Mao-Lin Shen

Mao-Lin Shen

Auto-generated profile, based on publicly available information

Mao-Lin Shen

Bjerknes Centre for Climate Research Geophysical Institute University of Bergen Bergen Norway

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 December 2016 in Tellus
Verified
Journal Article
Open access
PDF file
FCFrançois Counillon
François Counillon
NKNoel Keenlyside
Noel Keenlyside
IBIngo Bethke
Ingo Bethke
YWYiguo Wang
Yiguo Wang
SBSebastien Billeau
Sebastien Billeau
MBMats Bentsen
Mats Bentsen
Flow-dependent assimilation of sea surface temperature in isopycnal coordinates with the Norwegian Climate Prediction Model

9 January 2016 in Geophysical Research Letters
Verified
Journal Article
Open access
PDF file
MSMao‐Lin Shen
Mao‐Lin Shen
NKNoel Keenlyside
Noel Keenlyside
FSFrank Selten
Frank Selten
WWWim Wiegerinck
Wim Wiegerinck
GDGregory S. Duane
Gregory S. Duane
We construct an interactive ensemble of two different climate models to improve simulation of key aspects of tropical Pacific climate. Our so‐called supermodel is based on two atmospheric general circulation models (AGCMs) coupled to a single ocean GCM, which is driven by a weighted average of the air‐sea fluxes. Optimal weights are determined using a machine learning algorithm to minimize sea surface temperature errors over the tropical Pacific. This coupling strategy synchronizes atmospheric variability in the two AGCMs over the equatorial Pacific, where it improves the representation of ocean‐atmosphere interaction and the climate state. In particular, the common double Intertropical Convergence Zone error is suppressed, and the positive Bjerknes feedback improves substantially to match observations well, and the negative heat flux feedback is also much improved. This study supports the concept of supermodeling as a promising multimodel ensemble strategy to improve weather and climate predictions.

18 September 2019 in Journal of Geophysical Research: Atmospheres
Verified
Journal Article
Open access
PDF file
FLF. Li
F. Li
YOYvan Orsolini
Yvan Orsolini
NKN. Keenlyside
N. Keenlyside
MSM.‐L. Shen
M.‐L. Shen
FCF. Counillon
F. Counillon
YWY. G. Wang
Y. G. Wang
Snow initialization has been previously investigated as a potential source of predictability at the subseasonal‐to‐seasonal (S2S) timescale in winter and spring, through its local radiative, thermodynamical, and hydrological feedbacks. However, previous studies were conducted with low‐top models over short periods only. Furthermore, the potential role of the land surface‐stratosphere connection upon the S2S predictability had remained unclear. To this end, we have carried out twin 30‐member ensembles of 2‐month (November and December) retrospective forecasts over the period 1985–2016, with either realistic or degraded snow initialization. A high‐top version of the Norwegian Climate Prediction Model is used, based on the Whole Atmosphere Community Climate Model, to insure improved coupling with the stratosphere. In a composite difference of high versus low initial Eurasian snow, the surface temperature is strongly impacted by the presence of snow, and wave activity fluxes into the stratosphere are enhanced at a 1‐month lag, leading to a weakened polar vortex. Focusing further on 7 years characterized by a strongly negative phase of the Arctic Oscillation, we find a weak snow feedback contributing to the maintenance of the negative Arctic Oscillation. By comparing the twin forecasts, we extracted the predictive skill increment due to realistic snow initialization. The prediction of snow itself is greatly improved, and there is increased skill in surface temperature over snow‐covered land in the first 10 days, and localized skill increments in the mid‐latitude transition regions on the southern flanks of the snow‐covered land areas, at lead times longer than 30 days.

1 May 2011 in Journal of Marine Systems
Verified
Journal Article
Closed access
MSMao-Lin Shen
Mao-Lin Shen
YU-HENG TSENGYU-HENG TSENG
YU-HENG TSENG
SJSen Jan
Sen Jan

9 March 2023 in Journal of Advances in Modeling Earth Systems
Verified
Journal Article
Open access
FCFrançois Counillon
François Counillon
NKNoel Keenlyside
Noel Keenlyside
SWShuo Wang
Shuo Wang
MDMarion Devilliers
Marion Devilliers
AGAlok Gupta
Alok Gupta
MSMao‐Lin Shen
Mao‐Lin Shen
A supermodel connects different models interactively so that their systematic errors compensate and achieve a model with superior performance. It differs from the standard non‐interactive multi‐model ensembles (NI), which combines model outputs a‐posteriori. Supermodels with Earth system models (ESMs) has not been developed because it is technically challenging to combine models with different state space. Here, we formulate the first supermodel framework for ESMs and use data assimilation to synchronise models. The ocean of three ESMs is synchronised every month by assimilating pseudo sea surface temperature (SST) observations generated by them on a common grid to handle discrepancies in grid and resolution. We compare the performance of two supermodel approaches to that of the NI. In the first (EW), the models are connected to the equal‐weight multi‐model mean, while in the second (SINGLE), they are connected to a single model. Both versions achieve synchronisation in the ocean and in the atmosphere, where the ocean drives the variability. The time variability of the supermodel multi‐model mean SST is reduced compared to observations, most where synchronisation is not achieved and is lower‐bounded by NI. The damping is larger in EW, for which variability in the individual models is also damped. Hence, under partial synchronisation, the unsynchronized variability gets damped in the multi‐model average pseudo‐observations, causing a deflation during the assimilation. The SST bias in individual models of EW is reduced compared to that of NI, and so is its multi‐model mean in the synchronised regions. A trained supermodel remains to be tested.

1 December 2017 in Chaos: An Interdisciplinary Journal of Nonlinear Science
Verified
Journal Article
Closed access
MSMao-Lin Shen
Mao-Lin Shen
NKNoel Keenlyside
Noel Keenlyside
BBBhuwan C. Bhatt
Bhuwan C. Bhatt
GDGregory S. Duane
Gregory S. Duane
The supermodel strategy interactively combines several models to outperform the individual models comprising it. A key advantage of the approach is that nonlinear improvements can be achieved, in contrast to the linear weighted combination of individual unconnected models. This property is found in a climate supermodel constructed by coupling two versions of an atmospheric model differing only in their convection scheme to a single ocean model. The ocean model receives a weighted combination of the momentum and heat fluxes. Optimal weights can produce a supermodel with a basic state similar to observations: a single Intertropical Convergence zone (ITCZ), with a western Pacific warm pool and an equatorial cold tongue. This is in stark contrast to the erroneous double ITCZ pattern simulated by both of the two stand-alone coupled models. By varying weights, we develop a conceptual scheme to explain how combining the momentum fluxes of the two different atmospheric models affects equatorial upwelling and surface wind feedback so as to give a realistic basic state in the tropical Pacific. In particular, we propose a mechanism based on the competing influences of equatorial zonal wind and off-equatorial wind stress curl in driving equatorial upwelling in the coupled models. Our results show how nonlinear ocean-atmosphere interaction is essential in combining these two effects to build different sea surface temperature structures, some of which are realistic. They also provide some insight into observed and modelled tropical Pacific climate.