Recent warming and its risk assessment on ecological and societal implications in Nepal

A predominantly mountainous country Nepal has a complex climatic pattern that varies from tropical in the south (Terai region) to arctic in the north (Himalayas). The gradual rise in temperature in the mountainous region has attracted great interest among the scientific community in general over recent years. However, recent warming in Nepal’s east-west and south-north temperature gradients and its implications for ecology and society based on facts and figures are still lacking. In this context, temperature data (1970–2016) of 76 meteorological stations from the Terai region to the Mountains were used in this study to analyze the annual and seasonal warming trends in the different physiographic regions of Nepal. We performed a hybrid analytical approach i.e. integrated statistical and theoretical tools to detect the warming trend and its ecological and societal implications across the country. The Eastern part of the country was found to be more warming than the Central and Western parts, showing an increased climatic sensitivity across the Khumbu (Mt. Everest region). The increasing trends of temperature have been found in all physiographic regions along an altitude gradient, i.e. Terai, Siwaliks, Lower Hills, and Upper Hills observed 0.15, 0.26, 0.68, and 0.57 °C per decade, respectively. Higher warming trend in Lower Hills than the Upper Hills showed that higher elevations experienced lesser degrees of warming trends than the lower elevations in the mountainous regions. Further, a higher warming trend was observed in the winter season than the other seasons in all regions except for Terai. Based on the warming trends in different physiographic regions, we also found a similar pattern of ecological impacts, where a higher warming region also experienced higher ecological impacts such as changes in water resources, phenology, etc. Lower Hills, Upper Hills, and Mountains experienced higher adverse impacts than the Terai and Siwaliks in the current global warming scenarios.


Introduction
Changes in average temperatures and temperature variation are major key indicators of climate change. The Sixth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, AR6) showed that human activity is the main responsible factor to cause global warming since the mid-twentieth century (IPCC 2021). Due to the continuous increase of greenhouse gases (GHG), the consequent increase in temperature can have Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. severe negative impacts on the environment, socioeconomic systems, and human life. Mountain areas, especially snow-and glacier-dominated regions, are often highly vulnerable to the impacts of climatic change. Glaciers are retreating in the Himalayas, and the mass of snow and ice has been continuously decreasing in the Greenland and Arctic sea over the last four decades (Bolch et al 2012, Tang et al 2014. The importance of snow and glacier-dominated mountains is well known due to their critical role in providing fresh water to large human populations. Snow cover and its associated phenomenon on the higher Himalayas influence water availability in the downstream area, especially during the onset of spring Bengtsson 2004, Barnett et al 2005), which have multiple impacts on agriculture, ecosystems, and hydropower.
The cost of climate change would mostly be borne by poor nations as the regions most likely to be negatively affected fall within the developing world. Combining the latest physical climate models, socioeconomic projections, and economic estimates of past impacts, Lemoine and Kapnick (2016) revealed that future warming could raise the expected economic growth in richer countries while reducing the expected economic growth in poorer countries. Nepal, a mountainous country is a sensible zone in the context of climate change in South Asia due to its fragile ecological system, rugged geophysical structure, and steep slopes, where glaciers are retreating rapidly in the Himalayas and the equilibrium line altitude (ELA) had been shifting upward (Kayastha andHarrison 2008, Khadka et al 2020). In the High Himalaya region, formation, growth, and likely outburst of glacial lakes are phenomena directly related to climate change and deglaciation. Consequently, some signals of negative impacts on river hydrology, agriculture, natural biodiversity, livelihoods, and human health have been observed in recent years (Shrestha and Aryal 2011). As we know the rate of warming in the Himalayas is greater than the global average, which has caused high ecological degradation in the mountainous country Nepal (Kattel 2021). The country has been facing shifts in species ranges and phenology, changes in the relative abundance of species, and water resources for several years (Gaire et al 2014, Lamsal et al 2017, Baniya et al 2018). In recent years, agriculture shifting also has started in the High Mountains. Furthermore, risk factors have increased due to the increased number of pests and invasive species in the Middle Mountains including some parts of the High Mountains (Chaudhary and Bawa 2011).
Previous studies (Shrestha et al 1999, Kattel and Yao 2013, Thakuri et al 2019, Yue et al 2020 reported distinct warming patterns and also revealed that the increasing trend in minimum temperature was negligible in comparison to maximum temperature in almost all parts of the country. However, the trends of climatic warming along east-west and south-north altitudinal gradients and its ecological impacts across the different physiographic regions of Nepal have not been comprehensively explored yet. To address this key issue, here primarily, we analyze the variability and trend of east-west and south-north temperature gradients of Nepal including the four different physiographic regions Terai, Siwaliks, Lower Hills, and Upper Hills, respectively. We then examine the annual and seasonal patterns of warming trends and associated ecological and societal implications in the region. This study provides cause and effect relationships between warming trends and adverse ecological impacts in different physiographic regions from a lower elevation to the High Mountains of Nepal. Further, it elaborates on which sectors have been facing more adverse impacts in those physiographic regions.

Materials and methods
2.1. Brief introduction to the study area Nepal, located on the lap of the East-West extended Himalayas, lies between 26°22′ and 30°27′ N and 80°04′ and 88°12′ E geographically (figure 1). Its climate varies from the tropical climate in the south to the arctic in the north. It has five distinct physiographic regions, such as Terai, Siwaliks, Middle Mountains (consists of Lower and Upper Hills), High Mountains, and the Himalayas, which are also distributed geographically from low to high latitudes, respectively. Terai and Siwaliks regions, which both have tropical to sub-tropical climates, are at the levels of 60-200 m and 200-1000 m, asl, respectively. Natural vegetation in the Terai and Siwliks regions consists of tropical evergreens forests (dominated by Dalbergia sisso and Acacia catechu) and tropical deciduous forests (dominated by Shorea robusta), respectively (Dhital et al 2013a). The other physiographic regions called Lower Hills, Upper Hills, High Mountains, and the Himalayas range from 1000-1700 m, 1700-2800 m, 2800-4000 m, and 4000-8848 m, asl, respectively. Middle Mountains experience sub-tropical climate in valley bottoms whereas warm temperate on valley sides and cool temperate on higher ridges (Karki et al 2016). This region has a high diversity of ecosystems and the vegetation consists of a mixture of many species (pines, oak, rhododendrons, poplars, walnuts, etc). The High Mountains have cool temperate to sub-alpine climates and the vegetation consists of valuable forests including spruce, fir, cypress, juniper, and birch (MoE 2008). The Himalayas experience alpine with a nival climate above the snow line, and have very little vegetation with summer grazing pastures in the lower elevation. The national mean temperature is around 15°C and normally increases from north to south. The highest temperature of the year occurs in May or early June and starts decreasing from October and reaches the minimum in December or January. Average annual rainfall was reported 1500-2500 mm and tends to decrease with elevation above Lower Hills. However, 70-80% of the total annual rainfall occurs during the summer monsoon season (Shrestha 2000, Sharma et al 2021. In general, precipitation decreases from east to west during the summer monsoon season (Pokharel et al 2019).

Data analysis
Monthly maximum temperature data (monthly means of daily maximum temperature) of 76 stations from 1970-2016 were made available from the Department of Hydrology and Meteorology (DHM), Government of Nepal for this study (figure 1). Hereafter, the maximum temperature is simply referred as temperature. The selected stations are distributed almost evenly across the Terai, Siwaliks, Lower Hills, and Upper Hills regions. Both parametric (linear regression) and non-parametric (Mann-Kendall and Sen's slope estimator) methods were used to detect temperature trends in each season and in the annual time scale. Based on DHM, the seasons were classified as pre-monsoon (March-May), monsoon (June-September), post-monsoon (October and November), and winter (December-February). In this study, simple linear regression was used to see the data fluctuations of the individual station. This method is highly acceptable in the case of normally distributed variables and has been used to describe the relationship between one variable with other variables of interest (Mosmann et al 2004). However, we preferred non-parametric tests because these methods do not need any distribution form for the data and also can be used to derive accurate results in comparison to parametric methods (Yue et al 2002, Zhang andLu 2009).
As the non-parametric method is useful to detect trends in meteorological and hydrological time series data (e.g. Gadgil and Dhorde 2005, Sharma et al 2016, Mahmood and Jia 2019), here, we proposed the null hypothesis (H 0 ) for an independent time series data X i (I = 1, 2K, n) distributed identically. Our alternative hypothesis (H 1 ) was that a monotonic trend would exist in X i . We set the null hypothesis H 0 would be rejected at a significant level α if |Z| >Z (1 −α/2), where Z (1 −α/2) value would be the standard normal distribution with a probability of α/2. In this study, we set α = 0.05 significance level. Except for the high mountains, initially, we ran the analysis for annual and seasonal warming trends for the entire country by dividing it into three major parts horizontally: Eastern, Western, and Central. We then evaluated the different physiographic regions along the altitudinal gradient from the lower (Terai) to the higher elevation (Mountains). In this study, risk assessment on ecological implications of warming was assessed based on our results and a review of relevant literature. Further, adverse impacts on each physiographic region were summarized, and a qualitative projection graph (adverse ecological impacts of warming versus elevation) was finally prepared.

Results and discussion
Most stations of the DHM (Nepal) that we selected for our study were recorded either increasing or decreasing temperature trends on the mean annual and seasonal cycle. In figure 2, we have categorized the stations with significant positive temperature trends into 5 groups (0.2-0.4, 0.4-0.6, 0.6-0.8, 0.8-1, and >1°C per decade), however, the stations showing the only normal positive temperature trends are categorized into 6 groups (0.05-0.2, 0.2-0.4, 0.4-0.6, 0.6-0.8, 0.8-1, and >1°C per decade) with different (increasing) bullet sizes. The average temperature trend of all stations in the country was recorded as 0.4°C per decade, which is higher than the Northern Hemisphere and Global average warming trends. However, Shrestha et al (1999) found that the average warming in annual temperature was 0.  The different physiographic regions across the south-north gradients of Nepal showed high variability of increasing or decreasing temperature trends (table 1). Some temperature stations also exhibited variable increasing or decreasing trends in different seasons. In the Terai region, the highest PSPT (67%) and PSSPT (43%) were found during post-monsoon, whereas the highest PSNT (33%) in winter was also recorded. The average temperature trends in annual, winter, pre-monsoon, monsoon, and post-monsoon were recorded 0.1, 0.2, 0, 0.2, and 0.2°C per decade, respectively. In the Siwaliks region, 67-83% of stations were recorded increasing temperature trends, whereas 11-22% of stations were recorded decreasing temperature trends (table 1). The average trends for annual, winter, pre-monsoon, monsoon, and post-monsoon temperatures were measured as 0.23, 0.4, 0.2, 0.3, and 0.12°C per decade, respectively. In the Terai region, the discharge of the non-snow-fed rivers is reported declining continuously during the winter (Manandhar et al 2011, Dhital et al 2013b and the groundwater level is also depleting in most parts of this region, suggesting that such condition gradual temperature warming is occurring. In recent years, more fire events in residential areas are reported by the Nepal Government from the Terai region, which also indicates the increased frequency and duration of droughts in the region. In the Siwalik region alone, water-induced disaster events (flooding, landslide, debris flow) frequently occur every year during the monsoon and post-monsoon periods, so, the impacts of warming are not much noticeable in comparison to natural disasters. However, some ethnic communities (Chepang, Tamang) in this region have been facing more adverse impacts of warming, causing a decrease in agricultural production (maize, millet) for the last 2-3 decades (Piya et al 2013).
Higher warming trends were observed in both Lower Hills and Upper Hills. In the Lower Hills, the highest PSPT was recorded in winter (100%) and Lowest in post-monsoon (73%), whereas 18% PSNT was recorded in post-monsoon (table 1). The average trends in annual, winter, pre-monsoon, monsoon, and post-monsoon temperatures were 0.65, 0.82, 0.7, 0.6, and 0.6°C per decade, respectively. In the Upper Hills, 60-73% of stations were still recorded significant increasing temperature trends, while 7% of stations were recorded significant decreasing temperature trends in the post-monsoon season (table 1). The average trends in annual, winter, premonsoon, monsoon, and post-monsoon temperatures were recorded as 0.6, 0.72, 0.5, 0.4, and 0.64°C per decade, respectively. In the Lower Hills region, relatively higher impacts of warming on water resources are observed. The local community in the region has reported that springs are dried up in many places, which is thought to be associated with enhanced warming in this region. Water releasing rates in small, non-snow-fed rivers in some parts of the region have shown a continuous decreasing trend (Poudel andDuex 2017, Thapa et al 2020), suggesting a prolonged spell of drought is occurring. Unlike in the Lower Hills, phenological changes in plants are noticeable as a result of temperature warming in the Upper Hills. For example, plant species such as rhododendrons have started to flower a month earlier than in the past (Xu et al 2009). A previous study (Shrestha et al 2012) also revealed that the average start of the growing season seems to have advanced by two days per  There is no significant change in the total amount of mean annual precipitation which indicates a decrease in the snowfall and an increase in the erratic pattern of rainfall. Decreased snowfall resulted in water scarcity in high altitude residents. In some places of these regions, local people have noticed that precipitation falls in the form of rain instead of snow which did not happen in the past, causing more soil erosions as well as forming gullies (Fort 2015). The conditions have directly caused the daily lives of the local community. For example, the house roofs which are built with mud especially in Upper Mustang to protect from snow cover, have been impacted by the changing form of snow to water resulting in water leakage from the roof. The grassland cover in the upper region has declined with the gradual replacement of shrubs and other small trees. The loss of grasslands has led the herbivores, especially blue sheep populations to move to the lower region nearby human settlements in search of food, but in the meantime, snow leopard preys have declined with further increase in human-wildlife confrontations . New pests such as mosquitos have already adapted to climate warming in the upper region triggering mosquito-borne diseases such as malaria, dengue, etc (Chaudhary and Bawa 2011). Despite those negative impacts of warming, some positive outcomes have been achieved from climate warming in the region. For example, farmers have modified seasonal cropping calendars, apple cultivation and other new crops plantation have been shifted further upward, and resulted in increased production in both quantity and quality (Manandhar et al 2014).
In the Himalayas, the melting of snow and ice is a natural process. However, new glacial lakes formation, old glacial lakes expansion, and snow avalanches with GLOF are increasing in recent decades (Bajracharya and Mool 2009, Shrestha et al 2017, Chand and Watanabe 2019, which reveal the multiple impacts of increased warming in this region. These phenomena have intensified vulnerability in the lower region, especially in the Lower Hills region due to densely populated areas. In some past GLOF events, human settlements including cattle and cropland, hydropower stations, bridges, and other infrastructures in the downward areas especially in Hilly regions have been partially or fully destroyed, and also human lives were lost nearby stream (Khanal et al 2015, Bajracharya et al 2020). Some previous studies (Manandhar et al 2013, Nepal 2016 showed that discharge on snowfed rivers was found to be increasing in winter and pre-monsoon season, which might be the additional contribution of increasing snow and ice melting. It may have provided some positive feedback on fishery and agriculture activities but will have higher negative consequences in the future. In sum, all physiographic regions were reported more adverse impacts of warming, which are shown in the projection graph ( figure 4). This demonstrates the cause and effect relationship of warming trends with increasing elevation. Though Hilly regions including both Lower Hills and Upper Hills experienced higher ecological impacts than the Siwaliks and Terai regions.

Conclusion and recommendations
The temperature is increasing at a high rate in most parts of Nepal, except for Terai and some parts of the Siwaliks region. The warming trend in the Terai region has not been observed in pre-monsoon. More than 75% of the stations have shown increasing trends at annual and seasonal cycles. Winter season pronounced high warming trend in almost all parts of the country. The warming trend in the Eastern part is higher in comparison to the Central and Western parts of the country. About 70% of PSSPT in Lower Hills and Upper Hills reveals alarming warming trends in these regions. The highest increasing trend of temperature in the Lower Hills is recorded as 0.82°C per decade in winter, which shows that higher elevations may experience lesser degrees of warming trends than the lower elevations mostly above the Siwaliks. Similarly, higher ecological impacts of warming have been reported in the upper region than those in Terai and Siwaliks. As marked ecological impacts are likely to be occurring in the hilly regions implicating water resources in Lower Hills and phenology in Upper Hills. Invasive species and new diseases in those areas are causing more problems for agricultural activities and cattle farming. Transhumant pastoralism is highly impacted in both regions. Changes in snowfall patterns and high ablation rates in the High Mountains have made mountain life more complicated. The increasing trend of snow avalanches with GLOF events has further increased the vulnerability of the downstream population. Those all ongoing changes in the environment have a high probability of ecosystem alteration.
As high as average 0.4°C per decade warming trend of Nepal's temperature observed annually in our study is unusually high and alarms the revaluation of the nation's current climate policies. The contrast of warming across the south-north gradients of the country further suggests that region-specific adaptation strategies are needed to tackle climate change. Both seasonal and annual trends of temperature warming indicate the importance of micro-level evaluation of ecological and societal responses to climate change in Nepal. The national adaptation program should be more focused on ecosystem management. Hopefully, this interdisciplinary research will be beneficial for the scientific community to know the current status of climate change in the mountainous region of the Central Himalayas and also will be fruitful for policymakers to prioritize necessary adaptation strategies for highly influenced regions and sectors.

Data availability statement
The data cannot be made publicly available upon publication because they are owned by a third party and the terms of use prevent public distribution. The data that support the findings of this study are available upon reasonable request from the authors.