Extreme rainfall event analysis over the state of Himachal Pradesh in India

Extreme rainfall events (EREs) are very localized and intense events witnessing heavy rainfall resulting in flash floods and landslides in the Himalayan region. The trend of EREs in the state of Himachal Pradesh shows an increasing trend, so in this study, an attempt is being made for quantifying the climatological feature of these EREs in the state of Himachal Pradesh in the western Himalayan mountain region. Multi-source observed datasets for the long-term period, i.e., 1901 to 2020, are considered for the analysis. The climatology analysis indicates that northern and southern parts of Himachal Pradesh receive comparatively more rainfall, and the state witnessed a decreasing trend of the rainfall in the current decade. The classifications of these EREs in terms of monsoon and non-monsoon with a different threshold of daily rainfall are being analyzed, and it is found that almost 90% of extreme events are observed in the monsoon season. The zone centered around 32.5° N and 76.25° E seems to be the hotspot for the frequent EREs in the mountainous state. The spatial analysis of the ERE trend also indicated that almost 40% of the state witnesses the heavy rainfall throughout the years. This observational study will surely help in understanding the dynamics of these EREs in the higher altitude regions and can be used for hydro-meteorological disaster management and mitigation.


Introduction
Extreme rainfall in the mountainous region generally results in deadly disasters as the associated flash flood and landslide caused loss of humans, animals, cropland, infrastructure, etc. The extreme rainfall events (days) cause huge damage to the ecosystem and the economy of a state. Several studies inferred an increase in extreme rainfall events (EREs) with climate change particularly a rise in the major parameters like temperature and rainfall (Alexander et al. 2006;Kharin et al. 2007Kharin et al. , 2013Fischer and Knutti 2016). The increase in the ERE also represents the signature of regional climate change (Easterling et al. 2000;Kyselý and Beranová 2009;Zhang and Zhai 2011;. Basically, these studies are based on long-term climatology. The climate in the Himalayan mountain regions is characterized by tropical and subtropical climate conditions (Pant et al. 2018). The seasonal cycle of the major climate parameters, i.e., temperature and rainfall, is mainly influenced by the orography, topographical distribution, and the regional atmospheric dynamics like the small-scale convective events, and as a result, the Himalayas witnesses different variability in different zones (Bookhagen and Burbank 2006).
Few studies using the weather station observation data and the India Meteorological Department (IMD) longterm data analysis show the intra-regional trend of annual, monsoon, and winter rainfall in the north-western Himalayas (Guhathakurta and Rajeevan 2008;Sontakke et al. 2010;Bhutiyani et al. 2010;Dimri and Dash 2012). In recent time, the global interest is the change of climate due to the increase in the high-impact extreme rainfall events as many studies has the evidence both from observational analysis (Alexander et al. 2006;Goswami and Ramesh 2006;Fischer and Knutti 2016;Buerkert et al. 2020;Hanlon et al. 2021) and the future climate projection studies Orlowsky and Seneviratne 2012;Kharin et al. 2007Kharin et al. , 2013Fischer et al. 2013;Done et al. 2015;Bokhari et al. 2018;Chen et al. 2019;Hanlon et al. 2021).
Vanden Broucke et al. (2019) studied the dependency of topography and time scale on the future projection of extreme precipitation in high resolution. The complexities involved in high-resolution projection over mountain environments are essential in revealing the impacts of climate change (Moraga et al. 2021). Several studies used the long-term climate data and reanalysis product over the Mountain regions (Durand et al. 2009;Crochet 2007;Talchabhadel et al. 2018;Saddique et al. 2020). The ERE-induced disasters in the Himalayan region are also reported in earlier studies Talchabhadel et al. 2018;Bhardwaj et al. 2019Bhardwaj et al. , 2021Umar et al. 2021), which refers that the rainfall variability is mainly controlled by the intensity and frequency of such EREs (Das et al. 2006). The ERE estimations are required for a spectrum of applications in different domains like weather and climate, hydrology, agriculture, health, energy, water resource, and ecosystem (Valipour and Eslamian 2014;Valipour et al. 2017;Poveda et al. 2020). The ERE analysis can be used as a guidance for the modeling of landslide and avalanche prediction. However, there are not many studies that explore the ERE analysis in the Himalayan state like Himachal Pradesh which witnessed many EREs like cloud burst and flash flooding which also induced the landslide hazard in the state (Bhan et al. 2004(Bhan et al. ,2015Chaudhuri et al. 2015;Dimri et al. 2017). Some mesoscale modeling studies were also carried out for assessing the model capability in simulating such events (Das et al. 2006). The study of the spatial and temporal variability and analysis of ERE in the state will help in understanding the zone of hot spots of such events along with the physical mechanism. Daily gridded (0.25° × 0.25°) precipitation datasets from IMD (Pai et al. 2014) are used in this study for the historical period (1901-2020) over the state of Himachal Pradesh, India ( Fig. 1).
ERE is very much a user-defined parameter where the daily or hourly rainfall amount is considered to describe the heavy rainfall events as ERE. In this work, the ERE over the state of Himachal Pradesh situated in the Indian Himalayan range is analyzed using the IMD rainfall data. The state of Himachal Pradesh situated in the latitude (30° 15′-33° 15′ N) and longitude (75° 30′-79° E) in the western Himalaya has a wider elevation in the range 243-3028 m (Fig. 1). The current study includes different analyses like annual and monsoon rainfall climatology and interannual variability of the total rainfall in summer and winter monsoon. The ERE analysis with different thresholds is considered, and the frequency analysis of the number of rainy days in each category is quantified. Similarly, the interannual variability of the maximum daily intensity of rainfall is also presented for the years 1901-2020. The trend analysis of the ERE for both monsoon and non-monsoon seasons are analyzed and presented.

Results and discussion
The annual and monsoon rainfall climatology over the state of Himachal Pradesh is presented in Fig. 2 which indicates that the northern, north-western, and southern parts receive better rainfall as compared to the central and eastern parts that remain in a low rain zone. As the state does not fall under the core monsoon zone, so it is obvious that the monsoon rainfall intensity is also low in climatology sense. The maximum rainfall recorded is around 16 mmd −1 in the northwestern part of the state during the monsoon season which is mainly dominated by the orographic precipitation. The trend of monsoon rainfall in the state also shows a decreasing value indicating the regional climate change in the mountain state of Himachal Pradesh.
The interannual variability analysis of average rainfall during monsoon and winter over the state of Himachal Pradesh (Fig. 3) shows there is a slight decreasing trend in the monsoon rainfall and almost no trend in the winter rainfall. The rainfall quantity also decreased after 1950 as compared to the values before it during monsoon season whereas the winter rainfall shows less variability but shows less quantity in the recent decade consistently.
The interannual variability of the average rainfall days over the state of Himachal Pradesh during 1951-2020 ( Fig. 4) indicates the variability is high with 4 days standard deviation. Also, the impact of ENSO on the rainy day distribution is clearly visible in the analysis which indicates that more (less) rainy days are observed during El-Nino (La-Nina) years in the state of Himachal Pradesh.
The daily maximum rainfall recorded in all the year during 1901 to 2020 is presented in Fig. 5, it indicates the mean maximum attained daily rainfall is about 214 mm, and the variability of the time series is about 68 mm. To quantify the temporal assessment of the extreme rainfall days, the days in a year that received maximum rainfall over any part in the state of Himachal Pradesh are presented in Fig. 6 which clearly indicates that 90% of the time, the extreme rainfall events occurs in the monsoon season.
To analyze the temporal changes in the extreme rainfall event, the dates in each year when the maximum rainfall  Fig. 6. It is clear that almost all year the maximum rainfall occurs in a day during monsoon season, i.e., during July and August months. During the 16 years, the maximum rainfall day is observed in the non-monsoon season as indicated in the figure. The month-wise analysis (Fig. 6b) of the maximum rainfall for the 120 years shows that the numbers are very high in August (55) followed by July (28).
Analysis of 120 years rainfall data also indicated that the more frequent daily maximum rainfall event occurred in the active monsoon months, i.e. 55 years in August followed by 28 years in July (Fig. 7). It is inferred that the maximum rainfall in the Himachal Pradesh was received in the monsoon season. The frequency analysis shows that the area close to latitude 32.75 and longitude 76.25 witnessed a maximum of 36 extreme events followed by the region confined to longitude 77.75 and latitude 30.5 where 11 ERE The extreme rainy day (rainfall with different thresholds like 25 mm, 50 mm, and 100 mm per day) is analyzed for all the years, and the trend analyses for the monsoon and non-monsoon season are carried out and presented in Figs. 8 and 9, respectively. The northern and western parts show a decreasing trend both in monsoon and nonmonsoon seasons, and only a small region confined to 30-32° N along 76.8° E shows a slightly positive trend (i.e., 0.1 day/year) for all the three thresholds. For the threshold, 25 mmd −1 and 50 mmd −1 , the central eastern parts show no trend while for the extreme threshold of 100 mmd −1 , the rainy day trend shows a negative value. For non-monsoon years, eastern and north-eastern parts show a positive trend for the 25-mm and 50-mm threshold rainy days, and in the western parts, there is a negative trend in the rainy days over the state of Himachal Pradesh. The analysis clearly indicates that with a 25 mmd −1 threshold, the rainy days show a positive (negative) trend in the 42.2% (57.8%) region, and the value is 40% and 60% for the non-monsoon season where the rainfall intensity is comparatively low. For the 50-mmd −1 threshold, the positive and negative trends were shared by almost close to 50% while in the non-monsoon year, the positive trend was observed in 69% of regions in the state of Himachal Pradesh. In the 100-mmd −1 category, the rainy day shows positive in 41%, and in the non-monsoon season, the positive trend value is observed in 58% of the region; the values are almost reverse in the regions with negative trend, i.e., 60% and 40%, respectively (Table 1).

Conclusions
The present study shows that there is a clear sign of regional climate change over the state of Himachal Pradesh situated in the Himalayan region in India. The climatology of rainfall in annual and monsoon seasons reveals that the northern and southern parts receive more rainfall in the state, and the net rainfall in the recent decade shows a decreasing trend which is mainly due to the decrease in the thermal contrast between the Tibetan Plateau and the tropical Indian Ocean as explored earlier (Duan et al. 2006) and a clear cut signature of regional climate change. The assessment of extreme rainfall analysis shows that the maximum rainfall occurred in the state is 214 mm with interannual variability of 64 mm in the 120 years of rainfall analysis. The temporal analysis of the ERE with different thresholds like 25 mmd −1 , 50 mmd −1 , and 100 mmd −1 is quantified, and it is worth noting that 90% of the time of the extreme rainfall, an event occurs in monsoon season, i.e., in the month of July-August. It is also seen that in 16 years, the state witnessed the maximum daily ERE in the months during the non-monsoon season. The ERE frequency analysis also indicates that the region close to 32.5° N latitudes and 76.25° E longitudes experienced a large number of events (i.e., about 38%) in the study region, and this is mainly because of the rise in the air temperature in all season and the anthropogenic interventions resulting huge change in the land use land cover (LULC) of the region. The spatial trend analysis for the ERE with different user-defined thresholds for monsoon and non-monsoon season clearly shows a decreasing trend of these events in both the season in the north and western part. For the lowest and medium thresholds (25 mmd −1 and 50 mmd −1 ), the ERE analysis of central and eastern parts almost shows no change in the trend during monsoon while the trend is quite positive in the non-monsoon months indicating an increase in the non-monsoonal extreme rainfall events during pre-monsoon and winter season. Similarly, it is also quantified that the spatial coverage in the distribution of EREs is almost above 40% for all the threshold values throughout the year, and for the extreme category (events with rainfall > 100 mmd −1 ) the trend is positive in 41% (58%) in the monsoon (non-monsoon) season over the state of Himachal Pradesh. The patterns of ERE revealed the orographic control in the high-altitude regions resulting more of such events. The ERE variability and trend analysis clearly advocate the climate change over the mountain ecosystem with visible extreme events particularly heavy rainfall which causes landslides and associated disasters, and also, these changes in climate pattern will have direct implications in the sectors like water resource, agriculture, health, and energy. This study will be helpful in formulating the hydro-meteorological disaster mitigation of climate change; however, more studies with high-resolution