Sensitivity of Diferent Physics Schemes in the Simulation of Heat Wave Events over Bangladesh Using WRF-ARW Model Sahadat Jaman 1 , Md. Jafrul Islam 1 , Ashik Imran 1 , Md. Kamruzzaman 1 , M. A. K. Mallik 2 , Pappu Paul 3 and I. M. Syed 1 1 Department of Physics, University of Dhaka, Dhaka-1000, Bangladesh 2 Bangladesh Meteorological Department, Agargaon, Dhaka-1207, Bangladesh 3 Department of Meteorology, University of Dhaka, Dhaka-1000, Bangladesh ( Received: 4 January 2022 ; Accepted : 6 April 2022) Abstract Heat waves (HWs) are an extreme temperature condition that has a direct impact on human lives. In recent years, a large number of people have died all over the world due to hot weather. The purpose of this study is to predict HWs accurately to mitigate the casualties caused by them. Two HW events are selected for this study (Event-1: 0000 UTC of 18 May to 0000 UTC of 25 May 2015, Event-2: 0000 UTC of 05 April to 2100 UTC of 08 April 2015). At frst, sensitivity tests have been done using diferent combinations of physics schemes. Sensitivity of Planetary boundary layer (PBL) and surface layer (SL) schemes combinations (YSU-Revised MM5, YSU-MM5, MYJ-Eta, and ACM2-Revised MM5) and land surface models (RUC, Noah, Noah-MP, and CLM4) are investigated to predict Comfort Index (CI), which is identifed by using Physiological Equivalent Temperature (PET). To simulate PET, the primary meteorological variables 2-m air temperature (T2), 2-m relative humidity (rh2), mean radiant temperature (TMRT), wind speed at 10 m (ws10), and cloud cover data have been used. These parameters were simulated by the WRF model using both single and nested domains. The experiments found that the combination of the YSU-MM5 scheme and the Noah land surface model predicted the WRF simulated variables very well. The study also found that the CI exists between the slight heat stress to extreme heat stress and the maximum PET values were found to be 47.6 ºC and 48.5 ºC for Rajshahi and Khulna event respectively. Keywords: Heat waves, Climate change, Land surface model, Surface Layer, PET, CI. I. Introduction A Heat Wave (HW), a period of successive hot weather along with high humidity, afects both the natural and human environment. Their most life-threatening impacts are the degeneration of human thermal comfort conditions which consequently increase various heat related diseases. It is well established that extreme heat poses a serious health risk, causing many deaths each year 1 . Every year a large number of people die because of heatstroke and heat exhaustion from the excessive heat exposure in the various regions over the world. In 2021, more than 1,400 people died in western North America because of HW with more than 800 excess deaths than the expected in western Canada 2 . Nowadays, in Indian subcontinent HWs are increasing alarmingly because of climate change, infuenced by human or various devastating hazards. HWs were ranked as 4 th of the 10 deadliest natural disasters in South Asia in 2015 3 . In India, around 17000 people died due to extreme HW in between 1970 to 2019 4 . Bangladesh is also one of the world’s most climate vulnerable countries, with HW frequency and severity expected to rise in future 5 . An eight day long HW in 2008 killed approximately 3800 people, mostly elderly, in Bangladesh 6 . Several cities in Bangladesh have experienced a signifcant increase in the surface urban heat island intensity (SUHII) over the past 20 years (2000–2019), with the highest rise of 19 . °C at night for Chattogram city 7 . The study of Meehl et al. showed that the intensity, frequency and duration of HWs will also increase in the second half of the 21 st century 8 . At the end of the century, the maximum temperature is expected to increase by 2°C to 2.5°C under RCP4.5 (Representative Concentration Pathway) scenario, and by up to 4°C for the RCP8.5 scenario 9 . To mitigate the casualties early warning systems of HWs have already been placed in developed countries 10 , which have been known to save lives 11 . According to Bangladesh Meteorological Department (BMD), four types of HWs are categorized in Bangladesh based on temperature. These are mild (36 ° -38 ° C), moderate (38 ° -40 ° C), severe (40 ° -42 ° C) and very severe HWs (>42 ° C) 12 . To develop HW warning system, WRF model is a valuable tool to compute primary meteorological variables. In this paper, the study investigated the variations of PBL, SL and LSMs schemes which are strongly related to the HWs. Afterwards, using the WRF model output parameters as the input to the Rayman model, the study calculated the PET and compared it to the observed PET to have a proper combination of schemes to predict the HWs. II. Model and Methodology Identifcation of Heat Wave Events HW event is identifed by using an extreme temperature criterion based on IPCC’s (Intergovernmental Panel on Climate Change) defnition of extreme temperature index, TX90p 13 . As stated by the index TX90p, a HW day is *Author for correspondence. e-mail: imsyed@du.ac.bd ISSN 1022 - 2502, E-ISSN 2408-8528 DOI: https://doi.org/10.3329/dujs.v70i1.60384 Dhaka Univ. J. Sci. 70(1): 70-78, 2022 (January)