RESEARCH ARTICLEInterannual variability of monsoon onset and withdrawalin BangladeshCarlo Montes1| Nachiketa Acharya2| Mathew A. Stiller-Reeve3,4|Colin Kelley5| S. M. Quamrul Hassan61International Maize and WheatImprovement Center (CIMMYT),Texcoco, Mexico2Center for Earth System Modeling,Analysis, and Data (ESMAD), Departmentof Meteorology and Atmospheric Science,The Pennsylvania State University,University Park, Pennsylvania, USA3University of Bergen, Centre for Climateand Energy Transformation (CET),Faculty of Social Sciences, Bergen,Norway4NORCE Norwegian Research Centre,Bergen, Norway5Appalachian State University, Boone,North Carolina, USA6Bangladesh Meteorological Department,Dhaka, BangladeshCorrespondenceCarlo Montes, International Maize andWheat Improvement Center (CIMMYT),Texcoco, Mexico.Email: c.montes@cgiar.orgAbstractThis research investigates the interannual variability of monsoon onset andwithdrawal in Bangladesh, both of which are major climate features shaping mul-tiple societal activities. There is considerable research on the monsoon timing inSouth Asia, but with much less focus on Bangladesh. We applied a local monsoononset and withdrawal definition to observations and the latest-generationhigh-resolution gridded precipitation data from the Climate Hazards Center forthe period 1981 through 2018. We analyzed the interannual variability in mon-soon timing in Bangladesh and its teleconnection with the sea surface tempera-ture anomalies (SSTA) over the Pacific Ocean (El Niño Southern Oscillation,ENSO) and the Indian Ocean (IO). The monsoon starts with early significantrains in northeast Bangladesh and propagates westward, and a similar pattern isobserved for withdrawal, which tends to be more homogeneous in time andspace. A high spatial and temporal variability in monsoon onset and withdrawalis observed in Bangladesh, with a within-country average range of around1 month despite it being a country of relatively small size. The associationbetween monsoon onset and withdrawal and ENSO and IO is addressed at thecountry and regional level by analyzing composites for different ENSO and IOphases and associated atmospheric circulation and moisture transport. A similarassociation between monsoon onset and ENSO and IO phases was found, withgenerally earlier (later) onset dates duringthenegative(positive)phaseofENSOand IO. Monsoon withdrawal shows a clearer association with ENSO, with earlier(later) dates during the positive (negative) phase. Monsoon withdrawal is earlierduring the negative IO phase. SSTA-induced anomalies in circulation and mois-ture transport contribute to anomalies in monsoon timing. Results suggest bothENSO and IOD can be potentially used as sources of predictability of monsoononset and withdrawal over specific regions of Bangladesh.KEYWORDSrainy season, sea surface temperature, South Asian monsoon, teleconnectionReceived: 10 February 2021Revised: 16 September 2021Accepted: 17 September 2021DOI: 10.1002/asl.1069This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, providedthe original work is properly cited.© 2021 The Authors.Atmospheric Science Letterspublished by John Wiley & Sons Ltd on behalf of Royal Meteorological Society.Atmos Sci Lett.2021;22:e1069.wileyonlinelibrary.com/journal/asl1of10https://doi.org/10.1002/asl.1069
1|INTRODUCTIONThe main goals of this work are to statistically analyze the pri-mary regional patterns of variability in monsoon onset andwithdrawal in Bangladesh and assess the interannual influ-ence of large-scale teleconnections. These are important issuesfor a country that features a monsoonal climate with rainfallconcentrated typically from June to September and even ear-lier in the northeast region of the country (Ahmed &Karmakar 1993; Stiller-Reeve et al. 2015). This seasonality isresponsible for shaping multiple activities that directly dependon the timing and amount of the summer rains, as well asassociated changes in relevant meteorological variables suchas air temperature and humidity. Both the monsoon onsetand withdrawal are crucial attributes for agricultural planningin terms of the timing of land preparation, sowing, and trans-planting dates of summer crops during the main growing sea-son, and dates of planting and harvest of winter crops(Acharya & Bennett 2021; Ray et al. 2015).The timing of the South Asian monsoon has beenwidely studied, with previous studies mostly focused on theSouth Asian domain and on large-scale features (Bombardiet al. 2019; Wang & Fan 1999). In this way, multiple criteriaand atmospheric variables have been used to characterizethe monsoon timing variability and associated mechanisms(Carvalho et al. 2016; Wang et al. 2009; Zeng & Lu 2004),while others have assessed long-term trends (Bollasimaet al. 2013; Singh et al. 2014). Since the monsoon onset andwithdrawal are typically influenced by local and regionalmechanisms (Wang et al. 2017), multiple approaches havebeen proposed to explain its dynamics. In addition to thebasic large-scale features, namely, continental heating, thenorthward shift in meridional wind, and the associatedincrease in water vapor transport, other large-scale andregional mechanisms have been documented to drive thetransition. The latter includecirculation anomalies inducedby intraseasonal oscillations (Karmakar and Misra, 2019),the passage of deep convection over the adjacent ocean(e.g., the Bay of Bengal) propagating into the continent, andthe influence of the Tibetan plateau on circulation(Fasullo & Webster 2003; Yanai et al. 1992). Furthermore,large-scale earlier or advanced monsoon onset have beenassociated with sea surface temperatures (SSTs) anomaliesover the Indian and Pacific oceans (Lau & Yang 1997; Sunet al. 2017). Likewise, El Niño (La Niña) years have beenreported as associated with later (earlier) than normal onsetof the monsoon season in India (Adamson & Nash 2014;Joseph et al. 1994; Xavier et al. 2007) and farther East(Wang et al. 2013).Targeted studies that evaluate the monsoon inBangladesh have been very limited. The study of Ahmedand Karmakar (1993) can be regarded as the first toestablish a climatology of monsoon onset andwithdrawal. Recently, Stiller-Reeve et al. (2015)highlighted the heterogeneity of results obtained from aset of often-used monsoon onset definitions using coarse-resolution rainfall data and the differences between theseresults and the perception from local stakeholders inBangladesh. Here, we take a step further and discussinterannual variability of monsoon timing acrossBangladesh and possible teleconnections with the large-scale features. There is a need to consider the regionalityof the monsoon transitions in more detail in Bangladesh,as well as understanding the relationship between mon-soon timing and large-scale features such as the SST overthe Pacific and Indian oceans in order to generate predic-tive statistical models. This is why we aim to characterizethis variability using high-resolution data and localmethods, and also analyze how the resulting patternsrelate with SST.2|DATA AND METHODS2.1|Datasets2.1.1 | Stations and gridded precipitationDaily gridded precipitation data from the Climate HazardsGroup InfraRed Precipitation with Station product(CHIRPS v2; Funk et al. 2015) were used. This griddedproduct is generated by merging high-resolution TropicalRainfall Measuring Mission 3B42 v7 thermal infrared-derived satellite precipitation (Huffman et al. 2007), groundtruth rainfall observations from rain gauges, monthly pre-cipitation climatology from the Climate Hazards Group'sPrecipitation Climatology (CHPClim; Funk et al. 2015), andsimulated precipitation from the National Oceanic andAtmospheric Administration (NOAA) Climate Forecast Sys-tem V2. Data from January 1981 to December 2018 wereselected at the highest available spatial resolution of0.050.05, which in the current case allows a good spa-tial coverage of the relatively small area of Bangladesh. Inaddition, observed precipitation data from rain gauges wereprovided by the Bangladesh Meteorological Departmentand used here as a backgroundground truth observationalreference. These rainfall data correspond to daily time seriesspanning the period from January-1981 to December-2017for a total of 27 stations. Only stations with missing data lessthan 10% were included.2.1.2 | SST and atmospheric reanalysisEl Niño-Southern Oscillation (ENSO) and Indian OceanSST (IO) indices were generated in order to assess the2of10MONTESET AL.