence timing of the ESP maximum. Both the CESM2‐LE and CMIP6 projections indicated that ESP will be enhanced and the phase of maximum ESP will be delayed, from peaking around mid‐June in the present to late June in the future. These changes can be attributed to an enhanced short‐wave trough over southwest China and a late intensification of the southwesterly wind ahead of the short‐wave trough, which help transport more moisture from the north of the South China Sea to SCTW, particularly in mid‐to‐late June. In addition, relative to CMIP6, CESM2‐LE showed less uncertainty in the projected increase in ESP and phase delay. This finding highlights that model diversity may play a more important role than internal variability in attributing the uncertainty of projected changes in ESP over SCTW.
Earth's forests harbor extensive biodiversity and are currently a major carbon sink. Forest conservation and restoration can help mitigate climate change; however, climate change could fundamentally imperil forests in many regions and undermine their ability to provide such mitigation. The extent of climate risks facing forests has not been synthesized globally nor have different approaches to quantifying forest climate risks been systematically compared. We combine outputs from multiple mechanistic and empirical approaches to modeling carbon, biodiversity, and disturbance risks to conduct a synthetic climate risk analysis for Earth's forests in the 21st century. Despite large uncertainty in most regions we find that some forests are consistently at higher risk, including southern boreal forests and those in western North America and parts of the Amazon.
Full-text available
The connection between the Atlantic meridional overturning circulation (AMOC) and the Atlantic multidecadal variability (AMV) is inspected in a suite of pre-industrial integrations from the 6th phase of the Coupled Model Inter-comparison Project (CMIP6), using a change-point detection method to identify different AMOC-AMV co-variability regimes. A key finding of this study is that models robustly simulate multi-decadal windows where the AMV and the AMOC are essentially uncorrelated. These regimes coexist with longer periods with relatively high correlation. Drops and recoveries of correlation are found to be often abrupt and confined in a temporal window of the order of 10 years. Phenomenological evidence suggests that the no-correlation regimes may be explained by drops in the variance of the AMOC: a less variable meridional heat transport leads to a suppressed co-variability of the AMV, leaving a larger role for non-AMOC drivers, consistent with a non-stationary AMOC-stationary noise interpretative framework.
Full-text available
The current generation of Earth system models exhibits large inter-model differences in the simulated climate of the Arctic and subarctic zone, with differences in model structure and parametrizations being one of the main sources of uncertainty. One particularly challenging aspect in modelling is the representation of terrestrial processes in permafrost-affected regions, which are often governed by spatial heterogeneity far below the resolution of the models' land surface components. Here, we use the MPI Earth System model to investigate how different plausible assumptions for the representation of the permafrost hydrology modulate the land-atmosphere interactions and how the resulting feedbacks affect not only the regional and global climate, but also our ability to predict whether the high latitudes will become wetter or drier in a warmer future. Focusing on two idealized setups that induce comparatively "wet" or "dry" conditions in regions that are presently affected by permafrost, we find that the parameter settings determine the direction of the 21st-century trend in the simulated soil water content and result in substantial differences in the land-atmosphere exchange of energy and moisture. The latter leads to differences in the simulated cloud cover and thus in the planetary energy uptake. The respective effects are so pronounced that uncertainties in the representation of the Arctic hydrological cycle can help to explain a large fraction of the inter-model spread in regional surface temperatures and precipitation. Furthermore, they affect a range of components of the Earth system as far to the south as the tropics. With both setups being similarly plausible, our findings highlight the need for more observational constraints on the permafrost hydrology to reduce the inter-model spread in Arctic climate projections.
Full-text available
The aerosol module is included into the INM RAS climate model. The module computes the evolution of main aerosols: sea salt, mineral dust, sulfate aerosol, and black and organic carbon. Aerosol surface fluxes, advection, gravitational sedimentation, surface absorption, and scavenging by precipitation are taken into account to compute aerosol concentration variations. Model aerosol distribution is used to compute radiation fluxes. The ten-year run of the climate model is performed. The climatology of model aerosol is considered. The aerosol mass, integral source values, optical thickness, and radiative forcing are presented. The results are compared with the data of other models and observations.
Full-text available
The monthly Extended Reconstructed Sea Surface Temperature (ERSST) dataset, available on global 2° × 2° grids, has been revised herein to version 4 (v4) from v3b. Major revisions include updated and substantially more complete input data from the International Comprehensive Ocean–Atmosphere Data Set (ICOADS) release 2.5; revised empirical orthogonal teleconnections (EOTs) and EOT acceptance criterion; updated sea surface temperature (SST) quality control procedures; revised SST anomaly (SSTA) evaluation methods; updated bias adjustments of ship SSTs using the Hadley Centre Nighttime Marine Air Temperature dataset version 2 (HadNMAT2); and buoy SST bias adjustment not previously made in v3b. Tests show that the impacts of the revisions to ship SST bias adjustment in ERSST.v4 are dominant among all revisions and updates. The effect is to make SST 0.1°–0.2°C cooler north of 30°S but 0.1°–0.2°C warmer south of 30°S in ERSST.v4 than in ERSST.v3b before 1940. In comparison with the Met Office SST product [the Hadley Centre Sea Surface Temperature dataset, version 3 (HadSST3)], the ship SST bias adjustment in ERSST.v4 is 0.1°–0.2°C cooler in the tropics but 0.1°–0.2°C warmer in the midlatitude oceans both before 1940 and from 1945 to 1970. Comparisons highlight differences in long-term SST trends and SSTA variations at decadal time scales among ERSST.v4, ERSST.v3b, HadSST3, and Centennial Observation-Based Estimates of SST version 2 (COBE-SST2), which is largely associated with the difference of bias adjustments in these SST products. The tests also show that, when compared with v3b, SSTAs in ERSST.v4 can substantially better represent the El Niño/La Niña behavior when observations are sparse before 1940. Comparisons indicate that SSTs in ERSST.v4 are as close to satellite-based observations as other similar SST analyses.
The objects of the American Meteorological Society are "the development and dissemination of knowledge of meteorology in all its phases and applications, and the advancement of its professional ideals." The organization of the Society took place in affiliation with the American Association for the Advancement of Science at Saint Louis, Missouri, December 29, 1919, and its incorporation, at Washington, D. C., January 21, 1920. The work of the Society is carried on by the Bulletin, the Journal, and Meteorological Monographs, by papers and discussions at meetings of the Society, through the offices of the Secretary and the Executive Secretary, and by correspondence. All of the Americas are represented in the membership of the Society as well as many foreign countries.
The stratospheric climate and variability from simulations of sixteen chemistryclimate models is evaluated. On average the polar night jet is well reproduced though its variability is less well reproduced with a large spread between models. Polar temperature biases are less than 5 K except in the Southern Hemisphere (SH) lower stratosphere in spring. The accumulated area of low temperatures responsible for polar stratospheric cloud formation is accurately reproduced for the Antarctic but underestimated for the Arctic. The shape and position of the polar vortex is well simulated, as is the tropical upwelling in the lower stratosphere. There is a wide model spread in the frequency of major sudden stratospheric warnings (SSWs), late biases in the breakup of the SH vortex, and a weak annual cycle in the zonal wind in the tropical upper stratosphere. Quantitatively, �metrics� indicate a wide spread in model performance for most diagnostics with systematic biases in many, and poorer performance in the SH than in the Northern Hemisphere (NH). Correlations were found in the SH between errors in the final warming, polar temperatures, the leading mode of variability, and jet strength, and in the NH between errors in polar temperatures, frequency of major SSWs, and jet strength. Models with a stronger QBO have stronger tropical upwelling and a colder NH vortex. Both the qualitative and quantitative analysis indicate a number of common and long�standing model problems, particularly related to the simulation of the SH and stratospheric variability.
The results of simulations of the World Ocean sea surface hight (SSH) in by various versions of the Climate Model of the Institute of Numerical Mathematics, Russian Academy of Sciences, are compared with the CNES-CLS09 fields of the mean dynamic topography (deviation of the ocean level from the geoid). Three models with different ocean blocks are considered which slightly differ in numerical schemes and have various horizontal spatial resolution, i.e., the INMCM4 model, which participated in the Climate Model Intercomparison Project (CMIP Phase 5, resolution of 1° × 1/2°); the INMCM5 model, which participates in the next project, CMIP6 (resolution of 1/2° × 1/4°); and the advanced INMCM-ER eddy-resolving model (resolution of 1/6° × 1/8°). It is shown that an increase in the spatial resolution improves the reproduction of ocean currents (with Agulhas and Kuroshio currents as examples) and their variability. A probable cause of relatively high errors in the reproduction of the SSH of Southern and Indian oceans is discussed.
The radiative code of the atmospheric model of the Institute of Numerical Mathematics (IVM), Russian Academy of Sciences, (DNM model) is described. The code uses spectral transmission functions and the δ-Eddington approximation to take into account the absorption and scattering of radiation in the atmosphere due to atmospheric gases, aerosols, and clouds. A very simple regularization procedure in combination with the nonmonotonic factorization method is used to find a stable solution to the ill-conditioned system of δ-Eddington equations. Computation algorithms are presented, and the results obtained are compared to both the data of benchmark line-by-line (LBL) calculations and the model data of ICRCCM international radiative programs. It is found that the DNM model yields a high accuracy of computing the thermal and solar radiation.
The reproduction of dynamic processes in the stratosphere at extratropical latitudes is considered in calculations of the atmospheric module of the global climate model of the Institute of Numerical Mathe� matics, Russian Academy of Sciences, with an upper boundary of 0.2 hPa (~60 km) for the period from 1979 to 2008 in comparison with the data observational. Changes in temperature, zonal wind, activity of planetary waves, heat fluxes in the lower stratosphere, and sudden stratospheric warmings with the displacement and splitting of the polar vortex, as well as the distribution of associated circulation anomalies in the troposphere, are analyzed.
The experience of inclusion of ozone dynamics and chemistry into the atmospheric model of INM RAS is presented. To reduce systematic errors in the ozone seasonal cycle, the form of differential equations and finite-difference scheme in the dynamic block was changed. The influence of orographic gravity wave drag on ozone distribution is shown.