- Email: [email protected]
Assessment of 2010 flash flood causes and associated damages in Dir Valley, Khyber Pakhtunkhwa Pakistan
Author’s Accepted Manuscript Assessment of 2010 Flash Flood Causes and Associated Damages in Dir Valley, Khyber Pakhtunkhwa Pakistan Shakeel Mahmood, Amin ul Haq Khan, Saif Ullah www.elsevier.com/locate/ijdr
PII: DOI: Reference:
S2212-4209(15)30074-1 http://dx.doi.org/10.1016/j.ijdrr.2016.02.009 IJDRR322
To appear in: International Journal of Disaster Risk Reduction Received date: 18 September 2015 Revised date: 26 February 2016 Accepted date: 28 February 2016 Cite this article as: Shakeel Mahmood, Amin ul Haq Khan and Saif Ullah, Assessment of 2010 Flash Flood Causes and Associated Damages in Dir Valley, Khyber Pakhtunkhwa Pakistan, International Journal of Disaster Risk Reduction, http://dx.doi.org/10.1016/j.ijdrr.2016.02.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Assessment of 2010 Flash Flood Causes and Associated Damages in Dir Valley, Khyber Pakhtunkhwa Pakistan
Shakeel Mahmood1, 2, Amin ul Haq Khan3, Saif Ullah4 1
Department of Geography, University of Peshawar, KP, Pakistan Department of Geography, Government College University Lahore, Pakistan Email: [email protected], [email protected] 3 Sustainable Development Study Center, Government College University Lahore, Pakistan 4 Department of Mathematics, Government College University Lahore, Pakistan Email: [email protected] 2
Abstract This paper identifies the underlying causes and damages caused by 2010 flash flood, which is considered one of the worst hydrological disasters in the country (Pakistan). Questionnaire based survey and structured interviews were conducted to investigate physical and economic damages in the selected case study Dir Valley. Global Positioning System (GPS) survey was also conducted to acquire location of damages and Geographical Information System (GIS) is used to visualize land use, land cover and damage data. Buffer analysis is used to delineate the effected strata on both sides of Panjkora River to assess the flood damages. Climate change phenomenon along with peculiar topography of the region are considered as the underlying causes as monsoon winds interacted with the jet stream abnormally causing unprecedented high intensity rainfall in the valley. The steep topography of the area caused rain water to accumulate rapidly in the Panjkora River, overpowering the withholding capacity of the river. The phenomenon resulted in severe flash flood which annihilated whatever came in its way. Upper zone in the target area is badly affected by flood with maximum damaged houses and human causalities followed by lower zone with high number of damaged bridges. This study will bring the attention of disaster management and other relevant authorities to focus on flood risk reduction by enhancing the retention capability of watershed in upstream areas that will reduce the risk in low lying areas. Key Words: Flash Flood; Causes; Damages; Risk Reduction; GPS; GIS. 1 Introduction For quite some time, disasters around the World are on the rise, not so much due to shorter recurrence of natural extreme events, but more so due to spatial concomitant of hazards with rising human vulnerabilities across the globe. Disasters rising trends, especially those of flash floods can be commonly witnessed in most recent disaster related literature. Natural disasters are increasing tremendously over the earth surface disrupting physical and cultural environment (Fendler, 2008). Floods are the serious socio-natural disaster (Zhang et al., 2002; Krausmann and Mushtaq, 2008) affecting social and economic aspects of life (Wang et al., 1995). Changes in meteorological conditions and land-use pattern have increased floods frequency in many parts of the World (Lehner et al., 2006). Flash flood is hydro-meteorological hazard (Pande, 2010) and occurs after five or six hours of heavy rain in any watershed with high level of destructive discharge (Zhou et al., 2000; Korutny and Kichigina, 2006; Ruin et al., 2008). Its occurrence often associated with orographic effect augmenting violent weather conditions and steep gradient promoting rapid collection of stream flow (Rossa et al., 2010; Gaurav, 2011). Pande (2010) has investigated that cloudburst, breaching of landslides dams, and glacial lakes outburst are factors of flash floods genesis. It has potential to transports large mass of
regolith in the dead channels resulting massive erosion. Mazzarana (2009) expressed that there are scientific evidences of extreme precipitation events which entails that extreme flood events might become more frequent in different regions of the World. Hydrological processes in mountains are strongly influenced by temperature, precipitation, vegetation cover, and gradient (Groisman et al., 2004; Tariq and Giesen, 2011). Mountains; the water towers roughly cover 27% of the plant’s earth surface and draining water through the local drainage system (Viviroli et al., 2003). Areas with altitude ranging from 2000m to 4000m are high risk zones because of steep topography, high stream density, and heavy rain (Viviroli and Weingartner, 2004). The potential for flash flood casualties and damages has been increasing because of development in the proximity of rivers and inappropriate land uses (Groisman et al., 2004). The study area; Upper Dir is a mountainous area with high drainage density, steep topography, and sometimes receives heavy rain because of extra-ordinary activity of monsoon or cloudburst. In the year 2003, flash flood was generated by cloudburst, while in 2010 heavy monsoon rain was the main flood generating factor in the entire valley, inflected heavy losses on life, properties of the local population and infrastructure. This paper comprises of five sections. The first section is regarding introduction, second section describes the geographical conditions of the study area, third section elaborates the methodology of the research work, fourth section comprises of results and discussion and last section concludes the study. 2
Geographical Conditions of the Study Area
The study area; Dir valley is situated in the North-West of Pakistan and about 180Km from the capital city Islamabad. It extends from 35º 04' to 35º 46' N-latitude and 71º 32' to 72º 22' E-longitude. It is bounded from the North and North-West by district Chitral, from the East by Swat valley, from the South-West by Afghanistan, and in South it has common outlet of Panjkora River (Figure 1). It is bounded from all sides by mountains except the South-East with total geographical area of 3,699Km2 (Government of Pakistan (GoP), 1999). The total population was 575,000 as enumerated in the 1998 census with an increase of 58.7% since 1981. The average annual growth rate was 2.8% during this period and population density was 156persons/Km2. The estimated population till 2014 is 805, 000 (Government of Pakistan (GoP), 2014). The elevation of study area varies from 5577m in the North to 844m in the South. The length of the valley is 120Km with average gradient of 2.1%. Physically, Hinduraj Mountains, the off-shoots of Hindukush Mountains act as watershed between Dir valley and the surrounding districts. These mountains have snow covered peaks, accumulated snow in the valleys and glaciers in the North-East, where elevation exceeds 5000m. River Panjkora originates from these mountains and its tributaries are Rivers Kohistan, Barawal, Usherai and Dir (Figure 2). Upper zone is narrow with deep gorge and lower zone is broad. Average width of the channel is 0.7Km. River Panjkora is the main tributary of River Swat. Most of the area is covered by temperate forests in higher altitude while low altitude areas are covered by degraded sub-tropical scrub forest (Champion et al., 1965). These forests are economically important in generating revenue, source of timber, and fire wood. Furniture industry of the study area depends on these forests. Agriculture is the main economic activity of the people and terraced farming is common, and cattle, sheep and goat rearing is common as well. Soil and climate are favorable for the cultivation of crops including wheat, maize, rice, onion, potato, tomato, and fruits; apple, apricot, pears, walnuts, and citrus fruits (GoP, 1999).
3
Data Acquisition and Analysis Methods
Primary and secondary data have been collected to achieve the objectives of this research work. Primary data were collected through questionnaire survey, interviews, personal observation and GPS. Prior to field survey, frequent field visits and meeting with local people were arranged to estimate the extent of flood damage and the identification of most affected sites. Detail micro-level investigation and analysis of the flood affected areas were carried out based on initial field visits and eleven sample sites; Sharingal, Barikot, Kalkot, Palam, Tarpatar, Barawal Bandai, Wari, Akhagram, Khal, Turmong and Baroon (Figure 7), were selected randomly from the entire valley along the river. Questionnaire based survey was conducted in all the sample sites using random sampling techniques within 200m buffer zone on both sides of the river. Spatial sampling technique was used to acquire the geo-location of damaged structures; buildings, bridges, road and retaining walls in the entire valley using GPS. Similarly elevation data along the channel was acquired using GPS to calculate gradient (which is the change in elevation over the distance covered along the surface) of the channel using the following formula.
Fig. 2. Gradient of the Channel = Gradient of the Channel
Difference in height between two po int s 100 . Horizontal dis tan ce between two po int s
y 2 y1 rise 100 100 , x2 x1 run
where “rise” is the change in elevation between selected points and “run” is the horizontal distance between them (Graph based on the above formula is given in Figure 5).
Fig. 1. Location of Upper Dir and different landuses with villages
Separate structured questionnaires were designed for households head and concerned government department officials. Questionnaire survey was conducted directly in the sample sites by asking face to face questions from 339 flood affected people in local language “Pashto”. They were mainly asked regarding the causes and damages incurred by flood, nature of damages, and estimated rehabilitation cost. This constituted approximately 65% of the people living in the selected buffer zone. Concerned officials were also interviewed to explore the causes of flood, extent of damages, and estimated rehabilitation cost in “million Pakistan Rupees (PKR)”. Damage estimation cost of private property including houses, buildings, land loss, crops and livestock collected in field survey according to community perception. The average value for rehabilitation cost was calculated using computer based statistical program MS Office 2007. The estimated rehabilitation cost of damaged water supply schemes collected from Public Health Department, District Upper Dir, Khyber Pukhtoon Khawa (KPK), and damaged bridges, retaining walls and roads estimated rehabilitation cost is collected from Construction and Works Department, District Upper Dir, KPK. Secondary data collected from different concerned government departments. Rainfall and temperature data were collected from Pakistan Meteorology Department, Regional Office Peshawar. Population statistics were collected from Population Census Organization, Islamabad. Topographic map of the study area was collected from Survey of Pakistan (SoP), Islamabad to prepare land use and land cover maps using ArcGIS 9.3. Digital Elevation Model (DEM) with 30m resolution was downloaded from online geo-database of United State Geological Survey (USGS). Hydrologic analysis performed on DEM using ArcHydro to analyze hydro-geometric network of the study area. Similarly, Buffer analysis was applied to highlight the structures and settlements in 200m buffer zone. Finally, results were presented in the form of maps with 360 dots per inch (dpi) and graphs. 4 Results and Discussions From the obtained data and results, assessments of causes and damages of flash flood 2010 have been made. 4.1 Causes Assessment Pakistan is prone to floods because of its hydro-meteorological conditions (Atta-ur-Rahman and Khan, 2010), and every year different nature of floods have been occurring including monsoon floods, flash floods, urban floods, and coastal floods (Zaidi, 2011). However, excessive rainfall has been the main cause of many flood events (Hunter et al., 2005; Delrieu et al., 2005). In the year 2010, disastrous flood generated by the extra-ordinary activity of monsoon rainfall, melting of snow and glaciers in Pakistan (Atta-ur-Rahman and Khan, 2011; Tariq and Giesen, 2011; Atta-urRahman and Khan, 2013). Meteorological setup of Pakistan was unique in the last week of July, 2010 in the context of spatial coverage, intensity and duration of clouds compared to the past meteorological events, because of the interaction of monsoon winds and westerlies over the North-West of Pakistan. This caused unprecedented rain in the entire country, particularly in the North-western and western parts of Pakistan. This rainfall became the major factor of high surface run-off while steep topography accelerated the accumulation of rain water in the channel and produced flash flood in Rivers Panjkora
Fig. 3. Upper Dir Drainage Pattern and Digital Elevation Model developed from 30m SRTM data.
Fig. 4. Dir Valley, Mean Monthly Rainfall of Past 40 Years & Mean Monthly Rainfall of 2010. Source: PMD 2011 and Swat. This generated historic flood peak in River Kabul inundated most of the Peshawar Plain. Similarly, this highest peak became the flood generating factor in Indus River. Normally, duration of previous flood peaks ranged from 12 to 24 hours, while the duration of this flood was 110 hours (Asrar-ul-Haq and Zaidi, 2011; Syvitski and Brakenridge, 2012). Dir Valley is located in the NorthWestern Mountains, where the mentioned unprecedented rain generated destructive flash flood. This flood brought huge damage to social, economic and physical setup of the study area. The target area has long winter and short summer seasons. December to February, are the coldest months with mean maximum and minimum temperature in January is 11ºC and -2ºC, respectively. While, July and August are the hottest months with mean maximum and minimum temperature of 32º C and 16.3º C, respectively. In summer, the source of rainfall is local relief and monsoon winds, while Western Depressions have been the source of rainfall in winter. There is variation in annual rainfall since 1967.
Upper Zone
Middle Zone Lower Zone
Fig. 5. Panjkora River Gradient based on elevation data acquired through GPS. Source: Field Survey, 2011
Fig. 6. Community Perceptions Regarding Flood Causes Source: Field Survey 2011
Lowest rainfall is recorded in 1971 and 2001, while highest in 1979, 1986, and then in 2010. Analysis revealed that in the last four decades there is positive shift in rainfall particularly in July. In the year 2010 increase in mean monthly rainfall has taken place from May to September. The normal for the months of May, June, July and August is 89mm, 56mm, 156 and 149mm while the rainfall in the year 2010 for the same months is 110mm, 76mm, 393mm and 285mm respectively (Figure 4). In July, 2010 rainfall was maximum because of the heavy downpour from 27th to 30th July with intensity of 98mm/day, while on 29th July its intensity was 6.5mm/hour (Pakistan Meteorological Department (PMD), 2011). This high intensity rain became the main flash flood generating factor in the entire valley which was never ever experienced by this valley in the past record. July and August are the hottest months with maximum temperature higher than 32ºC resulting melting of snow and glaciers in the upper zone. This melt water is the main sources of discharge in summer season. The high intensity rain has accelerated melting. Thus the melt water has contributed in the genesis of high discharge which was beyond the capacity of the channel and generated massive flash flood with disastrous effect on human life, property and infrastructure. Topography of the target area is mountainous. River valley is narrow and steep in upper zone, while slope is decreasing towards the lower zone and width is increases. Gradient of channel varies from 3% in upper zone to 2.5% in middle zone and 2.1% in lower zone. The average gradient of the entire channel is 0.8% which has intensified the flash flood (Figure 5). There is a system of interconnected sub-watersheds and these characters of the study area have caused rapid collection of rain and melt water in the River Panjkora and resulted in disastrous flash flood. The factor of deforestation is also very important, because forest cleared areas accelerated the surface run-off and sheet erosion which also intensified the potential for flood. People are cutting trees as fuel sources because there is no other source of fuel. Thus heavy rainfall and steep topography have resulted flash flood which was further intensified by melt water and deforestation. In the target area, people were asked regarding flood generating factors because local people have indigenous knowledge. In Palam and Tarpatar 100% respondent view point was, “heavy monsoon
rainfall” while in rest of the sample sites more than 60 percent respondents view point was “heavy monsoon rainfall, and melting of snow and glacier” resulted this disastrous flood (Figure 6). According to elders of the community and certain officials; the principal cause of flood was rain-water. They also said that forest cover has been decreasing which has increased surface run-off and sheet erosion. It was observed in the target area that local people has people has cleared forest for the purpose of fuel and construction of houses. Most of the areas located on slopes are exposed to the action of snowfall and rainfall. In the winter season frost action weathered the exposed rocks and increased the fragility of the slopes. During monsoon seasons high surface runoff has resulted massive erosion and rapid collection of rain water from barren surfaces in torrents and streams which has generated high flood in the River Panjkora River. 4.2
Damages Assessment
Analysis revealed that flash flood in the year 2010 seriously affected human and animal lives, and undercut foundations of buildings, retaining walls and bridges, eroded topsoil, and changed river morphology. Within eleven selected sample sites, damages were assessed through extensive field surveys, observations and interviews. Historically, population clusters and human activities are concentrated in the proximity of river. Mostly people are living in mud-brick and mud-stone houses in the proximity of river which are less resistant to flood. Flood damaged large number of houses in the upper zone; Kalkot, Barikot, Sharingal, Palam, and Tarpatar where stream density is high and gradient is more than 3%. In all sample sites completely and partially damaged houses were 422 and 45 with estimated economic loss of 295.4 and 18 million PKR, respectively. The number of damaged houses is 93% of the total in the selected buffer zone. Kalkot was severely affected village in terms of damaged houses, because of abrupt change in gradient and hill torrents draining into the main channel from both sides. Here, cluster of houses were built in river proximity, 156 houses were completely damaged by flood with estimated economic loss of 109.2 million PKR. Damaged shops and restaurants were located in lower zone. The total number of damaged shops was 40 with estimated economic loss of 40 million PKR. Akhagram is the most affected site in terms of damaged shops. At this location shops were built to provide daily use goods where the channel is narrowest in the entire reach accompanied by interlocking spurs. This has generated high pressure resisted by spurs; the projecting part of the mountains into the river. This fluvial landform forms curves in the channel with outer and inner bank. The outer bank of the curve is backed by spur projecting into the channel that has diverted flow to the inner bank where shops were located. This surge shifted the flow with high hydraulic pressure towards inner bank, busting the concrete buildings and demolished everything there. Damaged buildings are located over the inner bank because outer bank has steep slope with no construction. Two bridges were also completely damaged by this flow at the same location. Distribution of damaged shops is concentrated in lower zone, because these structures were built in river proximity. Three restaurants were completely damaged with estimated loss of 27 million PKR. In the upper zone damaged buildings are located on both sides of the channel. In lower zone human settlements are dispersed and commercial buildings have been constructed on the left bank of river along Dir-Chitral Road. People have been living in the river proximity on both sides with main population clusters including the selected sample sites. Panjkora drainage system comprises of interconnected sub-watershed with perennial streams. In the upper zone, sub-watershed has maximum gradient and high drainage density,
while in lower zone the drainage density is low. High drainage density has made hindrance in the movement of people from one village to other situated across the river as well as along the river. Bridges and culverts overcome the problem accessibility. Flood damaged most of these structures. Partially and completely damaged bridges were 36 with estimated cost of 125.6 million PKR. Analysis revealed that highest recorded discharge, low height of bridges, high flow energy at bridging points, tree logs and timber have damaged these structures (Figure 7). Road and retaining walls were damaged by both the main river and tributary streams. In lower zone damages to road were high because of nearness to river level and maximum hydraulic pressure exerted by highest discharge. In upper zone, two factors are responsible for damages to road and retaining wall. First is the narrowness of channel and second is interlocking spurs. Maximum energy in the narrow parts of channel resulted high level of damages to the road and other structures. Similarly, the spurs have shifted the flow to the inner bank of the channel that undercut base of the retaining wall and road. In upper zone, this character of the valley has also resulted landslide that further intensified the flood. Total estimated economic loss resulted by damaged roads and retaining wall was 410 million and 180 million PKR, respectively. People are earning their livelihood mostly from agricultural activities practiced as terraced farming. Fields in the proximity of river were damaged by flash flood. In the upper zone, agriculture land in the proximity of river was eroded by flood while, in the lower zone land and crops were damaged by the temporary deposition of sand and pebbles. The layer of sand and pebbles over the agriculture land is ranging from 0.8m to 2m. The average width of the flood plain in lower zone is 1.5km on right side and 0.5km is on left side of the river. Over the right bank the alluvium was 5m deep with undulating surface towards river. River has shifted the course towards the agriculture land on right side by eroding 250m strip of 5m thick agriculture land. In Barikot 2.3ha, Kalkot 1.9ha, and Palam1.2ha of agriculture land has damaged by the flood causing economic loss of 4.6, 3.8 and 2.4 million PKR, respectively. In rest of the areas, damaged agriculture land was less than 1ha. Total economic loss to agriculture land and crops was estimated as 19.54 million PKR. In sample sites of Kalkot and Barikot, 161 livestock were lost with estimated cost of 2.468 million PKR. In Barikot 60 livestock including cow, sheep, and goat were taken by flood with estimated cost of one million PKR followed by Kalkot with estimated loss of 0.78 million PKR (Figure 8). Although the study area receives rainfall throughout the year but it is not sufficient for agricultural activities. River Panjkora is an important source of irrigation. Water is diverted in narrow channels to irrigate fields in river proximity. These channels are lined and unlined with 1m width and 0.5m depth. Flood has damaged the heads of these channels in all sites with estimated economic loss of 180 million PKR. Water supply schemes were also severely affected by flood. Most of the damages were concentrated in the upper zone with total estimated cost of 57.5 million PKR. The heads of water supply schemes are in the upstream of hill torrents. Destructive flow has completely washed away the heads of these water supply schemes
Fig. 7. Distribution of damaged houses, shops, restaurants, road and bridges developed in ArcGIS from GPS data.
Flood has been very lethal for human and animal lives, because it causes human injuries and fatalities. Most of the human causalities have taken place in the upper zone where gradient is more than 3%. In Palam and Tarpatar, impact of flood on human life was highest with total death cases of 25 and 20, respectively. Most of the victims were working in their fields in the proximity of river. They were not aware because of the lack of any warning or emergency system. In the study area, total fatalities and injuries were 66 and 69, respectively (Figure 9). Analysis revealed that 2010 flash flood adversely affected the study area and resulted severe damages to infrastructure, buildings, agriculture and human life. In all sample sites, completely and partially damaged houses were 467 with estimated cost of 313.4 million PKR, completely and partially damaged bridges were 30 and 6, respectively with estimated rehabilitation cost of 125.6 million PKR; damaged roads and retaining walls were 4.1Km and 6Km with estimated rehabilitation cost of 410 and 180 million PKR, respectively; total length of damaged irrigation channels was 36Km with estimated rehabilitation cost of 180 million PKR. Similarly, 25 water supply schemes were completely damaged with estimated rehabilitation cost of 57.5 million PKR. The number of damaged shops and restaurants were 40 and 3 with estimated economic loss of 40 and 27 million PKR, respectively. Agriculture was also badly affected by flood with total estimated economic loss of 23 million PKR which was simply unaffordable for subsistence agriculture. The most affected sector was infrastructure with damaged roads, retaining walls, water supply schemes, and bridges. In the study area most affected sites are Kalkot, Sharingal and Barikot with total estimated economic damages in selected sectors were 244.35, 233.43, and 178.8 million PKR, respectively—located in the upper zone. In Akhagram the total estimated cost of damages was 171 million PKR with highest economic loss in lower zone while, Wari is the least affected site because human activities and building are away from the river. The estimated economic loss is 39.2 million PKR. Similarly in Khal, Baroon and Tormang losses are less because people are living away from the river with few buildings in river proximity. Total estimated economic loss caused by the 2010 flood in selected sectors of the study area is more than 1350 million PKR, which is a significant loss to the very weak provincial and national financial economy. Geographically the terrain of target area is mountainous with narrow and steep sub-valleys. These valleys are more in the upper zone because of the high density of streams as compare to lower zone. Fast flowing streams and torrents drain into the main channel from both sides. This character of the upper zone has produced disastrous high flow in the channel. Local people have built their house in torrents, stream and main river proximity to make water accessible because they have never experience such high discharge in past like in 2010. In lower zone people have built restaurants and departmental stores with foundation in or near channel. This level of exposure has made the human life, property, and infrastructure more vulnerable to flash flood and resulted huge physical and economic damages in the study area.
Fig. 8. Site-wise economic loss and gradient of the Panjkora channel. Source: Field Survey, 2011
Fig. 9. Human Fatalities and Injuries Source: Field Survey, 2011
5
Conclusions
This study concludes that flood in the year 2010 was highly hazardous and was generated by extraordinary activity of monsoon rain, high drainage density, and steep gradient of the watershed. Average gradient of 2.1% developed by the River Panjkora from upper zone to lower zone in the reach of 120Km. Under-cut of the fragile slope caused landslides. Damage estimation cost revealed that in upper zone flood was very severe, in middle zone moderate and again severe in the lower zone. This flood was damaging and destructive in the entire valley and damaged buildings, infrastructure, and agriculture with huge economic losses. Human and animal lives were also severely affected. It was further explored that the flood water undercut the foundation of road and retaining walls in the proximity of river and resulted damage to road and bridges. Infrastructure was badly affected by flood. Kalkot, Barikot and Sharingal were with maximum economic loss, because of their lactation, where
hill torrents drains into the main river from both sides, at this location channel is broad and people had built houses and shops, and were eventually damaged by flood. Still these sites are at high risk in future for such extreme events because gradient is more than 3%, tributary stream from both sides drain into the river and human settlements are in the river proximity. In future, construction and commercial activities should ban in the river proximity to minimize the above mentioned damages. Plantation on the slope is highly recommended to enhance the retention capability of the catchment. Rehabilitation of road and bridges should be based on the highest recorded discharge. Embankments are highly recommended in the lower zone to protect the agriculture land. References 1. Ashfaq M, Ehsanullah, 2011. Floods in Pakistan (Lessons learned and way forward) in agriculture sector- a success story. Pakistan Engineering Congress 32:168-196. 2. Asrar-ul-Haq, Zaidi S M, 2011. Flood 2010: The event, issue and way forward. Pakistan Engineering Congress 32:53-74. 3. Atta-ur-Rahman, Khan A N, 2011. Analysis of 2010-flood causes and associated socioeconomic damages in Hindukush region. Nat Hazards 59:1239-1260. doi: 10.1007/s110690011-9830-8. 4. Atta-ur-Rahman, Khan A N, 2013. Analysis of 2010-flood causes, nature and magnitude in the Khyber Pakhtunkhwa, Pakistan. Nat Hazards 66:887-904. 5. Champion H G, Seth S K, Khattak G M, 1965. Forest Types of Pakistan. Forest Institute Peshawar, Pakistan. 6. Choudhury N Y, Paul A, Paul B K, 2004. Impact of coastal embankment on flash flood in Bangladesh: A case study. Applied Geography 24:241-258. 7. Delrieu G, 2005. The catastrophic flash-flood event of 8–9 September 2002 in the Gard region, France: A first case study for the Cévennes-Vivarais Mediterranean hydro-meteorological Observatory. Journal of Hydrometeorology 6:34–52. 8. Fendler R, 2008. Floods and safety of establishments and installations containing hazardous substances conclusions on a research project of the German Umweltbundesamt. Nat Hazards 46:257–263. 9. Gaurav K, Sindha R, Panda P K, 2011. The Indus flood of 2010 in Pakistan: A perspective analysis using remote sensing data. Nat Hazards 59:1815-1826. doi: 10.1007/s11069-011/95696. 10. Government of KP (GoKP), 2011. Flood Report 2010. Irrigation and Drainage Authority, Peshawar, KP. 11. Government of Pakistan (GoP), 2000. Annual Flood Report, 2000. Ministry of Water and Power, Federal Flood Commission, Government of Pakistan, Islamabad. 12. Government of Pakistan (GoP), 1999. District census report of Dir, 1999. Population Census Organization, Islamabad. 13. Government of Pakistan (GoP), 2014. Population Census of Pakistan, Islamabad. 14. Groisman P Y, Knight R W, Karl T R, Easterling D R, 2004. Hydrological cycle over the contiguous United State: Trends. Journal of Hydrometeorology 5:64-85. 15. Hunter, N M, Horritt M S, Bates P D, 2005. An adaptive time step solution for raster-based storage cell modeling of flood plain inundation. Advance Water Resources 28:975–991.
16. Korutny M L, Kichigina N V, 2010. Geographical analysis of river floods and their causes in southern East Siberia. Hydrological Sciences Journal 51(3): 450-464. 17. Krausmann E, Mushtaq F, 2008. A qualitative Natech damage scale for the impact of floods on selected industrial facilities. Nat Hazards 46:179-197. 18. Lehner B, Alcamo J, Henrichs T, Frank K, 2006. Estimating the impact of global change on flood and drought risks in Europe: A continental, integrated analysis. Climate Change 75:273– 299. 19. Mazzarana B, Hubl J, Fuchs S, 2009. Improving risk assessment by defining consistent and reliable system scenarios. Nat Hazard Earth Sys 9:145-159. 20. Pande K, 2010. Flash flood disasters in Uttarakhand. Disaster Prevention and Management 19(5):565-570. 21. Pakistan Meteorological Department (PMD), 2011. Regional Meteorological Observatory, Peshawar, KPK. 22. Rossa A, Laudanna D G, Borga F, Zanon F, Settin T, Leuenberger D, 2010. Radar-driven highresolution hydro-meteorological forecasts of the 26th September 2007 Venice flash flood. Journal of Hydrology 394:230-244. 23. Ruin I, Creutin J D, Anquetin S, Lutoff C, 2008. Human exposure to flash floods relation between flood parameters and human vulnerability during a storm of September 2002 in Southern France. Journal of Hydrology 361:199-213. 24. Syvitski J P M, Brakenridge G R, 2012. Causation and avoidance of catastrophic flooding along the Indus River, Pakistan. The Geological Society of America 4-10. doi: 10.1130/GSATG165A.1. 25. Tariq M A U R, Giesen V D N, 2012. Floods and flood management in Pakistan. Physics and Chemistry of the Earth 47(48)11-20. doi: 10.1016/j.pce.2011.08.014. 26. Viviroli D, Weingartner R, 2004. The hydrological significance of mountains: From regional to global scale. Hydrology and Earth System Science 8(6):1016-1029. 27. Viviroli D, Weingartner R, Messerli B, 2003. Assessing hydrological significance of the World’s mountains. Mountain Research and Development 1:32-40. 28. Zaidi S S, 2011. Pakistan Flood 2010: An opportunity to build back better. Pakistan Engineering Congress 32:179-196. 29. Zhang, J Q, Zhou C, Xu K, Watanabe M, 2002. Flood disaster monitoring and evaluation in China. Environmental Hazards 4:33-43. 30. Zhou C, Yang, C J, Baoiln L I, Wang S, 2000. Flood monitoring using multi-temporal AVHRR and RADARSAT imagery. Photogrammetric Engineering & Remote Sensing 66(5): 633-638.
PII: DOI: Reference:
S2212-4209(15)30074-1 http://dx.doi.org/10.1016/j.ijdrr.2016.02.009 IJDRR322
To appear in: International Journal of Disaster Risk Reduction Received date: 18 September 2015 Revised date: 26 February 2016 Accepted date: 28 February 2016 Cite this article as: Shakeel Mahmood, Amin ul Haq Khan and Saif Ullah, Assessment of 2010 Flash Flood Causes and Associated Damages in Dir Valley, Khyber Pakhtunkhwa Pakistan, International Journal of Disaster Risk Reduction, http://dx.doi.org/10.1016/j.ijdrr.2016.02.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Assessment of 2010 Flash Flood Causes and Associated Damages in Dir Valley, Khyber Pakhtunkhwa Pakistan
Shakeel Mahmood1, 2, Amin ul Haq Khan3, Saif Ullah4 1
Department of Geography, University of Peshawar, KP, Pakistan Department of Geography, Government College University Lahore, Pakistan Email: [email protected], [email protected] 3 Sustainable Development Study Center, Government College University Lahore, Pakistan 4 Department of Mathematics, Government College University Lahore, Pakistan Email: [email protected] 2
Abstract This paper identifies the underlying causes and damages caused by 2010 flash flood, which is considered one of the worst hydrological disasters in the country (Pakistan). Questionnaire based survey and structured interviews were conducted to investigate physical and economic damages in the selected case study Dir Valley. Global Positioning System (GPS) survey was also conducted to acquire location of damages and Geographical Information System (GIS) is used to visualize land use, land cover and damage data. Buffer analysis is used to delineate the effected strata on both sides of Panjkora River to assess the flood damages. Climate change phenomenon along with peculiar topography of the region are considered as the underlying causes as monsoon winds interacted with the jet stream abnormally causing unprecedented high intensity rainfall in the valley. The steep topography of the area caused rain water to accumulate rapidly in the Panjkora River, overpowering the withholding capacity of the river. The phenomenon resulted in severe flash flood which annihilated whatever came in its way. Upper zone in the target area is badly affected by flood with maximum damaged houses and human causalities followed by lower zone with high number of damaged bridges. This study will bring the attention of disaster management and other relevant authorities to focus on flood risk reduction by enhancing the retention capability of watershed in upstream areas that will reduce the risk in low lying areas. Key Words: Flash Flood; Causes; Damages; Risk Reduction; GPS; GIS. 1 Introduction For quite some time, disasters around the World are on the rise, not so much due to shorter recurrence of natural extreme events, but more so due to spatial concomitant of hazards with rising human vulnerabilities across the globe. Disasters rising trends, especially those of flash floods can be commonly witnessed in most recent disaster related literature. Natural disasters are increasing tremendously over the earth surface disrupting physical and cultural environment (Fendler, 2008). Floods are the serious socio-natural disaster (Zhang et al., 2002; Krausmann and Mushtaq, 2008) affecting social and economic aspects of life (Wang et al., 1995). Changes in meteorological conditions and land-use pattern have increased floods frequency in many parts of the World (Lehner et al., 2006). Flash flood is hydro-meteorological hazard (Pande, 2010) and occurs after five or six hours of heavy rain in any watershed with high level of destructive discharge (Zhou et al., 2000; Korutny and Kichigina, 2006; Ruin et al., 2008). Its occurrence often associated with orographic effect augmenting violent weather conditions and steep gradient promoting rapid collection of stream flow (Rossa et al., 2010; Gaurav, 2011). Pande (2010) has investigated that cloudburst, breaching of landslides dams, and glacial lakes outburst are factors of flash floods genesis. It has potential to transports large mass of
regolith in the dead channels resulting massive erosion. Mazzarana (2009) expressed that there are scientific evidences of extreme precipitation events which entails that extreme flood events might become more frequent in different regions of the World. Hydrological processes in mountains are strongly influenced by temperature, precipitation, vegetation cover, and gradient (Groisman et al., 2004; Tariq and Giesen, 2011). Mountains; the water towers roughly cover 27% of the plant’s earth surface and draining water through the local drainage system (Viviroli et al., 2003). Areas with altitude ranging from 2000m to 4000m are high risk zones because of steep topography, high stream density, and heavy rain (Viviroli and Weingartner, 2004). The potential for flash flood casualties and damages has been increasing because of development in the proximity of rivers and inappropriate land uses (Groisman et al., 2004). The study area; Upper Dir is a mountainous area with high drainage density, steep topography, and sometimes receives heavy rain because of extra-ordinary activity of monsoon or cloudburst. In the year 2003, flash flood was generated by cloudburst, while in 2010 heavy monsoon rain was the main flood generating factor in the entire valley, inflected heavy losses on life, properties of the local population and infrastructure. This paper comprises of five sections. The first section is regarding introduction, second section describes the geographical conditions of the study area, third section elaborates the methodology of the research work, fourth section comprises of results and discussion and last section concludes the study. 2
Geographical Conditions of the Study Area
The study area; Dir valley is situated in the North-West of Pakistan and about 180Km from the capital city Islamabad. It extends from 35º 04' to 35º 46' N-latitude and 71º 32' to 72º 22' E-longitude. It is bounded from the North and North-West by district Chitral, from the East by Swat valley, from the South-West by Afghanistan, and in South it has common outlet of Panjkora River (Figure 1). It is bounded from all sides by mountains except the South-East with total geographical area of 3,699Km2 (Government of Pakistan (GoP), 1999). The total population was 575,000 as enumerated in the 1998 census with an increase of 58.7% since 1981. The average annual growth rate was 2.8% during this period and population density was 156persons/Km2. The estimated population till 2014 is 805, 000 (Government of Pakistan (GoP), 2014). The elevation of study area varies from 5577m in the North to 844m in the South. The length of the valley is 120Km with average gradient of 2.1%. Physically, Hinduraj Mountains, the off-shoots of Hindukush Mountains act as watershed between Dir valley and the surrounding districts. These mountains have snow covered peaks, accumulated snow in the valleys and glaciers in the North-East, where elevation exceeds 5000m. River Panjkora originates from these mountains and its tributaries are Rivers Kohistan, Barawal, Usherai and Dir (Figure 2). Upper zone is narrow with deep gorge and lower zone is broad. Average width of the channel is 0.7Km. River Panjkora is the main tributary of River Swat. Most of the area is covered by temperate forests in higher altitude while low altitude areas are covered by degraded sub-tropical scrub forest (Champion et al., 1965). These forests are economically important in generating revenue, source of timber, and fire wood. Furniture industry of the study area depends on these forests. Agriculture is the main economic activity of the people and terraced farming is common, and cattle, sheep and goat rearing is common as well. Soil and climate are favorable for the cultivation of crops including wheat, maize, rice, onion, potato, tomato, and fruits; apple, apricot, pears, walnuts, and citrus fruits (GoP, 1999).
3
Data Acquisition and Analysis Methods
Primary and secondary data have been collected to achieve the objectives of this research work. Primary data were collected through questionnaire survey, interviews, personal observation and GPS. Prior to field survey, frequent field visits and meeting with local people were arranged to estimate the extent of flood damage and the identification of most affected sites. Detail micro-level investigation and analysis of the flood affected areas were carried out based on initial field visits and eleven sample sites; Sharingal, Barikot, Kalkot, Palam, Tarpatar, Barawal Bandai, Wari, Akhagram, Khal, Turmong and Baroon (Figure 7), were selected randomly from the entire valley along the river. Questionnaire based survey was conducted in all the sample sites using random sampling techniques within 200m buffer zone on both sides of the river. Spatial sampling technique was used to acquire the geo-location of damaged structures; buildings, bridges, road and retaining walls in the entire valley using GPS. Similarly elevation data along the channel was acquired using GPS to calculate gradient (which is the change in elevation over the distance covered along the surface) of the channel using the following formula.
Fig. 2. Gradient of the Channel = Gradient of the Channel
Difference in height between two po int s 100 . Horizontal dis tan ce between two po int s
y 2 y1 rise 100 100 , x2 x1 run
where “rise” is the change in elevation between selected points and “run” is the horizontal distance between them (Graph based on the above formula is given in Figure 5).
Fig. 1. Location of Upper Dir and different landuses with villages
Separate structured questionnaires were designed for households head and concerned government department officials. Questionnaire survey was conducted directly in the sample sites by asking face to face questions from 339 flood affected people in local language “Pashto”. They were mainly asked regarding the causes and damages incurred by flood, nature of damages, and estimated rehabilitation cost. This constituted approximately 65% of the people living in the selected buffer zone. Concerned officials were also interviewed to explore the causes of flood, extent of damages, and estimated rehabilitation cost in “million Pakistan Rupees (PKR)”. Damage estimation cost of private property including houses, buildings, land loss, crops and livestock collected in field survey according to community perception. The average value for rehabilitation cost was calculated using computer based statistical program MS Office 2007. The estimated rehabilitation cost of damaged water supply schemes collected from Public Health Department, District Upper Dir, Khyber Pukhtoon Khawa (KPK), and damaged bridges, retaining walls and roads estimated rehabilitation cost is collected from Construction and Works Department, District Upper Dir, KPK. Secondary data collected from different concerned government departments. Rainfall and temperature data were collected from Pakistan Meteorology Department, Regional Office Peshawar. Population statistics were collected from Population Census Organization, Islamabad. Topographic map of the study area was collected from Survey of Pakistan (SoP), Islamabad to prepare land use and land cover maps using ArcGIS 9.3. Digital Elevation Model (DEM) with 30m resolution was downloaded from online geo-database of United State Geological Survey (USGS). Hydrologic analysis performed on DEM using ArcHydro to analyze hydro-geometric network of the study area. Similarly, Buffer analysis was applied to highlight the structures and settlements in 200m buffer zone. Finally, results were presented in the form of maps with 360 dots per inch (dpi) and graphs. 4 Results and Discussions From the obtained data and results, assessments of causes and damages of flash flood 2010 have been made. 4.1 Causes Assessment Pakistan is prone to floods because of its hydro-meteorological conditions (Atta-ur-Rahman and Khan, 2010), and every year different nature of floods have been occurring including monsoon floods, flash floods, urban floods, and coastal floods (Zaidi, 2011). However, excessive rainfall has been the main cause of many flood events (Hunter et al., 2005; Delrieu et al., 2005). In the year 2010, disastrous flood generated by the extra-ordinary activity of monsoon rainfall, melting of snow and glaciers in Pakistan (Atta-ur-Rahman and Khan, 2011; Tariq and Giesen, 2011; Atta-urRahman and Khan, 2013). Meteorological setup of Pakistan was unique in the last week of July, 2010 in the context of spatial coverage, intensity and duration of clouds compared to the past meteorological events, because of the interaction of monsoon winds and westerlies over the North-West of Pakistan. This caused unprecedented rain in the entire country, particularly in the North-western and western parts of Pakistan. This rainfall became the major factor of high surface run-off while steep topography accelerated the accumulation of rain water in the channel and produced flash flood in Rivers Panjkora
Fig. 3. Upper Dir Drainage Pattern and Digital Elevation Model developed from 30m SRTM data.
Fig. 4. Dir Valley, Mean Monthly Rainfall of Past 40 Years & Mean Monthly Rainfall of 2010. Source: PMD 2011 and Swat. This generated historic flood peak in River Kabul inundated most of the Peshawar Plain. Similarly, this highest peak became the flood generating factor in Indus River. Normally, duration of previous flood peaks ranged from 12 to 24 hours, while the duration of this flood was 110 hours (Asrar-ul-Haq and Zaidi, 2011; Syvitski and Brakenridge, 2012). Dir Valley is located in the NorthWestern Mountains, where the mentioned unprecedented rain generated destructive flash flood. This flood brought huge damage to social, economic and physical setup of the study area. The target area has long winter and short summer seasons. December to February, are the coldest months with mean maximum and minimum temperature in January is 11ºC and -2ºC, respectively. While, July and August are the hottest months with mean maximum and minimum temperature of 32º C and 16.3º C, respectively. In summer, the source of rainfall is local relief and monsoon winds, while Western Depressions have been the source of rainfall in winter. There is variation in annual rainfall since 1967.
Upper Zone
Middle Zone Lower Zone
Fig. 5. Panjkora River Gradient based on elevation data acquired through GPS. Source: Field Survey, 2011
Fig. 6. Community Perceptions Regarding Flood Causes Source: Field Survey 2011
Lowest rainfall is recorded in 1971 and 2001, while highest in 1979, 1986, and then in 2010. Analysis revealed that in the last four decades there is positive shift in rainfall particularly in July. In the year 2010 increase in mean monthly rainfall has taken place from May to September. The normal for the months of May, June, July and August is 89mm, 56mm, 156 and 149mm while the rainfall in the year 2010 for the same months is 110mm, 76mm, 393mm and 285mm respectively (Figure 4). In July, 2010 rainfall was maximum because of the heavy downpour from 27th to 30th July with intensity of 98mm/day, while on 29th July its intensity was 6.5mm/hour (Pakistan Meteorological Department (PMD), 2011). This high intensity rain became the main flash flood generating factor in the entire valley which was never ever experienced by this valley in the past record. July and August are the hottest months with maximum temperature higher than 32ºC resulting melting of snow and glaciers in the upper zone. This melt water is the main sources of discharge in summer season. The high intensity rain has accelerated melting. Thus the melt water has contributed in the genesis of high discharge which was beyond the capacity of the channel and generated massive flash flood with disastrous effect on human life, property and infrastructure. Topography of the target area is mountainous. River valley is narrow and steep in upper zone, while slope is decreasing towards the lower zone and width is increases. Gradient of channel varies from 3% in upper zone to 2.5% in middle zone and 2.1% in lower zone. The average gradient of the entire channel is 0.8% which has intensified the flash flood (Figure 5). There is a system of interconnected sub-watersheds and these characters of the study area have caused rapid collection of rain and melt water in the River Panjkora and resulted in disastrous flash flood. The factor of deforestation is also very important, because forest cleared areas accelerated the surface run-off and sheet erosion which also intensified the potential for flood. People are cutting trees as fuel sources because there is no other source of fuel. Thus heavy rainfall and steep topography have resulted flash flood which was further intensified by melt water and deforestation. In the target area, people were asked regarding flood generating factors because local people have indigenous knowledge. In Palam and Tarpatar 100% respondent view point was, “heavy monsoon
rainfall” while in rest of the sample sites more than 60 percent respondents view point was “heavy monsoon rainfall, and melting of snow and glacier” resulted this disastrous flood (Figure 6). According to elders of the community and certain officials; the principal cause of flood was rain-water. They also said that forest cover has been decreasing which has increased surface run-off and sheet erosion. It was observed in the target area that local people has people has cleared forest for the purpose of fuel and construction of houses. Most of the areas located on slopes are exposed to the action of snowfall and rainfall. In the winter season frost action weathered the exposed rocks and increased the fragility of the slopes. During monsoon seasons high surface runoff has resulted massive erosion and rapid collection of rain water from barren surfaces in torrents and streams which has generated high flood in the River Panjkora River. 4.2
Damages Assessment
Analysis revealed that flash flood in the year 2010 seriously affected human and animal lives, and undercut foundations of buildings, retaining walls and bridges, eroded topsoil, and changed river morphology. Within eleven selected sample sites, damages were assessed through extensive field surveys, observations and interviews. Historically, population clusters and human activities are concentrated in the proximity of river. Mostly people are living in mud-brick and mud-stone houses in the proximity of river which are less resistant to flood. Flood damaged large number of houses in the upper zone; Kalkot, Barikot, Sharingal, Palam, and Tarpatar where stream density is high and gradient is more than 3%. In all sample sites completely and partially damaged houses were 422 and 45 with estimated economic loss of 295.4 and 18 million PKR, respectively. The number of damaged houses is 93% of the total in the selected buffer zone. Kalkot was severely affected village in terms of damaged houses, because of abrupt change in gradient and hill torrents draining into the main channel from both sides. Here, cluster of houses were built in river proximity, 156 houses were completely damaged by flood with estimated economic loss of 109.2 million PKR. Damaged shops and restaurants were located in lower zone. The total number of damaged shops was 40 with estimated economic loss of 40 million PKR. Akhagram is the most affected site in terms of damaged shops. At this location shops were built to provide daily use goods where the channel is narrowest in the entire reach accompanied by interlocking spurs. This has generated high pressure resisted by spurs; the projecting part of the mountains into the river. This fluvial landform forms curves in the channel with outer and inner bank. The outer bank of the curve is backed by spur projecting into the channel that has diverted flow to the inner bank where shops were located. This surge shifted the flow with high hydraulic pressure towards inner bank, busting the concrete buildings and demolished everything there. Damaged buildings are located over the inner bank because outer bank has steep slope with no construction. Two bridges were also completely damaged by this flow at the same location. Distribution of damaged shops is concentrated in lower zone, because these structures were built in river proximity. Three restaurants were completely damaged with estimated loss of 27 million PKR. In the upper zone damaged buildings are located on both sides of the channel. In lower zone human settlements are dispersed and commercial buildings have been constructed on the left bank of river along Dir-Chitral Road. People have been living in the river proximity on both sides with main population clusters including the selected sample sites. Panjkora drainage system comprises of interconnected sub-watershed with perennial streams. In the upper zone, sub-watershed has maximum gradient and high drainage density,
while in lower zone the drainage density is low. High drainage density has made hindrance in the movement of people from one village to other situated across the river as well as along the river. Bridges and culverts overcome the problem accessibility. Flood damaged most of these structures. Partially and completely damaged bridges were 36 with estimated cost of 125.6 million PKR. Analysis revealed that highest recorded discharge, low height of bridges, high flow energy at bridging points, tree logs and timber have damaged these structures (Figure 7). Road and retaining walls were damaged by both the main river and tributary streams. In lower zone damages to road were high because of nearness to river level and maximum hydraulic pressure exerted by highest discharge. In upper zone, two factors are responsible for damages to road and retaining wall. First is the narrowness of channel and second is interlocking spurs. Maximum energy in the narrow parts of channel resulted high level of damages to the road and other structures. Similarly, the spurs have shifted the flow to the inner bank of the channel that undercut base of the retaining wall and road. In upper zone, this character of the valley has also resulted landslide that further intensified the flood. Total estimated economic loss resulted by damaged roads and retaining wall was 410 million and 180 million PKR, respectively. People are earning their livelihood mostly from agricultural activities practiced as terraced farming. Fields in the proximity of river were damaged by flash flood. In the upper zone, agriculture land in the proximity of river was eroded by flood while, in the lower zone land and crops were damaged by the temporary deposition of sand and pebbles. The layer of sand and pebbles over the agriculture land is ranging from 0.8m to 2m. The average width of the flood plain in lower zone is 1.5km on right side and 0.5km is on left side of the river. Over the right bank the alluvium was 5m deep with undulating surface towards river. River has shifted the course towards the agriculture land on right side by eroding 250m strip of 5m thick agriculture land. In Barikot 2.3ha, Kalkot 1.9ha, and Palam1.2ha of agriculture land has damaged by the flood causing economic loss of 4.6, 3.8 and 2.4 million PKR, respectively. In rest of the areas, damaged agriculture land was less than 1ha. Total economic loss to agriculture land and crops was estimated as 19.54 million PKR. In sample sites of Kalkot and Barikot, 161 livestock were lost with estimated cost of 2.468 million PKR. In Barikot 60 livestock including cow, sheep, and goat were taken by flood with estimated cost of one million PKR followed by Kalkot with estimated loss of 0.78 million PKR (Figure 8). Although the study area receives rainfall throughout the year but it is not sufficient for agricultural activities. River Panjkora is an important source of irrigation. Water is diverted in narrow channels to irrigate fields in river proximity. These channels are lined and unlined with 1m width and 0.5m depth. Flood has damaged the heads of these channels in all sites with estimated economic loss of 180 million PKR. Water supply schemes were also severely affected by flood. Most of the damages were concentrated in the upper zone with total estimated cost of 57.5 million PKR. The heads of water supply schemes are in the upstream of hill torrents. Destructive flow has completely washed away the heads of these water supply schemes
Fig. 7. Distribution of damaged houses, shops, restaurants, road and bridges developed in ArcGIS from GPS data.
Flood has been very lethal for human and animal lives, because it causes human injuries and fatalities. Most of the human causalities have taken place in the upper zone where gradient is more than 3%. In Palam and Tarpatar, impact of flood on human life was highest with total death cases of 25 and 20, respectively. Most of the victims were working in their fields in the proximity of river. They were not aware because of the lack of any warning or emergency system. In the study area, total fatalities and injuries were 66 and 69, respectively (Figure 9). Analysis revealed that 2010 flash flood adversely affected the study area and resulted severe damages to infrastructure, buildings, agriculture and human life. In all sample sites, completely and partially damaged houses were 467 with estimated cost of 313.4 million PKR, completely and partially damaged bridges were 30 and 6, respectively with estimated rehabilitation cost of 125.6 million PKR; damaged roads and retaining walls were 4.1Km and 6Km with estimated rehabilitation cost of 410 and 180 million PKR, respectively; total length of damaged irrigation channels was 36Km with estimated rehabilitation cost of 180 million PKR. Similarly, 25 water supply schemes were completely damaged with estimated rehabilitation cost of 57.5 million PKR. The number of damaged shops and restaurants were 40 and 3 with estimated economic loss of 40 and 27 million PKR, respectively. Agriculture was also badly affected by flood with total estimated economic loss of 23 million PKR which was simply unaffordable for subsistence agriculture. The most affected sector was infrastructure with damaged roads, retaining walls, water supply schemes, and bridges. In the study area most affected sites are Kalkot, Sharingal and Barikot with total estimated economic damages in selected sectors were 244.35, 233.43, and 178.8 million PKR, respectively—located in the upper zone. In Akhagram the total estimated cost of damages was 171 million PKR with highest economic loss in lower zone while, Wari is the least affected site because human activities and building are away from the river. The estimated economic loss is 39.2 million PKR. Similarly in Khal, Baroon and Tormang losses are less because people are living away from the river with few buildings in river proximity. Total estimated economic loss caused by the 2010 flood in selected sectors of the study area is more than 1350 million PKR, which is a significant loss to the very weak provincial and national financial economy. Geographically the terrain of target area is mountainous with narrow and steep sub-valleys. These valleys are more in the upper zone because of the high density of streams as compare to lower zone. Fast flowing streams and torrents drain into the main channel from both sides. This character of the upper zone has produced disastrous high flow in the channel. Local people have built their house in torrents, stream and main river proximity to make water accessible because they have never experience such high discharge in past like in 2010. In lower zone people have built restaurants and departmental stores with foundation in or near channel. This level of exposure has made the human life, property, and infrastructure more vulnerable to flash flood and resulted huge physical and economic damages in the study area.
Fig. 8. Site-wise economic loss and gradient of the Panjkora channel. Source: Field Survey, 2011
Fig. 9. Human Fatalities and Injuries Source: Field Survey, 2011
5
Conclusions
This study concludes that flood in the year 2010 was highly hazardous and was generated by extraordinary activity of monsoon rain, high drainage density, and steep gradient of the watershed. Average gradient of 2.1% developed by the River Panjkora from upper zone to lower zone in the reach of 120Km. Under-cut of the fragile slope caused landslides. Damage estimation cost revealed that in upper zone flood was very severe, in middle zone moderate and again severe in the lower zone. This flood was damaging and destructive in the entire valley and damaged buildings, infrastructure, and agriculture with huge economic losses. Human and animal lives were also severely affected. It was further explored that the flood water undercut the foundation of road and retaining walls in the proximity of river and resulted damage to road and bridges. Infrastructure was badly affected by flood. Kalkot, Barikot and Sharingal were with maximum economic loss, because of their lactation, where
hill torrents drains into the main river from both sides, at this location channel is broad and people had built houses and shops, and were eventually damaged by flood. Still these sites are at high risk in future for such extreme events because gradient is more than 3%, tributary stream from both sides drain into the river and human settlements are in the river proximity. In future, construction and commercial activities should ban in the river proximity to minimize the above mentioned damages. Plantation on the slope is highly recommended to enhance the retention capability of the catchment. Rehabilitation of road and bridges should be based on the highest recorded discharge. Embankments are highly recommended in the lower zone to protect the agriculture land. References 1. Ashfaq M, Ehsanullah, 2011. Floods in Pakistan (Lessons learned and way forward) in agriculture sector- a success story. Pakistan Engineering Congress 32:168-196. 2. Asrar-ul-Haq, Zaidi S M, 2011. Flood 2010: The event, issue and way forward. Pakistan Engineering Congress 32:53-74. 3. Atta-ur-Rahman, Khan A N, 2011. Analysis of 2010-flood causes and associated socioeconomic damages in Hindukush region. Nat Hazards 59:1239-1260. doi: 10.1007/s110690011-9830-8. 4. Atta-ur-Rahman, Khan A N, 2013. Analysis of 2010-flood causes, nature and magnitude in the Khyber Pakhtunkhwa, Pakistan. Nat Hazards 66:887-904. 5. Champion H G, Seth S K, Khattak G M, 1965. Forest Types of Pakistan. Forest Institute Peshawar, Pakistan. 6. Choudhury N Y, Paul A, Paul B K, 2004. Impact of coastal embankment on flash flood in Bangladesh: A case study. Applied Geography 24:241-258. 7. Delrieu G, 2005. The catastrophic flash-flood event of 8–9 September 2002 in the Gard region, France: A first case study for the Cévennes-Vivarais Mediterranean hydro-meteorological Observatory. Journal of Hydrometeorology 6:34–52. 8. Fendler R, 2008. Floods and safety of establishments and installations containing hazardous substances conclusions on a research project of the German Umweltbundesamt. Nat Hazards 46:257–263. 9. Gaurav K, Sindha R, Panda P K, 2011. The Indus flood of 2010 in Pakistan: A perspective analysis using remote sensing data. Nat Hazards 59:1815-1826. doi: 10.1007/s11069-011/95696. 10. Government of KP (GoKP), 2011. Flood Report 2010. Irrigation and Drainage Authority, Peshawar, KP. 11. Government of Pakistan (GoP), 2000. Annual Flood Report, 2000. Ministry of Water and Power, Federal Flood Commission, Government of Pakistan, Islamabad. 12. Government of Pakistan (GoP), 1999. District census report of Dir, 1999. Population Census Organization, Islamabad. 13. Government of Pakistan (GoP), 2014. Population Census of Pakistan, Islamabad. 14. Groisman P Y, Knight R W, Karl T R, Easterling D R, 2004. Hydrological cycle over the contiguous United State: Trends. Journal of Hydrometeorology 5:64-85. 15. Hunter, N M, Horritt M S, Bates P D, 2005. An adaptive time step solution for raster-based storage cell modeling of flood plain inundation. Advance Water Resources 28:975–991.
16. Korutny M L, Kichigina N V, 2010. Geographical analysis of river floods and their causes in southern East Siberia. Hydrological Sciences Journal 51(3): 450-464. 17. Krausmann E, Mushtaq F, 2008. A qualitative Natech damage scale for the impact of floods on selected industrial facilities. Nat Hazards 46:179-197. 18. Lehner B, Alcamo J, Henrichs T, Frank K, 2006. Estimating the impact of global change on flood and drought risks in Europe: A continental, integrated analysis. Climate Change 75:273– 299. 19. Mazzarana B, Hubl J, Fuchs S, 2009. Improving risk assessment by defining consistent and reliable system scenarios. Nat Hazard Earth Sys 9:145-159. 20. Pande K, 2010. Flash flood disasters in Uttarakhand. Disaster Prevention and Management 19(5):565-570. 21. Pakistan Meteorological Department (PMD), 2011. Regional Meteorological Observatory, Peshawar, KPK. 22. Rossa A, Laudanna D G, Borga F, Zanon F, Settin T, Leuenberger D, 2010. Radar-driven highresolution hydro-meteorological forecasts of the 26th September 2007 Venice flash flood. Journal of Hydrology 394:230-244. 23. Ruin I, Creutin J D, Anquetin S, Lutoff C, 2008. Human exposure to flash floods relation between flood parameters and human vulnerability during a storm of September 2002 in Southern France. Journal of Hydrology 361:199-213. 24. Syvitski J P M, Brakenridge G R, 2012. Causation and avoidance of catastrophic flooding along the Indus River, Pakistan. The Geological Society of America 4-10. doi: 10.1130/GSATG165A.1. 25. Tariq M A U R, Giesen V D N, 2012. Floods and flood management in Pakistan. Physics and Chemistry of the Earth 47(48)11-20. doi: 10.1016/j.pce.2011.08.014. 26. Viviroli D, Weingartner R, 2004. The hydrological significance of mountains: From regional to global scale. Hydrology and Earth System Science 8(6):1016-1029. 27. Viviroli D, Weingartner R, Messerli B, 2003. Assessing hydrological significance of the World’s mountains. Mountain Research and Development 1:32-40. 28. Zaidi S S, 2011. Pakistan Flood 2010: An opportunity to build back better. Pakistan Engineering Congress 32:179-196. 29. Zhang, J Q, Zhou C, Xu K, Watanabe M, 2002. Flood disaster monitoring and evaluation in China. Environmental Hazards 4:33-43. 30. Zhou C, Yang, C J, Baoiln L I, Wang S, 2000. Flood monitoring using multi-temporal AVHRR and RADARSAT imagery. Photogrammetric Engineering & Remote Sensing 66(5): 633-638.