Climatic indicators
Changes in average annual temperature (AAT)
Climate change is often manifested as changes in temperature. The greater the fluctuations in temperature, greater will be the uncertainty. The increasing temperature is affecting the availability of water, biodiversity, ecosystem boundaries, distribution and duration of rainfall and content of carbon in the soil (Ives, 2005; Xu et al., 2009; Kotlia and Joshi, 2013; Liang et al., 2013; Kohler et al., 2014). To find out the spatial pattern of exposure to temperature variability, the actual change in average annual temperature (AAT) for a period of 117 years from 1901 to 2017 using the annual temperature records was calculated (Fig. 4). Huge variability is recorded in temperature over BRB in space and time. The AAT was noted maximum 22.6°C during 2016, followed by 22.5°C (2002), and 22.4°C (1999). Whereas, the minimum was observed 20°C in 1917, preceded by 20.3°C (1912), and 20.4°C (1913). Overall, it was noted that BRB has experienced relatively warmer temperatures after the first half of the 20th century.
Spatially, higher temperatures were observed in the LBRB and the UBRB owing to its higher altitudes experience comparatively lower temperatures throughout the year. During the study period, the Firozpur district had the highest temperature with an LTA temperature of 25.4°C, followed by Jalandhar (24.8°C) and Amritsar (24.8°C). In addition, all other areas with high temperatures were part of LBRB, whereas all of the low temperature areas, including Lahaul and Spiti (9°C), Kinnaur (12.7°C) and Kullu (17.5°C) were part of UBRB. A range of 16.4°C was noted between the highest (Firozpur) and lowest (Lahaul & Spiti) temperature districts located at the southern and northern ends of the basin, respectively. The LTA temperature for the entire study area was calculated as 21.2°C.
During the study period of 117 years, a total of 62 years had an AAT of less than LTA and 48 years were recorded with temperatures warmer than LTA. Although the number of years with an AAT less than LTA temperature was more, the overall linear trend indicated an increasing temperature. Annual departures of temperature in the basin from the LTA vary between 1.3°C (2016) and-1.1°C (1917). A huge range of 2.6°C was observed in the AAT between the warmest and the coldest years. The spatial distribution of AAT for the years 1901 and 2017 is presented in Table 2. It was noted that warming is more pronounced in the UBRB. The UBRB was 1.05°C warmer in 2017 as compared to 1901, while the LBRB was 0.37°C warmer than in 1901. The comparative analysis of the beginning year (1901) and the end year (2017) has revealed an increase in the mean temperature of all the districts for calculation of exposure index. It is asserted that, higher the normalised value, greater the exposure to climate change. Hence, the highest values were assigned to the districts with high exposure and vice versa.
Changes in diurnal temperature range (DTR)
The diurnal temperature range (DTR) is another important indicator used in assessing climate change because it provides a more detailed description of the complicated variations in daily maximum and minimum temperatures than the mean surface temperature (Braganza et al., 2004; Qu et al., 2014). DTR was obtained by subtracting the minimum temperature from the maximum temperature. The lesser the fluctuations in DTR, the smaller the probability of getting affected from climatic variability. Therefore, high values were assigned to the districts with the highest exposure to diurnal temperature change. It was observed that DTR was declining during the study period. Along with maximum temperatures, the minimum temperatures also increased throughout the study area.
The range of temperature over the study area increased from north to south, i.e., the UBRB witnessed less temperature variability than the LBRB. Temporal fluctuations in DTR are presented in Fig. 5. Maximum DTR was noted in 1970 with complete spatial unity as all parts of the basin were noted with maximum DTR in the same year. DTR has shown a sudden decline during 1951–1960 and a considerable increase in the following decade. A detailed description of spatio-temporal variations in DTR is presented in Table 3. Although the lowest DTR was noted in 1962, huge variations were observed spatially (Fig. 6). DTR fell the most in the upper basin, including Lahaul and Spiti (-0.22°C), Chamba (-0.21°C), and Kullu (-0.20C°) districts in Himachal Pradesh, and the least in the lower basin, including Firozpur (-0.06°C), Kapurthala (-0.09°C), and Jalandhar (-0.09°C) districts in Punjab. To be taken into consideration is that the upper basin is experiencing temperature increases with much more intensity as compared to the lower basin. The DTR for the entire basin kept on increasing for the first five decades of the twentieth century, and for the second half it had an undulating behaviour. The diurnal temperature range has decreased from 12.5°C in 1901 to 12.4°C in 2000.
The analysis has revealed that Kinnaur (9.3°C), Lahaul & Spiti (10.6°C) and Kullu (10.7°C) had the minimum diurnal temperature range (DTR) in 1901. During the year 2000, DTR was found to be the lowest again in Kinnaur (9.2°C), Lahaul & Spiti (10.3°C) and Kullu (10.5°C) districts. However, the DTR in these districts was relatively lesser than 1901. On the other side, in 1901, the maximum DTR was recorded in Firozpur (14°C), Jalandhar (13.5°C), and Kapurthala (13.5°C) and the DTR decreased to 13.9°C in Firozpur, 13.4°C in Jalandhar, and 13.4°C in Kapurthala during 2000. The highlight was that, the maximum fall in DTR has been noted in the areas of upper basin including Lahaul & Spiti (-0.22), Chamba (-0.21), and Kullu (-0.20) districts of Himachal Pradesh and minimum fall has been noted in the lower basin in Firozpur (-0.06), Kapurthala (-0.09) and Jalandhar (-0.09) districts of Punjab.
Changes in average annual rainfall
Another manifestation of climate change is the pattern of quantum and the distribution of rainfall. Rainfall is also one of the main indicators in the studies of climate change impacts that influence the hydrologic system and agriculture of a region. An analysis of the distribution of rainfall over a region, both temporally and spatially, provides a better understanding of climatic variability. The average annual rainfall (AAR) for a period of 117 years (1901–2017) recorded in the entire Beas River basin is presented in Fig. 7. The long-term average annual rainfall (LTA) noted was 743.4 mm. The analysis of the AAR revealed that the minimum rainfall was 400.7 mm during the year 1918, preceded by 428 mm (1952), and 461.3 mm (1987). On the other hand, maximum rainfall was 1166.4 mm in 1976, followed by 1163.9 mm (1978), and 1125 mm.
During the study period of 117 years, a total of 66 years received rainfall less than the long-term average, whereas 51 years were observed with more rainfall than the LTA. Although the number of years with lesser rainfall than LTA was greater than those received more rainfall than LTA, the overall linear trend in rainfall was upward. Annual rainfall departures from the LTA in the basin range from − 346.9 mm in 1918 to 419.1 mm in 1976. During 1901–1950, a total of 35 years received rainfall less than LTA, as compared to the second half, that received lesser rainfall only in 27 years. A considerable increase in rainfall was observed during latter half in the 20th century. Moreover, it is also noted that the first quarter of the twenty-first century has experienced a comparatively higher amount of downpour than the first quarter of the twentieth century.
Spatially, Kullu district of UBRB has received the maximum amount of rainfall during last 117 years with an LTA rainfall of 966.4 mm, followed by Shimla (938.9 mm) and Kinnaur (918.6 mm). In addition, all other high rainfall districts were part of UBRB. whereas all of the low rainfall receiving districts, including Firozpur (303.2 mm), Amritsar (475.2 mm) and Kapurthala (504.4 mm) were part of the LBRB. A range of 663.1 mm was noted between the highest (Kullu) and lowest (Firozpur) rainfall receiving districts located at the northern and southern ends of the basin, respectively. Further, it was observed that northern, north-eastern, and south-eastern parts of the study area receive much rainfall, whereas the north-western, southern, and south-western parts receive a comparatively lesser amount of rainfall. The maximum variation in annual rainfall was observed in lower parts of the basin. The average annual rainfall increased during the second half in all the districts except Kinnaur. Maximum increase was observed in the low-lying areas of Punjab and the rainfall increase kept on decreasing towards the north and the hilly parts of the UBRB (Kumar and Rao, 2021).
Changes in monsoon season rainfall
The Indian summer monsoon rainfall is controlled mainly by the south-west monsoon. It occurs from June to September every year and plays an important role in the agricultural production in India. The study and prediction of monsoon rainfall variability has been a matter of great importance to both society and the scientific community because a deficit or excess in summer monsoon rainfall in a year leads to drought or flood disasters respectively, causing great impacts on the agriculture and economic activities of the region (Webster et al., 1998). Similarly, the maximum amount of rainfall in the study area is received during the south-west monsoon season. Hence, the monsoon season rainfall has been taken into consideration in addition to the average annual rainfall.
During monsoon season, Amritsar, Firozpur, Jalandhar, and Kapurthala districts were noted with significant increases in rainfall at a rate of 1.2, 1, and 0.8 mm/year respectively. In addition, Chamba, Hamirpur, Kangra, Una, Gurdaspur, and Hoshiarpur districts were also observed with increasing trends, but the trends were not statistically significant. Moreover, negative trends were also observed at some locations during the monsoon season. Although Kinnaur, Kullu, Lahaul & Spiti, Mandi, and Una districts had negative trends in rainfall, these trends were not statistically significant. Overall, huge spatial variability was noted during the monsoon season. The southern and south-western parts of BRB have shown a statistically significant increase in rainfall during 1901–2017.
Hazard specific indicators
Area prone to floods
The hazard event is not the sole driver of risk because the severity of the impacts depends strongly on the level of exposure of societies and socio-ecological systems to such events (Alford, 1992; UNISDR, 2004; Birkmann, 2006). Frequent occurrences of hazards such as floods are becoming common features in the mountainous and plain regions of northern India. Furthermore, the growing population and the expansion of human activities on fragile land trigger disasters in the lower Himalayan region. The larger the area prone to flood events, the greater the exposure. The occurrence of floods is limited to the rainy season when almost 80% of the annual rainfall is received. It was noted that, the exposure to flood events in BRB is primarily restricted to lower regions and the upper regions does not face extensive flooding. Although, the amount of annual rainfall received by UBRB is more than double the amount received by LBRB, the mountainous relief and steep slopes provide suitable topographic conditions for rapid runoff. The analysis of spatial distribution of floods in the LBRB highlighted that Jalandhar district has the highest percentage of area prone to floods followed by Amritsar and Hoshiarpur. All these three districts have more than 80% area prone to floods.
Area prone to landslides
Landslides cause a large-scale disruption of natural resources, economic valuables, and human lives. With particular reference to the Himalaya, a large number of landslides occur every year, causing extensive damage to human lives, properties, and natural resources (ADB, 2010; Prasad et al., 2016). The Himalayan ranges are formed of tectonically active younger geological formations and the exposure of these juvenile and not so stable steep slopes in various Himalayan ranges, has increased at a rapid rate recently due to activities like deforestation, road cutting, terracing, etc. Another hazard in the basin is the occurrences of landslides, however, the incidences of damages caused due to landslide events are limited only to UBRB.
The classification of intensity of landslide and levels of risk is adopted from the BMTPC vulnerability atlas. The classification has four levels of risk such as severe to very high, high, moderate to low and Unlikely (Fig. 8). For calculation of exposure index, the first two levels of risk i.e., severe to very high and high are taken into account. The hilly and mountainous relief of Himachal Pradesh is liable to suffer landslides during south-west monsoon season. Moreover, such events can also occur due to high intensity earthquakes. It was noted that the high-altitude districts with steep slopes are more prone and exposed to landslides, particularly during the rainy season. The maximum area prone to landslides lie in the districts of Lahaul and Spiti, Chamba and Kinnaur. Whereas, the highest percentage of the total population prone to landslides resides in the districts of Kangra, Shimla and Mandi (Table 4). Around 13% population in Kangra district is prone to landslides. While the population in the lower basin does not face landslide hazard events.
Area with slope greater than 25 degrees
Slope is a hazard-specific indicator that determines the sensitivity of a region. Various climate-dependent parameters affecting the sub-surface hydrology led to slope instability, which may result in landslide activity (Dehn et al., 2000). Therefore, slope stability is one of the vital indicators primarily in the hilly and mountainous regions for assessing the exposure to hydro-meteorological hazards (Dijkstra and Dixon, 2010). Steep topographical features imply lack of availability of flat land, instability, and inaccessibility. These areas are more susceptible of being adversely affected due to changes in the climate. Hypsometric analysis was carried out to measure the topographic area-elevation relationship and slope distribution in the study area. Advanced Land Observing Satellite (ALOS) Global Digital Surface Model-2018 (ALOS World 3D-30m (AW3D30) Version 2.1) was used to generate contours and derive slope. The slope in degrees was divided into six categories (Fig. 9) such as very gentle (less than 5), gentle (5–10), moderate (10–15), moderately steep (15–25), steep (25–35) and very steep (above 35).
The BRB is lowest in the ‘Bet’ areas of the alluvial plains and highest near its headwaters with elevations ranging between 166 m in the plains to more than 6500 m in the Great Himalayan range highlighting the great heterogeneity in physiographical characteristics. Around 55% of the basin is occupied by areas with elevation less than 1000 m. As far as slope is concerned, BRB has huge slope variations ranging from very gentle to very steep. The steep to very steep slopes is the chief characteristic feature of the UBRB, while the LBRB has very gentle to gentle slopes. It is noted that, the slope gradient in the study area keeps on declining as one moves from north-east to north-west. Moreover, the percent area under slope more than 25° is greater in the districts of UBRB. In Lahaul & Spiti and Kullu more than 50% of district area has slope greater than 25°. For the analysis we have taken into account the district area under steep and very steep slopes only. The steep topographic environment and monsoon climate combine to produce landslide problems in the Himalaya. Therefore, areas with more than 25-degree slope were considered highly sensitive regions to climate change.
Area prone to wind damage
High wind is a component of weather that can pose many threats to life and property (Adelekan, 2012). Wind speed and turbulence intensity over mountainous terrain with different topographical characteristics, such as escarpments, cliffs, ridges, and hills, are quite different from those over flat terrain. The IPCC’s assessment highlighted the likely increase in the frequency and intensity of extreme weather events related to temperature, wind, and rain, consequent to global climate change. The occurrence of more extreme weather events, together with rapid and unplanned growth, poor environmental management, and poor socioeconomic conditions, has been largely responsible for the increasing vulnerability of societies to natural disasters.
The classification of wind speed and levels of risk is adopted from the BMTPC vulnerability atlas (Fig. 10). The classification has four levels of damage risk such as low, moderate, high and very high damage risk zones depending upon the nature of material used for construction of the buildings. For calculation of exposure index, we have taken into account the first two levels of risk i.e., very high damage risk zone (wind speed 50 m/s) and high damage risk zone (wind speed 47 m/s). In the study area, the UBRB has low exposure to wind-specific hazards while, most of the LBRB lies in the high to very high damage risk zone. Districts like Amritsar, Jalandhar and Kapurthala fall entirely in the very high wind damage risk zone while, other districts of Firozpur, Gurdaspur and Hoshiarpur also has considerably large area of the district in the category of high to very high damage risk zones.
Exposure Index (EI)
The normalized values for individual exposure indicators and the cumulative exposure for all the indicators are given in Table 5. It is noted that the districts of the UBRB, including Chamba, Kangra, Kullu, Kinnaur, and Lahaul & Spiti, were highly exposed to variability of temperature. While the districts of LBRB such as Amritsar, Firozpur, Hoshiarpur, Jalandhar, and Kapurthala were least exposed. Other districts, namely Hamirpur, Mandi, Shimla, Gurdaspur, and Una, were moderately exposed to temperature variability. Both maximum and minimum temperatures increased during the study period as a result the diurnal temperature range decreased from 12.54°C in 1901 to 12.40°C in 2017. A positive relationship was established between DTR and exposure to climate change. Districts of Chamba, Kangra, Lahaul & Spiti, and Gurdaspur recorded higher variations in DTR and, hence, were highly exposed. On the other hand, Kinnaur, Shimla, Jalandhar, Kapurthala, and Firozpur districts were least exposed to changes in DTR. In context to annual and monsoon rainfall, the spatial distribution of exposure is ironic to temperature as high exposure was noted in the districts of LBRB including Amritsar, Kapurthala, Gurdaspur, Hoshiarpur, Jalandhar, Una and Firozpur.
A positive relationship was established among various hydro-meteorological hazards and exposure because more the exposure to hazard events, greater the loss to life and property. Earthquakes, landslides, cloudbursts, avalanches, flash floods, and other natural disasters are common in UBRB. On the other hand, LBRB is particularly exposed to annual flood events, winds and dust storms. The normalized values for flood exposure were highest in Amritsar, Gurdaspur, Hoshiarpur, Jalandhar, and Kapurthala, while UBRB has no exposure to such events. Similarly, damage due to wind hazards was limited to the southern parts of the basin. Amritsar, Firozpur, Gurdaspur, Hoshiarpur, and Jalandhar were highly exposed to wind hazards. Whereas, the mountainous and hilly regions of UBRB were found to be completely devoid of any damage due to furious winds because the mountainous landscape does not allow the free movement of winds.
All the indicators were assigned equal weights to derive the cumulative values of exposure using normalized values of different indicators using a simple arithmetic mean equation. High exposure was recorded in the UBRB for the indicators of change in average annual temperature and change in the diurnal temperature range. While, the indicators of change in average annual rainfall, change in monsoon season rainfall, and percentage of people at risk of landslides showed moderate exposure. Furthermore, due to steep slopes and high altitudes, UBRB had very little to no exposure to annual flood events and wind damage. Overall, minimum and maximum exposure to climate change and hydrometeorological hazard evets were recorded in the districts of Kinnaur and Chamba respectively.
The lower basin is highly exposed to changes in average annual rainfall, changes in monsoon season rainfall, areas prone to flood hazards, and district areas prone to high wind hazards. It is interesting to know that, despite being highly exposed to individual indicators, no district has high exposure in cumulative climate change exposure index. The districts were either moderate or low exposure. Moderate exposure was noted in the districts of Chamba, Kangra, Kullu, Lahaul & Spiti, Mandi, Amritsar, Gurdaspur, Hoshiarpur, Jalandhar and Kapurthala. Low exposure was observed in Hamirpur, Kinnaur, Shimla, Una, and Firozpur districts of the study area.