key: cord-0839327-ps5jouas authors: Rahman, Mahzabin Binte; Salam, Roquia; Islam, Abu Reza Md. Towfiqul; Tasnuva, Anjum; Haque, Ubydul; Shahid, Shamsuddin; Hu, Zhenghua; Mallick, Javed title: Appraising the historical and projected spatiotemporal changes in the heat index in Bangladesh date: 2021-07-23 journal: Theor Appl Climatol DOI: 10.1007/s00704-021-03705-x sha: 5d7f4b8c05fbfa6051e6a3789d1570e6325a3fe8 doc_id: 839327 cord_uid: ps5jouas Climate change-derived extreme heat phenomena are one of the major concerns across the globe, including Bangladesh. The appraisal of historical spatiotemporal changes and possible future changes in heat index (HI) is essential for developing heat stress mitigation strategies. However, the climate-health nexus studies in Bangladesh are very limited. This study was intended to appraise the historical and projected changes in HI in Bangladesh. The HI was computed from daily dry bulb temperature and relative humidity. The modified Mann-Kendal (MMK) test and linear regression were used to detect trends in HI for the observed period (1985–2015). The future change in HI was projected for the mid-century (2041–2070) for three Representative Concentration Pathway (RCP) scenarios, RCP 2.6, 4.5, and 8.5 using the Canadian Earth System Model Second Generation (CanESM2). The results revealed a monotonic rise in the HI and extreme caution conditions, especially in the humid summer season for most parts of Bangladesh for the observed period (1985–2015). Future projections revealed a continuous rise in HI in the forthcoming period (2041–2070). A higher and remarkable increase in the HI was projected in the northern, northeastern, and south-central regions. Among the three scenarios, the RCP 8.5 showed a higher projection of HI both in hot and humid summer compared to the other scenarios. Therefore, Bangladesh should take region-specific adaptation strategies to mitigate the impacts of HI. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00704-021-03705-x. Climate change-derived heat extremes have become a matter of global concern with the rapid rise in temperature (Rajib et al. 2011; Rakib 2013; Ballester et al. 2016; Nissan et al. 2017; . The climate influences public health mainly through temperature extremes. Though cold extremes are more deadly than heat extremes (Dixon et al. 2005) , a gradual rise in earth temperature due to climate change has made heat extremes the primary concern to public health throughout the globe (Field et al. 2014 ). According to the United Nations Office for Disaster Risk Reduction (UNISDR, 2018) , nearly 166,000 people died between 1998 and 2017 due to heat extremes. Global Change (2018) report estimated about 9300 deaths per year by 2090 due to temperature extremes. The rising temperature would cause a gradual decrease in cold extreme-related fatalities and an increase in heat extreme-related health consequences. Therefore, the majority of public health implications in the future would be due to heat extremes. Many studies have been conducted to assess extreme heat effects on different human health aspects in recent decades, considering their severe consequences (Monteiro et al. 2013; McGregor et al. 2015; Ballester et al. 2016; Pal and Eltahir 2016; Modarres et al. 2018) . Temperature-based indices are generally used to appraise human exposure to extreme heat. Heat exposure is defined as a negative function of temperature, relative humidity, solar radiation, and wind speed that human experiences (Blazejczyk et al. 2012; Nguyen and Dockery 2016) . In contrast, the heat index (HI) is commonly known as the apparent temperature experienced by the human body. Therefore, the HI is generally used to measure human exposure to extreme heat. However, the measures of the HI vary according to the local climate and vulnerability of the population. Therefore, many HIs have been developed to assess human exposure to extreme heat (De Freitas and Grigorieva 2014) . According to Modarres et al. (2018) , out of 170 climate indices available to evaluate various climatic conditions, over 100 indices are designed to assess the nexus between meteorological parameters and HI. The human capability to adapt to heat is directly related to dry bulb temperature and humidity (Raymond et al. 2017) . Therefore, the HI based on temperature and humidity is most widely accepted to assess human exposure to extreme heat (Brouillet and Joussaume 2019) . The human temperature comfort decreases with the increase of temperature and relative humidity, and vice-versa. It is also related to the evaporative cooling of the human body or physical homeostasis (Kjellstrom et al. 2009; Buzan et al. 2015) . High temperature and humidity together cause human discomfort, which may lead to heatstroke. Sweating is an essential body mechanism by which the human body can transport away metabolic heat. At a higher dry bulb temperature and relative humidity, the human body's heat exchange process is ceased, which results in accumulating excessive temperature in the human body (Budd 2008) and accelerates the rate of morbidity and mortality (Basu and Samet 2002) . Therefore, HI estimated based on both dry bulb temperature, and relative humidity is most widely accepted to evaluate heat extremes. Though heat extreme-related casualties are increasing globally, the growing rate of health consequences is much higher in developing countries, mainly densely populated countries, due to higher exposure. South Asia is one of the most vulnerable regions regarding public health exposure to temperature rise (Singh et al. 2016) . Climate models projected more frequent and intense heat extremes in the region than in many other regions of the globe (UN ESCAP 2015; Im et al. 2017 ). Bangladesh, located in South Asia, is enormously susceptible to global climate change (Huq 2001; Rahman and Islam 2019; . The country has experienced some recent heat extreme occurrences, which caused high mortality, especially the death of males and people older than 65 years (Burkart et al. 2014) . Heat extreme-related mortality rate has increased by 20% in Bangladesh over the past decade (Nissan et al. 2017) . Climate change might increase the frequency and intensity of heat exposure in the forthcoming period (Kirtman and Coauthors 2013) . Battisti and Naylor (2009) reported an increase in heat extremes in tropical and subtropical regions in the future. Future forecasts of climate change scenarios in Bangladesh demonstrated a higher occurrence of extreme heat (Nissan et al. 2020) . The minimum heat exposure of consecutive 3 days might exceed up to 10 days in the country in the future (Nissan et al. 2017) . Although many researchers have investigated the changes and variabilities of Bangladesh's climate at the regional and country levels (Islam et al. 2017 , the climate-health relationship studies in the country are very rare Nissan et al. 2020) . The previous studies are focused mainly on HI anomaly assessment (Rajib et al. 2011) , extreme temperature variability (Rakib 2013) , HI trends in southwest Bangladesh (Rakib 2018) , effects of heat extreme (Burkart et al. 2014) , mortality due to high temperature (Burkart et al. 2011) , and heatwave prediction (Nissan et al. 2017) . No attempt has been taken to quantify the spatiotemporal change in historical (observed) and projected HI in different climatic regions of Bangladesh. This study intends to fill this research gap by quantifying the HI's spatiotemporal changes in Bangladesh due to climate change. The information generated in this study can be used for adaptation planning and reducing climate change impacts on disease and death, particularly the large number of marginally poor people involved in agriculture and construction and exposed to heat extremes in the country (McGregor et al. 2015; Havenith and Fiala 2016) . Bangladesh, a low-lying subtropical South Asian country, has a land area of 147,570 km 2 (Rahman and Islam 2019). Three dominant seasons are prevailing in Bangladesh, pre-monsoon hot summer (March to May), monsoon humid summer (June to August), and post-monsoon (September to November) and dry winter (December to February) seasons. The annual mean rainfall is approximately 2,400 mm, of which about 75% occurs in the monsoon from June to September. Long-term daily mean relative humidity, minimum temperature, maximum temperature, and evapotranspiration across Bangladesh are 80%, 21.39°C, 29.94°C, and 3.72 mm d −1 , respectively . January is the coldest, and April to October is the hot months in Bangladesh. Banglapedia (2014) demarcated Bangladesh into seven climatic sub-regions based on climate, land use patterns, and hydrogeologic settings (Figure 1 ). There is spatial variability in climate in Bangladesh (Ghose et al. 2021) . The western zone of the country is relatively drier compared to other areas (Islam et al. 2017 ). Bangladesh Meteorological Department (BMD) operates only 43 weather stations throughout the country. Weather stations across the country are not uniformly distributed. They are mostly concentrated in the southeastern and south-central regions. Many of the stations are set up after the 1990s and do not have a long-term record (www.bmd.gov.bd). Due to these shortcomings of long-term data availability, only 20 weather stations were selected to estimate the HI for the period 1985-2015. These selected stations encompass the seven climatic sub-regions of the country. The period between 1985 and 2015 is regarded as the warmest period on record (IPCC 2019). The global average surface temperature rising was 0. 87°C during 1985-2013 relative to 1850 -1900 (IPCC 2013 . For that reason, the daily climate dataset for the period 1985-2015 was used as the historical period. Daily dry-bulb temperature (T) (°C) and mean relative humidity (R) (%) data of 20 stations were acquired from the BMD. The missing data at all the stations were found less than 5%. The missing data were filled using the past data of the respective days or the data from the neighbouring stations after consulting with the expert of BMD. The BMD follows the World Meteorological Organization (WMO) guideline for weather dataset acquisition and record archiving. Nonetheless, a homogeneity test and autocorrelation of the dataset were performed to reveal any anomaly in the dataset . All the datasets were passed the quality control test conducted by BMD staff. The statistical downscaling model (SDSM) usually develops the statistical relationship of regional predictors with Figure 1 Map showing the location of the study area and meteorological stations along with climatic zones in Bangladesh forecasters utilizing the multiple linear regression model (Wilby and Dawson 2007) . Typically, this relationship is formed between station data (e.g., temperature) and NCEP variable, as a proxy of GCM simulation, for the study period. The GCM simulations of NCEP variables are then used in SDSM for historical simulation and future projections of the variable at the station level. In this study, two variables, temperature and relative humidity, were downscaling at 20 stations in Bangladesh (details in Section 2.2). The GCM data of CanESM2, obtained from the Canada Climate change site, was used for this purpose. The inputs of the downscaling model were selected by screening 26 NCEP variables based on their association with station data. The GCM simulations of the selected NCEP variables for three RCP scenarios (very low emission-RCP 2.6, medium emission-RCP 4.5, and a very high emission-RCP 8.5) were used to downscale the future projections of temperature and relative humidity for those three scenarios. The SDSM was calibrated with 70% of historical data and validated with the rest (30%). The performance of the downscaling model was evaluated using the coefficient of determination (R 2 ), standard error (SE), and Durbin Watson (DW) tests. All the statistical tests have given satisfactory results (Table S1 ), indicating the ability of downscaling models developed in this study in downscaling temperature and relative humidity. Figure S1 illustrates the key steps used in this SDSM. The downscaled temperature and relative humidity data were finally used to simulate HI for the three RCPs for the period of 2041-2070. The CanESM2 model can generate reasonable projections of worldwide climate variability (Arora et al. 2011; Wang and Zhang 2019; Chavaillaz et al. 2019; Yengoh and Ardö 2020) . The result produced by the CanESM2 model is found reliable based on its historical performance and accuracy (Javaherian et al. 2021) . Therefore, this model has been extensively used for climate projections in different regions, including South Asian countries (Das and The Intergovernmental Panel on Climate Change (IPCC) has regarded the GCMs as an important tool for assessing future climatic impacts. SDSM has been recognized as an efficient technique for downscaling GCM simulation (Gagnon et al. 2005) . Therefore, the present study utilized SDSM for downscaling CanESM2 simulations at the station level for the estimation of HI. For estimating the HI using the CanESM2 model, the temperature data collected from BMD were converted from°C to°F using the following Eq. (1) for easy calculation of the HI: The historical HI was computed from relative humidity (%) and temperature (°F) using Eq. (2). where the HI denotes heat index (in degrees Fahrenheit); T indicates ambient dry-bulb temperature (in degree Fahrenheit); R means relative humidity (percentage value between 0 and 100); and C 1 =−42.379; C 2 =2.04901523; C 3= 10.14333127; C 4 =−0.22475541; C 5 =−6.83783×10 −3 ; C 6 = −5.481717×10 −2 ; C 7 = 1.22874×10 -3 ; C 8 = 8.5282×10 −4 ; C 9 = −1.99×10 −6 The HI, used in this study, integrates bulb temperature and humidity and estimates health exposure (HE) to heat extreme (Anderson et al. 2013; Modarres et al. 2018) . The HI was demarcated into four classes, e.g., caution (80-90°F), extreme caution (90-105°F), danger (105-130°F), and extreme danger (>130°F) (Belding 1970; Modarres et al. 2018) . The HI for the historical period 1985-2015 was estimated from observed data, while it was calculated for the future period using GCM projected data for the period 2041-2070. Different techniques are available for assessing the trend of climatic variability. The scholars used Mann-Kendall (MK) (Mann 1945; Kendall 1975) and modified MK (MMK) tests for detecting trends (Vousoughi et al. 2013; Jhajharia et al. 2014; Dinpashoh et al. 2014; Zamani et al. 2017; Praveen et al. 2020; Islam et al. 2021) . The MK test has the problem of serial correlation, which may affect the trend result. Yue and Wang (2004) introduced the MMK test for eliminating the serial correlation effect on test significance. MMK test provides a more accurate assessment of the significance of the existing trends because the influence of the existing trend on sample serial correlation estimation has been effectively limited (Hamed and Rao 1998) . The nonparametric MMK test was utilized in this study to assess the HI trends. The original equation of the MK test (Mann 1945; Kendall 1955 ) statistics (S) is expressed by following Eqs. 3-4: Values of S indicate the direction of decreasing or increasing trend. The variance ) of S is followed by the Eq. 5: V*(S), is modified by Yue and Wang (2004) and given by the following Eq. 6: The correction factor is denoted by n/n* and expressed (Bayley and Hammersley 1946) by following Eq. 7: Test statistic Z is calculated by following Eq. 8: Positive Z values mean increasing trend and negative Z values mean decreasing trend. "modifiedmk" package of R statistical software (v3.5.3) was used for performing the MMK test. Sen (1968) proposed the method for measuring the change per time, which is given by the following equation: where β is the slope between x j and x k . The median N values of β i provides the Sen's slope of estimator β. The present study has utilized Sen's slope estimator for identifying the magnitude of the change of the historical HI (Sen 1968; Islam et al. 2019; Zinat et al. 2020 ). Linear regression was used for exploring the relationship between two variables of continuous data. The following equation is expressed the statistical form of linear regression: where Y denotes the dependent variable, a is constant, b means the coefficient and slope of the regression line, and X demonstrates the independent variable. Here, the historical HI is used as a dependent variable and time (years) as an independent variable. The value of the correlation coefficient (r) varies from −1 to 1. The absolute 1 denotes a strong correlation, and a value near zero indicates a weak correlation. This simple linear regression method has been applied to monitor the monotonic changes in the historical HI in this study Figures 2 and 3 show the temporal changes of the HI between hot summer and humid summer, respectively, in the projected period compared to the historical period. Figure 2 demonstrates that the HI values ranged from 88 ∘ − 92 ∘ F, indicating the extreme caution in the northern, western, south-central, and southeastern regions. In the northeastern region, the HI values range from 83 ∘ − 85 ∘ F, also indicating the caution HI. The projected HI slightly decreased than the historical HI in the southwestern region. On the contrary, considerable declination in HI (80 ∘ − 84 ∘ F) was noticed in the northwestern region compared to the historical (85 ∘ − 88 ∘ F) period during the hot summer (March-May). All the regions showed caution HI for the historical period (HI <90°C), except in the southwest (extreme caution HI). Figure 3 shows increase in HI in the northern, northeastern, and southcentral regions during humid summer (June-August). The rest of the regions would experience less HI than historical HI. Like the hot summer, the northwestern region would experience considerable declination of the HI compared to historical HI. For both hot summer and humid summer, the projected scenarios' ascending order in increasing HI is RCP8.5 > RCP4.5 > RCP2.6. The long-term station-wise mean annual variations of HI for hot summer and humid summer are shown in supplementary Figure of S2. Both hot summer and humid summer showed nearly the same projected HI for RCP 2.6, RCP 4.5, and RCP 8.5. The projected scenarios of RCP2.6, RCP 4.5, and RCP 8.5 for HI were higher than the historical HI in all the stations except for Bogra, Faridpur, and Chattogram stations for hot summer ( Figure S2a) and Bogra, Khepupara, Satkhira, and Cox's Bazar stations for humid summer ( Figure S2b ). It is shown in Figure S2 that Bangladesh would be facing higher HI in humid summer compared to hot summer in the future. The probability density function (PDF) plot was used to investigate the distributional change mean annual HI for the historical and projected periods. The PDFs of the HI for the seven climatic zones of Bangladesh predicted the potentially significant changes (Figure 4 ). Rightward shifting of the shape is a sign of the future-increasing trend of the mean annual HI. The future-expanding pattern in the HI in different zones can be ranked as Zone C (Northern), followed by Zone B (Northeastern), Zone G (South-central), Zone E (South-eastern), Zone F (South-western), Zone A (Western), and Zone D (North-western). Among these zones, Zone D (Northwestern) showed a remarkable decrease in HI in the future. The spatial changes of the HI during the historical and projected period (2041-2070) for three scenarios are presented in Figure 5 . The results show that the HI values in the northern zone would increase from 81-84°F to 85-86°F compared to the historical (observed) HI, indicating increasing temperature and relative humidity in the north. The HI of southwestern, south-central, and western zones also showed an increase. Surprisingly, the HI values would decrease in a few stations like Chattogram and Faridpur. Figure 6 demonstrates that most regions of Bangladesh, exposed to cautious conditions (8090-105°F during the historical period. Unlike the historical period (HI ranged from 89-92°F), the northeastern region would experience higher HI (92-94°F) in the future for all scenarios. On the contrary, the southeastern part of the southeastern (A zone) region would expose to the same HI level as the historical period. Most of Bangladesh that experienced extreme condition in the past would also experience extreme condition in the future (94-96°F and 96-98°F) during humid summer. The HI and the rate of HI change for the historical period of 1985-2015 are shown in Figure 7 . The figure demonstrates that HI has increased significantly (p > 0.01, 0.05, 0.1) in most of the country except (insignificant decreasing trend) some parts in the southwestern region and the northern and northeastern of zone G (south-central region) (Figure 7a ). Northern and southeastern regions experienced the most significant (p>0.01, MMK Z value ranged from 6-7) increasing trend in HI compared to other regions (Figure 7a ). The rate of change in HI was from −0.01 to 0.03°F per decade The present study investigated the changes in seasonal and spatial HI for historical and projected periods across Bangladesh. This study represents a crucial step to quantify temporal and spatial changes of HI over Bangladesh under three different future climate change scenarios. Most of the earlier works in Bangladesh are based on the maximum temperature and minimum temperature changes, the temporal pattern of extreme temperature condition, heatwave, etc. (Rakib 2013; Fahad et al. 2018; Khan et al. 2019; Nissan et al. 2020 ). However, Fahad et al. (2018) used a statistical downscaling model to project exposure to rising temperature in Bangladesh. They found that northern Bangladesh will be more affected due to heat extremes. The present study also found that the northern region would be more vulnerable to HI during the hot summer season due to increased temperature and precipitation (a high precipitation rate accelerates the percentage of relative humidity), which is consistent with the results of this study. Rajib et al. (2011) showed increases in HI in Bangladesh's southwestern and northwestern regions from 1961-2010. This study also reported increasing HI in the northern (contrary to Rajib et al. 2011), northeastern, and south-central regions (consistent with Rajib et al. 2011 ) of the country during the historical period . The most probable reason for the increasing the HI is unfavourable climatic conditions, environmental pollution, and the lessening and irregularity of rainfall, causing a rising trend in humidity in northwestern, central, and eastern Bangladesh. The right shift of the PDFs suggests that the onset of an increasing trend in HI mean and extremes and rare heatwaves in the future (Modarres et al. 2018) . This study revealed that the northern and northeastern regions have the highest probability of increasing HI. At the same time, Figure 4 Probability density function of the mean annual HI for the seven climatic zones of Bangladesh the northwestern region might experience a decreasing rate of HI in the future period. The cause of increasing HI is the rising humidity in the northwestern, central, and eastern regions due to unfavourable climatic conditions, environmental pollution, and the abatement and irregularity of rainfall (Rajib et al. 2011) . In this study, a remarkable change of the HI is detected in Barishal during the humid summer for three different future scenarios. In the far future, the HI value may turn into a dangerous condition in Barishal. Besides, the study also indicates that the HI would increase at an alarming rate at Rangpur, Chandpur, Cumilla, and Rajshahi. Surprisingly, HI would decrease at Bogra, Chattogram, and Faridpur in the future period. The temperature for higher RCP showed a decrease and rainfall to increase in Chattogram and Faridpur, leading to a reduction of the HI of these two stations (Alamgir et al. 2019; Islam et al. 2019; Rahman and Islam 2019) . Rakib (2013) demonstrated that the HI values are within a comfortable limit during the dry season. But it increases rapidly during summer, which is similar to our finding. For example, in hot summer, Rangpur would experience a change in HI at least 10% in the future (2041-2070) compared to the historical period. In contrast, Barishal would show the highest HI in humid summer as it is located near the coast. The high humidity is the cause of the elevated HI at Barishal. Prolonged exposure to such HI conditions may result in epidemiological problems such as human health hazards (Hansen et al. 2008) . Although the physiology of the human body may be adapted to a long-term change in climate (Hondula et al. 2015; Kjellstrom et al. 2016) , the outcomes recommend an urgent requirement for health care system guideline, robust early warning scheme, and healthcare plans to cope with the rising temperature. The rising HI may lead to a rise in heat-related death Figure 5 Spatial changes of the HI during the observed and projected period (2041-2070) under three different RCP scenarios over Bangladesh (Bowler 2005; Fischer et al. 2012; Hansen et al. 2012; Buzan et al. 2015; Li et al. 2018; Pal and Eltahir 2016) , the risk of novel infectious diseases like Coronavirus , air conditioning demand, human errors in the workplace, and an increasing trend of accidents. This indicates the need to shift work hours to avoid severe circumstances (Kjellstrom et al. 2016; Mora et al. 2017; Bodrud-Doza et al. 2020) . The increase in temperature and relative humidity over Northeastern Indian states, including Assam bordering Bangladesh, and the rising temperature and changes in rainfall pattern/ humidity may cause higher cases of vector-borne diseases . The present study has only employed the CanESM2 for future projection of the HI. The future study should consider more than one GCM and explore the best GCM for projecting the climate variability of Bangladesh. Station data were collected from only 20 meteorological stations for HI projections. The future study should collect data from more stations and project the HI from 2021. In the future, a quantitative risk assessment based on the HI in various climatic zones of Bangladesh at the regional and global scales should be performed, considering various demographic information. The present study intends to show the historical and future spatiotemporal changes in HI in seven climatic regions of Bangladesh. The HI in Bangladesh varies at the seasonal scale and among the climatic zones. The projected results revealed that HI's risk in Bangladesh would be more severe during humid summer than that of the hot summer season for all the three RCPs of RCP 2.6, RCP 4.5, and RCP 8.5 scenarios at both the temporal and spatial scales. This study has identified future hotspots of HI, particularly in the country's northern, northeastern, and south-central regions. The future-expanding pattern in the HI in different zones can be ranked as Zone C (northern). The outcomes obtained in this study by considering temperature with relative humidity can be regarded as more reliable for Figure 6 Seasonal changes of the observed HI and the projected HI (2041-2070) across Bangladesh for hot summer and humid summer under three different scenarios emerging emergency management strategy planning and early warning schemes. The results would help the planners and policymakers formulating and implementing adaptation strategies for minimizing the effects of future climate change. These results are also crucial for policymakers to take initiatives for coping with this alarming risk of heat extremes by preparing region-wise and season-specific different contingency plans. This was the first attempt to quantify the spatiotemporal change of the historical and projected HI in various climatic regions of Bangladesh. The inclusion of relative humidity with temperature for the projection of HI has made the estimation more realistic for a humid country like Bangladesh. The online version contains supplementary material available at https://doi.org/10.1007/s00704-021-03705-x. Downscaling and projection of spatiotemporal changes in temperature of Bangladesh Methods to calculate the heat index as an exposure metric in environmental health research Carbon emission limits required to satisfy future representative concentration pathways of greenhouse gases Comparison between statistical and dynamical downscaling of rainfall under Representative Concentration Pathways and projected spatiotemporal changes in the heat index in Bangladesh scenarios over the Gwadar-Ormara basin Evaluation of statistical model for future precipitation and temperature in drainage area of Jhelum river European seasonal mortality and influenza incidence due to winter temperature variability Drought in Bangladesh Relation between elevated ambient temperature and mortality: a review of the epidemiologic evidence Historical warnings of future food insecurity with unprecedented seasonal heat The effective number of independent observations in an autocorrelated time series The search for a universal heat stressindex Comparison of UTCI to selected thermal indices Psychosocial and socio-economic crisis in Bangladesh due to COVID-19 pandemic: a perception-based assessment Acclimation, heat shock and hardening Investigating the role of the relative humidity in the co-occurrence of temperature and heat stress extremes in CMIP5 Projections Wet-bulb globe temperature (WBGT)-its history and its limitations The effect of atmospheric thermal conditions and urban thermal pollution on all-cause and cardiovascular mortality in Bangladesh An analysis of heat effects in different subpopulations of Bangladesh Implementation and comparison of a suite of heat stress metrics within the Community Land Model version 4.5. Geosci Model Dev Exposure to excessive heat and impacts on labour productivity linked to cumulative CO2 emissions Downscaling monsoon rainfall over river Godavari Basin under different climate-change scenarios A comprehensive catalogue and classification of human thermal climate indices Effect of short-term and long-term persistence on identification of temporal trends Climate extremes and precipitation trends in Kelani river basin, Sri Lanka and impact on streamflow variability under climate change Heat mortality versus cold mortality: a study of conflicting databases in the United States Regional changes of precipitation and Temperature over Bangladesh using bias-corrected multimodel ensemble projections considering high-emission pathways Climate Change 2014: Impacts, adaptation, and vulnerability. Part A: Global and sectoral aspects. Contribution of working group II to the fifth assessment report of the intergovernmental panel on climate change Contrasting urban and rural heat stress responses to climate change An application of the statistical downscaling model (SDSM) to simulate climatic data for streamflow modelling in Québec Climate-induced rice yield anomalies linked to large-scale atmospheric circulation in Bangladesh using multi-statistical modeling Impacts, risks, and adaptation in the United States: The Fourth National Climate Assessment A Modified Mann-Kendall trend test for autocorrelated data Perception of climate change Humid tropical forest clearing from 2000 to 2005 quantified by using multitemporal and multiresolution remotely sensed data Thermal indices and thermophysiological modeling for heat stress Rising temperatures, human health, and the role of adaptation Climate change and Bangladesh Climate change 2014: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C abovepre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the globalresponse to the threat of climate change, sustainable development, and efforts to eradicate poverty Drought hazard evaluation in boro paddy cultivated areas of western Bangladesh at current and future climate change conditions Assessing recent impacts of climate change on design water requirement of Boro rice season in Bangladesh Trends in cooling and heating degree-days overtimes in Bangladesh? An investigation of the possible causes of changes Almoajel AM (2021) Spatiotemporal nexus between vegetation change and extreme climatic indices and their possible causes of change 2021) Prediction of changes in climatic parameters using CanESM2 model based on Rcp scenarios (case study): Lar dam basin Influence of climate on incidences of malaria in the Thar Desert, northwest India Trends in temperature over Godavari river basin in southern peninsular India Rank Correlation Methods Rank Correlation Methods Climate change projection for the marsyangdi river basin, Nepal using statistical downscaling of GCM and its implications in geodisasters Detection of changes and trends in climatic variables in Bangladesh during Contributionof working group I to the fifth assessment report of the intergovernmental panel on climate change Workplace heat stress, health and productivity -an increasing challenge for low and middleincome countries during climate change Heat, human performance, and occupational health: a key issue for the assessment of global climate change impacts Elevated increases in humanperceived temperature under climate warming Nonparametric tests against trend Robust global ocean cooling trend for the pre-industrial Common Era Future heat stress arising from climate change on Iran's population health Excess mortality and morbidity during the Global risk of deadly heat Daily indoor-to-outdoor Temperature and humidity relationships: a sample across seasons and diverse climatic regions Defining and Predicting Heat Waves in Bangladesh Targeted model evaluations for climate services: a case study on heat waves in Bangladesh Future Temperature in southwest Asia projected to exceed a threshold for human adaptability Analyzing trend and forecasting of rain-fall changes in India using non-parametrical and machine learning approaches Are precipitation concentration and intensity changing in Bangladesh overtimes? Analysis of the possible causes of changes in precipitation systems Analysis and prediction of rainfall trends over Bangladesh using Mann-Kendall, Spearman's rho tests and ARIMA model Health consequences of climate change in Bangladesh: an overview of the evidence, knowledge gaps and challenges Increase of heat index over Bangladesh: Impact of climate change Extreme Temperature Climatology and Evaluation of Heat Index in Bangladesh Characterization of climate change in southwestern Bangladesh: trend analyses of temperature, humidity, heat index, and rainfall Spatiotemporal patterns and synoptics of extreme wet-bulb temperature in the contiguous United States Downscaling of CMIP5 models output by using statistical models in a data scarce mountain environment (Mangla Dam Watershed) Potential of RT, Bagging and RS ensemble learning algorithms for reference evapotranspiration prediction using climatic data-limited humid region in Bangladesh The optimal alternative for quantifying reference evapotranspiration in climatic sub-regions of Bangladesh Assessment of present and future climate change over Kashmir Himalayas COVID-19 pandemic,socioeconomic crisis and human stress in resource-limited settings: a case from Bangladesh. Heliyon 6: e04063 Climate change, health and future well-being in South Asia The UN Economic and Social Commission for Asia and the Pacific The United Nations Office for Disaster Risk Reduction, annual report Trend analysis of groundwater using nonparametric methods (case study: Ardabil plain) Heat Stress Response to National-Committed Emission Reductions under the Paris Agreement Statistical downscaling model (SDSM) Climate change and the future heat stress challenges among smallholder farmers in East Africa The Mann-Kendall test modified by effective sample size to detect trend in serially correlated hydrological series Streamflow trend analysis by considering autocorrelation structure, long-term persistence, and Hurst coefficient in a semi-arid region of Iran Appraising drought hazard during Boro rice growing period in western Bangladesh Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations Acknowledgements We greatly acknowledge the Bangladesh Meteorological Department (BMD) for proving the necessary datasets in this work.Author contribution ARMTI conceptualized the study; MBR analyzed the data; MBR performed the GIS-based mapping; ARMTI, AT, SMI, JM, and ZH interpreted the results; MBR and AT searched the relevant literatures and drafted the initial manuscript. First extensive review and edits were performed by UH. All authors subsequent reviewed and edited for its clarity. Besides, a critical review was done by UH, ZH, SS, JM, and ARMTI. Final version was approved by all the authors. Data availability The time series datasets supporting the conclusions of this article will be made available upon request on the corresponding author.