Vol.:(0123456789) 1 3
Climate Dynamics
https://doi.org/10.1007/s00382-021-06030-1
Climate change response in wintertime widespread fog conditions
over the Indo‑Gangetic Plains
Dipti Hingmire
1,2
· Ramesh Vellore
1
· R. Krishnan
1
· Manmeet Singh
1,3
· A. Metya
1,2
· T. Gokul
1,2
· D. C. Ayantika
1
Received: 4 June 2021 / Accepted: 24 October 2021
© The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021
Abstract
This study investigates the influence of climate change on widespread fog conditions over the Indo-Gangetic Plains (IGP) of
north India using observations, reanalysis data of atmospheric parameters, coupled model inter-comparison project 6 (CMIP6)
projections following four future scenarios based on the shared socio-economic pathways (SSP126, SSP245, SSP370,
SSP585), and advanced analysis techniques including machine learning. Two parameters fog fraction and widespread fog
days (WFDs) are estimated in this study by functional mapping of fog observations with 8 atmospheric parameters for the
period 1981–2018 using three empirical/machine learning approaches. Of these, we note that the deep learning convolutional
neural network (CNN) exhibits superiority in performance by showing the mapping closer to the observed, and also offers
promising potential for operational purposes to provide fog outlooks for the IGP region. Temporal evolution of fog fractions
and WFDs is analyzed from the CMIP6 projections following the aforementioned four future scenarios using CNN for the
future periods of the twenty-first century. It is noted that there is a substantial enhancement in the CMIP6 projected fog frac-
tions as high as 57% during the period (2015–2045) relative to the historical (1981–2014) period, while the largest increase
of 154% is seen in projected WFDs. It is also seen that the near-future period (2015–2045) witnesses a larger prevalence of
WFDs, for all scenarios except SSP126, due to the combined effects of air pollution and greenhouse warming. The post-2046
periods, however, generally indicate signatures of decline in foggy days with widespread conditions relative to historical
period in most of the scenarios except SSP370. The severity in fog conditions following the high-emission scenarios SSP370
and SSP585 during this period comes from the relative impact of mitigation strategies of pollutants. The findings provide
insights into the possible future changes in widespread fog conditions suitable for the IGP region.
Abbreviations
AO Arctic oscillation
AOD Aerosol optical depth
CMIP6 Coupled Model Inter-comparison Project 6
CNN Convolutional neural network
CC Correlation coefficient
ECMWF European Centre for Medium-Range Weather
Forecasts
ERA5 Fifth-generation ECMWF atmospheric
reanalysis
EU Eurasian pattern
IGP Indo-Gangetic Plains
IMD India Meteorological Department
IVS Inter-annual variability skill score
LR Linear regression
RE Relative error
SSP126 Shared socioeconomic pathway 1(radiative
forcing level by 2100 is 2.6 W m
−2
)
SSP245 Shared socioeconomic pathway 2 (radiative
forcing level by 2100 is 4.5 W m
−2
)
SSP370 Shared socioeconomic pathway 3 (radiative
forcing level by 2100 is 7.0 W m
−2
)
SSP585 Shared socioeconomic pathway 5 (radiative
forcing level by 2100 is 8.5 W m
−2
)
STCF Short-term climate forcers
SVR Support vector regression
TR Total Ranking Score
USH Zonal wind shear
RH Relative humidity
WFD Widespread fog days
* Ramesh Vellore
rameshv@tropmet.res.in
1
Centre for Climate Change Research (CCCR), Indian
Institute of Tropical Meteorology, Ministry of Earth
Sciences (IITM-MoES), Pune, India
2
Department of Atmospheric and Space Sciences, Savitribai
Phule Pune University, Pune, India
3
Jackson School of Geosciences, University of Texas
at Austin, Austin, TX 78712, USA