Biological Conservation 192 (2015) 30–41

Contents lists available at ScienceDirect

Biological Conservation

journal homepage: www.elsevier.com/locate/bioc

Review of status, threats, and conservation management options for the endangered Indus River blind dolphin

Gill T. Braulik a,b,⁎,UzmaNoureenb, Masood Arshad b, Randall R. Reeves c a Wildlife Conservation Society, Tanzania Program, Zanzibar, Tanzania b WWF-Pakistan, Ferozepur Road, Lahore, Pakistan c Okapi Wildlife Associates, Hudson, Quebec, Canada article info abstract

Article history: The Indus River dolphin (Platanista gangetica minor) is a freshwater cetacean that occurs only in the Indus River Received 18 August 2015 system in Pakistan and India. This review provides a comprehensive summary of issues surrounding Indus Accepted 6 September 2015 dolphin conservation, and includes a description of their distribution, the threats they face and a discussion of Available online xxxx conservation and research activities, options and priorities. Information was gathered from published current and historical literature, newspaper articles, and unpublished reports. Prior to construction of the Indus Irrigation Keywords: system dolphins occurred in 3500 km of the Indus River system. Their range has now declined by 80% due to Freshwater dolphins Endangered species fragmentation of river habitat by barrages and large-scale diversion of water for irrigated agriculture. The sub- Dams species was estimated to number approximately 1450 individuals in 2011, and occurs in 6 subpopulations, Aquatic biodiversity three of which are probably too small to persist. There is an urgent need to evaluate whether dolphins move Habitat fragmentation through barrages, as this has the potential to extirpate upstream subpopulations. Pre-requisites for dolphin translocations to combat this problem are outlined. Mortality in fishing gear and high levels of chemical pollution in water courses are both increasing threats exacerbated by depleted flows. Research and conservation priorities include maintenance of river flows, mortality monitoring, canal rescues and community-based conservation. Conservation and research action is urgently needed to prevent the Indus dolphin from succumbing to the same fate as the baiji. © 2015 Elsevier B.V. All rights reserved.

Contents

1. Introduction...... 32 2. Materialsandmethods...... 32 3. TheIndusRiver...... 32 4. Statusofdolphinsandtheirhabitat...... 32 4.1. HistoricalinformationonIndusRiverdolphins...... 32 4.1.1. Historicaldistributionandabundance...... 32 4.1.2. DevelopmentoftheIndusbasinirrigationsystem...... 33 4.2. CurrentstatusofIndusRiverdolphins...... 33 4.2.1. Presentdistributionandabundance...... 33 4.2.2. TrendsinabundancebetweenGudduandSukkur...... 34 4.3. ShortsummaryofIndusdolphinecologyandbiology...... 35 5. Threats...... 35 5.1. Directcausesofmortality...... 35 5.1.1. Hunting...... 35 5.1.2. Fisheriesinteractions...... 35 5.1.3. Canalentrapment...... 35 5.2. Indirectcausesofmortality...... 36 5.2.1. Waterpollution...... 36 5.2.2. Freshwatersupplies...... 36

⁎ Corresponding author. E-mail addresses: [email protected] (G.T. Braulik), [email protected] (U. Noureen), [email protected] (M. Arshad), [email protected] (R.R. Reeves).

http://dx.doi.org/10.1016/j.biocon.2015.09.008 0006-3207/© 2015 Elsevier B.V. All rights reserved. G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 31

5.3. Climatechange...... 37 5.3.1. Damsandbarrages...... 37 5.4. Populationdynamicsanddownstreammigratoryattrition...... 37 6. Conservationoptionsandresearchpriorities...... 38 6.1. Mortalitymonitoring...... 38 6.2. Protectedareas(PAs)...... 38 6.3. Downstreammigration...... 39 6.4. Translocation...... 39 6.5. Rivermanagement...... 40 7. Conclusions...... 40 Acknowledgements...... 40 References...... 40

Fig. 1. The Indus River system, and the locations of irrigation barrages and dams. 32 G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41

1. Introduction peak flows of 20–30,000 m3/s occur between June and August when the river is fed by Himalayan melt-water and monsoon run-off, while The Indus River dolphin (Platanista gangetica minor) is an endan- flows as low as 300 m3/s occur in the dry season between December gered, obligate freshwater cetacean that occurs only in the Indus River and April. Human habitation is sparse but increases with proximity to system in Pakistan and India (Fig. 1). The subspecies is red-listed as En- the delta. The only large towns along the Indus are Dera Ismail Khan, dangered due to an 80% reduction in distributional range and a highly Sukkur and Hyderabad. The river is little used for commercial traffic, fragmented population (Braulik et al., 2014c). Indus dolphins, also probably because passage is repeatedly blocked by barrages, and the called ‘blind dolphins’ because of their reduced eyes and poor vision, few vessels present are oar-powered or motorised ferries and small are mammals of high conservation priority because of their evolution- fishing boats. ary distinctiveness and threatened status (Isaac et al., 2007). Although At present, the Indus plains are composed of desert, semi-desert, blind dolphins are potentially charismatic top predators, that could scrub and irrigated agricultural lands. However, several centuries ago function as ‘flagships’ for aquatic conservation, little is known about the native vegetation and fauna of the area was primarily forest and their basic biology. The factors that influence their survival and have grassland inhabited by numerous large mammals including the tiger driven their decline are not well understood, and the best strategy for (Panthera tigris), leopard (Panthera pardus), Asiatic cheetah (Acinonyx conserving them remains unclear. The Indus dolphin and the closely re- jubatus venaticus) and Indian rhino (Rhinoceros unicornis). All but the lated Ganges River dolphin (P. g. gangetica), also red-listed as Endan- leopard are now locally extinct. Freshwater mega-fauna in the Indus gered, are assigned to a monotypic family, the Platanistidae. This is River system previously included mugger crocodiles (Crocodylus one of the more ancient cetacean families that diverged approximately palustris) which were hunted extensively and are now found in only a 29 Million Years (MY) ago, 22 MY before modern marine dolphins arose few isolated areas of Sindh (Ahmad, 1999). The harmless, fish-eating (Xiong et al., 2009). Recent genetic studies showed that Indus and Gan- crocodilian, the gharial (Gavialis gangeticus), was once widespread but ges dolphins diverged from each other approximately 0.5 MY ago and, if is now extinct in Pakistan (Ahmad, 1999). Two species of otter, the they are shown to have morphological differences, the two subspecies smooth otter (Lutra perspicillata) and the Eurasian otter (Lutra lutra), may be recognised as separate species in the future (Braulik et al., were once common, but these animals were decimated by hunting for 2014b). their pelts and now persist in only a very few locations (WWF- Several reviews of Indus dolphin conservation status were published Pakistan unpublished). Eight species of turtles inhabit the Indus River in the 1980s and 1990s (Reeves, 1998; Reeves and Brownell, 1989; system, including four soft-shelled species that can reach more than Reeves and Chaudhry, 1998). There has been a substantial increase in 1 m in length and four smaller hard-shelled species. Freshwater turtles knowledge since then, and the purpose of this paper is to provide an used to be abundant, but a new illegal trade in soft-shelled turtle parts update and comprehensive summary of what is known about Indus dol- for use in Chinese traditional medicine has resulted in massive turtle phins today, including descriptions of their status and the major threats kills and greatly reduced wild turtle numbers (Pakistan Wetlands they face and a discussion of conservation and research activities and Programme/WWF-Pakistan, 2008). A commercially important fishery options. for the migratory shad (Hilsa ilisha) existed in the Indus River prior to construction of the barrages that blocked their migration. The fish 2. Materials and methods used to enter the Indus River in great numbers each year in the middle of January, ascended the river to spawn during June, July and August, Over a period of several decades we have compiled published and returned to the sea in November (Islam and Talbot, 1968). Before current and historical scientific papers from international and South construction of Sukkur barrage in 1932, Hilsa would migrate all the Asian journals, unpublished government and NGO reports, and way to present-day Taunsa barrage. The Kotri and Sukkur barrages Pakistani university theses, as well as newspaper and magazine articles contain fish ladders but these were inappropriately designed for use in English, French, Urdu or Sindhi that relate to Indus dolphins. All avail- by Hilsa. The fishery has totally collapsed resulting in the loss of able documents were summarised in preparation for this review. The around 9000 jobs and an important source of protein for local people review is organised as follows: 1) background information on the eco- (Moazzam, 1999). The Indus dolphin is one of the last aquatic mega- logical and social setting, 2) Indus dolphin historical and current status faunal species remaining in the Indus River system. and trends, 3) summary of threats to the dolphins including those caus- ing direct and indirect mortality, and 4) details of conservation options 4. Status of dolphins and their habitat and research priorities. 4.1. Historical information on Indus River dolphins 3. The Indus River 4.1.1. Historical distribution and abundance The Indus River originates in Tibet, flows through northwest India A detailed map of Indus dolphin distribution was produced by John and enters Pakistan in the north flowing for the entire length of the Anderson in 1879. Anderson sent letters to government offices across country to the Arabian Sea (Fig. 1). It has five main tributaries: the Jhe- British India asking for detailed information on river dolphins. He said lum, Sutlej, Chenab, Ravi and Beas. These rivers merge to form the Panj- “Notwithstanding that the inquiry was of a novel and rather unusual char- nad River, which has flow almost equivalent to the Indus mainstem. The acter, the replies were most complete and full of interest, and, more-over, Indus leaves the Himalayan foothills and enters the plains at Kalabagh examples of the dolphin were sent to me from the Indus, Ganges and Brah- just upstream of Jinnah Barrage, and then flows at a gentle gradient maputra ”. In the mid-1870s, Indus dolphins apparently never entered (average 13 cm/km), south-southwest, for 1600 km to the sea. the ocean, but were found throughout the year in the Indus, Jhelum, The river runs through semi-desert and irrigated agricultural land. It Ravi, Chenab and Sutlej Rivers from the Himalayan foothills to the estu- is broad, shallow and braided and naturally highly turbid. It is sand- ary, a linear range of around 3500 km. They were reported as constantly bedded and is constantly eroding its bed and banks, so consequently present at Kalabagh and in April as far upstream as Attock (Fig. 1). The there is very little rooted vegetation either submerged or on the reports all confirmed that dolphins occurred furthest upstream during banks. The configuration of channels, islands and sand bars is constantly the flood season and that their distribution contracted when river changing, and major changes occur during the annual flood. Air temper- flow was low. atures rise to 50 °C from May to September, and drop to near freezing It is difficult, almost 150 years later, to verify the information collat- between December and February; river water temperatures show sim- ed by Anderson but all other published reports from that time agree ilar but less extreme fluctuations. River discharge is highly seasonal: with it, and in general, it appears to be reliable. The only exception is G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 33 in Nepal which was not under British Administration, and where the mean fragment size have declined steadily as the Indus Basin rivers upper distribution of the Ganges River dolphin was later found to be have become progressively more subdivided (Fig. 2). 100 km further upstream than shown on Anderson's map (Kasuya and The partition of British India in 1947 saw creation of a new interna- Haque, 1972). Dolphins were reported by Anderson to extend their dis- tional border that bisected the Indus River system. The Indus Water tribution into the foothills of the mountains in the Indus and Jhelum Riv- Treaty was agreed in 1960 and the flows of the Indus, Jhelum and Che- ers, in the Beas and Sutlej they reached only to the base of the foothills, nab, amounting to 75% of the total, were allocated to Pakistan, and the and in the Ravi and Chenab their distributional limit was further down- water in the Ravi, Beas and Sutlej Rivers was allocated to India. This stream on the plains, apparently delimited by the Grand Trunk Road, the has had two significant consequences for Indus dolphins: 1) India has major transport route at the time. These small differences in the up- the rights to the Ravi and Sutlej therefore all the water in these rivers stream extent of distribution may be due to the seasonal range fluctua- is utilised within India, and they are now usually dry by the time they tions being reported differently in different rivers, or to the fact that enter Pakistan, and 2) most of Pakistan's water resources are in the differing habitat in each river resulted in different upstream distribu- west of the country but the greatest human population and the major tional limits. irrigated agricultural areas are in the east. This situation was addressed A regular steam boat service was maintained on the Indus only in the by the construction of massive link canals to transfer water from the early to mid-1800s (MacLagan, 1885), consequently, there are few ac- western rivers to those in the east so that agricultural lands south of counts of travel on the river that can be examined for dolphin sightings the Ravi and Sutlej could continue to be irrigated (Fig. 1). Prior to con- to verify Anderson's map. Alexander Burnes led the first British expedi- struction of the link canals, some flow remained in each river for its en- tion on the Indus travelling from the delta to Lahore. He reported dol- tire length so that land adjacent to the furthest downstream barrage phins in the Indus from the delta to Sukkur and also sighted several at could be irrigated. Opening of the link canals allowed for the complete the confluence of the Ravi and Chenab in July 1835 (Burnes, 1835). A diversion of a river's flow at upstream barrages as the river could be few years later, dolphins were reported to be present south of Thatta replenished downstream by a link canal. Since the 1970s, for several (Burnes, 1842)andtobe“very numerous” between Thatta and Sukkur months of the year the Ravi and Sutlej have been almost completely (Hall, 1848). In the 1860s, dolphins were noted to ascend the Punjab dry and no water is released through Khanki, Qadirabad, Trimmu and rivers (Adams, 1867), and a specimen collected from the Sutlej present- Panjnad barrages on the Chenab River, Balloki and Sidhnai on the Ravi ed to the Indian museum in the 1800s confirms their presence in that and Suleimanki and Islam on the Sutlej (Federal Flood Commission, river at that time (Anderson, 1879). At the far upstream end of their 2010)(Fig. 1). Water diversion has been steadily increasing and the cul- range there is a report from the 1840s that ‘before its junction with tivable area expanding as new canals are built, existing canals are ex- the Sutlej, the Beas is frequented with porpoises’ (Anon., 1846). This is tended, their capacity is increased and barrages are refurbished. the same area where dolphins were recently re-discovered in India Meanwhile, river discharge has been declining steadily (IUCN, 2011). (Behera et al., 2008). These few records are all in agreement with the distribution described by Anderson. 4.2. Current status of Indus River dolphins In 1874 it was reported that Indus dolphins were most abundant in the middle or lower third of their range (Jerdon, 1874), which is consis- 4.2.1. Present distribution and abundance tent with the present relatively high density area on the Indus in north- The historical range of Indus dolphins has been fragmented by bar- ern Sindh. Prior to large-scale water diversion, the Indus River had rages into 17 river sections. Dolphin sighting and interview surveys approximately four times the annual discharge of the Jhelum or Chenab, showed that dolphins have been extirpated from ten river sections, per- six times that of the Sutlej and thirteen times that of the Ravi (IUCN, sist in six sections and are of unknown status in the section of the Sutlej 2011). If discharge alone can be considered a crude indicator of dolphin River on the India-Pakistan border (Braulik et al., 2014a). Indus dolphins abundance, the Punjab tributaries may have supported lower dolphin now occur in five subpopulations on the Indus mainstem, bounded by densities and smaller numbers than the Indus historically, and the Jinnah, Chashma, Taunsa, Panjnad, Guddu, Sukkur and Kotri Barrages Jhelum and Chenab may have had greater dolphin abundance than (Fig. 1). A sixth subpopulation occurs in the Beas River in India the smaller Ravi, Sutlej and Beas. In 1901 it was reported that Ganges (Behera et al., 2008). River dolphins have been extirpated from the and Indus dolphins were not numerous and had once been far more Indus mainstem upstream of Jinnah Barrage and downstream of Kotri widespread, indicating that more than 100 years ago they were already barrage and from the five Indus tributaries in Pakistan (Braulik et al., perceived to be in decline (Blanford, 1901). 2014a). The linear extent of occurrence is now approximately 1000 km (Braulik, 2006) and approximately 99% of the dolphin popula- 4.1.2. Development of the Indus basin irrigation system tion occurs in only 690 km of river, which corresponds to an almost 80% The Indus plains are semi-arid, and the vast majority of the rain falls reduction in effective linear range since the 1870s (Reeves et al., 1991). during the summer with the result that agriculture is reliant on the abil- Three comprehensive surveys have been conducted to estimate ity to divert water from the rivers. Since the 1880s (just after Anderson abundance of Indus dolphins throughout their current range. These produced his map), 19 irrigation barrages have been constructed on the were conducted at five-year intervals, in 2001, 2006 and 2011 lower Indus within, or at the limits of, the former range of the dolphins (Table 1). All three surveys consisted of direct counts by three observers (Fig. 1). The Indus basin irrigation system is now claimed to be the larg- from a viewing platform on an oar-powered wooden vessel travelling est in the world. Barrages are low, gated diversion-dams composed of a downstream along a thalweg transect, a methodology described initial- series of gates (usually 60–70) that control the elevation of an upstream ly by Smith and Reeves (2000). In 2006 and 2011, direct counts were ‘head pond’ maintained not to store water, but to divert it into lateral conducted from vessels travelling in tandem, separated by 1 km, and canals. mark-recapture was used to correct for groups that were missed, The first six barrages were commissioned at the end of the 19th cen- resulting in higher point estimates of abundance along with measures tury and were located on the rivers of Punjab. Completion of Panjnad of precision (Braulik et al., 2012a; Noureen, 2013). The metapopulation barrage in 1933 split the former range of the Indus dolphin into two, abundance estimates in all three surveys were similar, ranging between separating animals in the Indus River from those in the five Punjab trib- 1200 and 1750 individuals for the entire subspecies (Table 1). The larg- utaries. By 1940, the tributaries were already fragmented into seven dif- est and most important subpopulation is located between Guddu and ferent sections by barrages whereas barrage construction had only just Sukkur barrages in northern Sindh. This short, 190 km river section begun on the Indus mainstem and dolphins could move relatively un- supports approximately 70% of all Indus dolphins, at encounter rates impeded along its length until the completion of several barrages approaching 10 individuals/km. Dolphin encounter rate and subpopula- around 1960. The longest stretch of unfragmented habitat and the tion abundance decline in each river section as one proceeds upstream. 34 G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41

Fig. 2. Illustration of the total number of fragments of Indus dolphin habitat against year, and the length of the largest portion of unfragmented habitat against year. The red line denotes the onset of barrage construction and the light grey line, the mean fragment length.

Between Jinnah and Chashma barrages only a few isolated individuals increase in dolphins over the same 5-year period (Braulik et al., were recorded in 2001 and 2006, and it is possible that dolphins have 2012a). Braulik et al. (2012a) discussed in detail whether this apparent since been extirpated from this area. trend was real or not and concluded that the lack of confidence intervals on all of the counts precluded any firm conclusions about population 4.2.2. Trends in abundance between Guddu and Sukkur growth rates. However those authors also concluded that all surveys in- Braulik et al. (2012a) reported that from 1974 to 2008 at least 22 dicate an increase in the subpopulation between Guddu and Sukkur dolphin counts were conducted between Guddu and Sukkur barrages. since the 1970s. Counts showed a steady increase from 138 individuals in 1974 to 902 Until the 1970s, dolphins were hunted in Pakistan (Anderson, 1879; in 2008, a statistically significant rate of increase equivalent to approx- Pilleri and Zbinden, 1973–74). Hunting was banned by the Sindh Wild- imately 5.65% per year. Similarly, two different groups conducted life Act (1972) and, in 1974, the Indus Dolphin Reserve was established dolphin surveys between Guddu and Sukkur in 2001 and 2006, and al- between Guddu and Sukkur barrages. The probable increase in dolphin though the abundance estimates generated in each year were different abundance between Guddu and Sukkur likely signifies a degree of due to differences in methodology, both groups recorded a 60–65% population recovery following the cessation of dolphin hunting. It is

Table 1 Summary of Indus dolphin subpopulation and metapopulation abundance estimates and estimation methods during the last 12 years.

Year Method Jinnah- Chashma- Taunsa- Guddu- Sukkur- Total Meta-population Reference Chashma Taunsa Guddu Sukkur Kotri estimate

2001 DC 2 84 259 602 18 965 (Braulik, 2006) –775a –1140a 1200 (Braulik et al., 2012a) 2006 DC 1 82 44b 1279 4b 1410 (Braulik et al., 2012a) MR-TV – 101 (CV = 44.1%) 52b (CV = 14.9%) 1289 (CV = 33.4%) – 1447 (CV = 57.2%) 1550–1750 (Braulik et al., 2012a) 2011c DC – 87 465c 570d 34c 1312 – (Noureen, 2013) 726e MR-TV – 96 (CV = 19.02%) – 701 (CV = 9.63%) – 797 1452 (Noureen, 2013) 857e (CV = 21.3%)

DC = Direct Count; MR-TV = Mark-recapture on data from tandem vessels; a = count was revised upwards to account for animals in 33.3 km that was not surveyed; b = entire section was not surveyed due to security concerns; c = Sukkur-Kotri was surveyed in 2010, and Taunsa-Guddu in 2012. These two counts were added to the rest of the 2011 survey data so that the 2011 metapopulation abundance estimate is combination of the three d = 33.8 km of the Indus downstream of Guddu Barrage and a 31.8 km side channel were not surveyed; e = dolphin count includes sighting rates extrapolated from adjacent river sections to account for unsurveyed sections. G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 35 unlikely that the increase is due to improvements in habitat or prey In the early 1970s, dolphin hunting was widespread in Sindh and availability, as new dams and barrages continue to be constructed, the several boats were equipped for their capture (Pilleri, 1972). Although Indus flood cycle and sediment transport processes have been greatly the Muslim majority disdained dolphin meat for religious reasons, disrupted, dry-season discharge has declined, and levels of pollution non-Muslims (not clear which religion) continued to consume it. The have increased dramatically as Pakistan becomes industrialised. Counts dolphin became a protected species in 1974 when the Wildlife Acts of recorded from all other Indus dolphin populations over the last 15 years Sindh, Punjab and the Khyber Phakhtunkhwa (KP) Provinces were have remained stable (Braulik et al., 2012a). passed, and a reserve for the Indus dolphin was declared in the 190 km stretch of river between Guddu and Sukkur barrages. Within a 4.3. Short summary of Indus dolphin ecology and biology few years, and following some prosecutions in the courts, dolphin hunt- ing in Sindh ceased (Bhatti and Pilleri, 1982). With enforcement of the The diet of Indus dolphins appears to vary according to location and/or ban on dolphin hunting in Sindh, the hunters moved upstream into season, but is generally composed of a large variety of bottom-dwelling fish Punjab to avoid the strict controls downstream. It was reported in (including catfish and carp) and prawns (Butt, 1977; Pilleri and Zbinden, 1977 that upstream of Guddu barrage many boats were equipped to 1973–74; Roberts, 1997). Dolphins surface rapidly and unpredictably, catch dolphins. Reeves et al. (1991) reported that Kehal fishermen making photography and collection of biopsies challenging. They never below Kalabagh were engaged in hunting dolphins, and also that dol- approach boats or bow ride, and individual photo-identification is phin hunts took place at Chashma, Ghazi Ghat and Taunsa in the early impractical because most animals lack unique features. 1980s. There is no evidence that Indus dolphin hunting has continued Indus dolphins are generally observed alone or in groups of two or anywhere in Pakistan since then. three, occasionally in aggregations of as many as 20 or 30 individuals. Prior to the construction of dams and barrages, the dolphins moved up- 5.1.2. Fisheries interactions stream during the floods and downstream during the low-water season. Mortality from accidental capture in fishing gear is the greatest The maximum lifespan may be 33–35 years (Ohsumi, 1979). Fe- threat to most cetaceans (Reeves et al., 2013). Indus dolphins are acci- males grow larger than males and both sexes become sexually mature dentally captured in nets when they stray into irrigation canals which, at 7–10 years of age. Gestation likely lasts around 10–11 months due to their narrow and shallow dimensions, are easily and heavily (Brownell, 1984). Limited observations of very small, solitary individ- fished. Net entanglement is a major concern between Sukkur and uals feeding on fish indicate that South Asian river dolphins are weaned Kotri barrages where the Indus flow is so severely depleted that fixed quickly, probably within a few months after birth, a situation that could nets span the river. However, in general, the Indus River main channel occur where there are no predators and a reliable food supply. Indus has not been intensively fished as fishing activity is concentrated in dolphins probably have an annual or biannual reproductive cycle, side channels and adjacent pools that are reported to be warmer and which would explain their observed high rate of population increase have higher fish densities (Khan, 1947). The low intensity of fishing in relative to most other cetaceans. the main channel is also partly due to the swiftness of the current, Indus dolphins produce echolocation clicks but no whistles. In cap- which limits the manoeuvrability of oar-powered boats. However en- tivity and in the wild they echolocate almost continuously, producing tanglement is an increasing threat as boats become mechanised and 20–50 clicks per second (Herald et al., 1969; Pilleri and Pilleri, 1987). better able to negotiate the main channel. Click frequencies are primarily 50–80 kHz, with a secondary peak at In the past a fish contractor system was in place in which the rights 160–200 kHz and a dominant frequency of around 80 kHz (Pilleri to fishing grounds were auctioned by the government and purchased by et al., 1976). The acoustic emission field is highly directional, extending powerful fish contractors. Contractors allowed fishing only on the con- in two relatively narrow cones dorsally and ventrally in front of the dol- dition that fishermen surrendered approximately 75% of the fish catch phin (Pilleri and Pilleri, 1987). In shallow water and in captivity Indus to them, and that the remainder was sold to them below market value dolphins have been observed to swim on their sides (Herald et al., (Jabbar, 2005). In 2007 the contractor system in Sindh was abolished 1969). and now local indigenous fishermen can obtain their own licences to fish. This action is likely to affect the other provinces soon (Anon., 5. Threats 2011). Although the new system is believed to be more equitable for fishermen, it has led to larger numbers of unskilled fishermen using A summary of threats, mitigation measures and current research is the river, and a coincident jump in the number of dolphin deaths. provided in Table 2. From 1993 to 2010 a total of 35 dolphins were reported dead between Guddu and Sukkur, however in 2011 alone 45 dead dolphins were re- 5.1. Direct causes of mortality ported, and another 15 from January to May 2012 (Waqas et al., 2012). The 2011 total includes at least 6 dolphins that were killed within 5.1.1. Hunting the Protected Area when insecticides were dumped into the river to in- Detailed accounts of the hunting bags of British officers, often total- crease fish catch (WWF-Pakistan, 2011). Since 2012 mortality rates ap- ling thousands of birds and mammals of numerous species, were regu- pear to have declined: only six deaths were recorded from 2013 to 2014, larly published in the 1800s and early 1900s. We could trace no records possibly because fisheries enforcement and community awareness ac- that included a dolphin or porpoise, therefore Indus dolphins apparent- tivities have increased. Additional research on this issue is essential ly were not targeted by colonial hunters. and could include studies to investigate the factors that influence sus- Freshwater dolphins were however killed by numerous indigenous ceptibility to bycatch and ways to reduce entanglements. groups along the Indus River, in Sindh and Punjab, over the course of several centuries (Anderson, 1879; McNair, 1908). The meat was used 5.1.3. Canal entrapment as food, and the oil as medicine and for lighting. Lowis (1915) gave a de- Indus dolphins occasionally enter irrigation canals through the flow- tailed explanation of how dolphins were captured by the local Kehal regulator gates adjacent to irrigation barrages. Canals run for hundreds people in Dera Ghazi Khan. They constructed a viewing platform in shal- of kilometres and are heavily fished and readily accessible to people. low water, and attached a fish to a nearby stake. A tethered tame otter Therefore canal-entrapped dolphins are at high risk. Dolphins trapped was released into the water and would try to reach the fish. The noise in canals generally remain close to the main river, but some have been of the otter would attract a dolphin and as it approached the dolphin known to travel over 170 km (Bhaagat, 2002; Khan, 2005). Each year would attempt to catch the fish attached to the stake, at which point all canals are drained for several weeks to allow for maintenance the fisherman would cast his net over the dolphin, capturing it. work. Even if trapped dolphins survive until canal closure, they will 36 G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41

Table 2 Details of the threats to the Indus River dolphin ranked by priority (in descending order from top to bottom) with details on research and mitigation.

Threat Priority Current/future/ Increasing/ More Mitigation Mitigation at Mitigation References for historical decreasing research strategy local/regional/ underway subspecies required? national level

Migration through Highest All Stable Yes Unclear, possibly translocation Local—primarily at No (Braulik et al., 2014a; barrages or possibly deterrents Guddu and Sukkur Toosy et al., 2009) barrages Decreasing Highest All Increasing Not Improved basin-wide water National/regional No (Braulik et al., 2014a; water flow essential management, maintenance of Braulik et al., 2012b) minimum environmental flows Negative High Current & Increasing Yes Unclear Local. especially Little WWF-Pakistan, interactions future Guddu to Sukkur unpublished with fisheries Canal entrapment High Current & Stable No Continuation of rescues Local. Sukkur Yes (Braulik, 2002) future barrage (Waqas et al., 2012) Pollution Medium Current & Unclear Yes Increased sewage treatment, Local/national Indirectly (WWF-Pakistan, future reduced industrial effluents 2011) Climate change Lower Current & Increasing Not Improved basin-wide water National/regional No None but linked to future essential management, maintenance of decreasing water flow minimum environmental flows Low genetic Lower Current & Stable Yes Possibly translocation National No (Braulik et al., 2014b) diversity future Hunting Lower Historical Absent No Continued enforcement of Not needed Not (Anderson, 1879; now ban on hunting needed Pilleri, 1972) certainly die when the canal is drained, and therefore a dolphin rescue 2007). More than three quarters of all Indus dolphins occur in the programme was initiated in 1992 by Sindh Wildlife Department and Indus below the Panjnad River confluence and therefore live down- WWF-Pakistan. Between 1992 and 2014, 147 trapped dolphins were stream of cities inhabited by more than 100 million people. Studies of documented; of that number, 113 were rescued successfully while 34 Ganges River dolphins showed substantial tissue accumulation of died (Waqas et al., 2012). Some animals died during capture and the heavy metals, organochlorine pesticides and polychlorinated biphenyls cause of death was believed to be capture-related stress rather than (PCBs) (Kannan et al., 1993) as well as accumulation of butyltins 5–10 drowning in the net. Animals that survived capture always survived times greater than the levels found in their prey (Kannan et al., 1997). transport and release although post-release survival was not monitored. Commonly used pesticides DDT, Cypermethrin, Deltamethrin and Endo- The number of dolphins reported in canals varies dramatically from sulfan were found in the tissue of three Indus dolphins that died in year to year, presumably due to differences in the numbers entering in Sukkur in January 2011 (WWF-Pakistan, 2011). Given the high pollut- the first place but also to differences in the efficiency of detection, ant loads in the some of the rivers, South Asian river dolphins are likely which depends on many logistical factors. Reports have been received to be exposed to some of the highest levels of pollutants of all cetaceans, occasionally of dolphins trapped in canals that originate from Guddu both through prey consumption and direct contact. Although this expo- or Taunsa barrage, but this is rare probably because there are fewer dol- sure may result in negative health effects now or in the future, this issue phins present above these barrages, and perhaps because of differences has yet to be investigated in any depth for this subspecies. in the location of the canals and different river morphology near the bar- rages. Research to understand when and how dolphins traverse the 5.2.2. Freshwater supplies gates, and what influences this process, would help in designing mea- The greatest threat to Indus dolphins originates from the irrigation sures to reduce the problem. network, in the form of habitat fragmentation by barrages and degrada- tion or removal of habitat due to the extraction of water (Braulik et al., 5.2. Indirect causes of mortality 2014a). Low dry-season river discharge, due to water extraction at irri- gation barrages, was shown to be the principal factor responsible for 5.2.1. Water pollution 1) the spatial pattern of the dolphins' range decline, 2) the temporal The magnitude of surface water pollution in Pakistan has increased pattern of subpopulation extirpation and 3) the speed of extirpation fol- at a dramatic rate as the country has rapidly industrialised and it is lowing habitat fragmentation (Braulik et al., 2014a). The probability now recognised a significant problem (Directorate of Land that an Indus dolphin subpopulation is still extant increases with in- Reclamation Punjab, 2007; Qadir et al., 2007). As of 2005, only around creasing median dry-season river discharge. Where dolphins have al- 8% of urban and industrial waste water, by volume, underwent even pri- ready disappeared, they were extirpated earlier and more quickly mary treatment, allowing more than 90% of industrial and municipal ef- from river sections where discharge was lower. Dolphins were more fluent to enter rivers and lakes (World Bank, 2005). The plains are likely to persist in the core of the former range because water diversions intensively cultivated with cotton, wheat and sugar cane and pesticide have been concentrated near the range periphery. Habitat fragmenta- use has increased at an annual rate of about 6% (World Bank, 2005). In- tion and degradation are intertwined and in combination they have secticides sprayed on crops mix with irrigation water, which leaches caused the decline of the Indus dolphin. through the soil and enters groundwater aquifers and also re-enters Reductions in flow affect dolphins directly by reducing the physical water courses. The discharge from tanneries, textile mills, wood and space available to them, reducing average water velocity and depth jute mills, sugar mills, distilleries and pulp and paper factories is the and increasing water temperature. During the dry season Indus dol- major source of chemical pollution. The Punjab rivers flow through phins are concentrated in the remaining deep pool habitat in the river, the industrial and agricultural heartland of Pakistan and as a conse- and they avoid shallow river sections (Braulik et al., 2012b). They select quence are more polluted than the Indus mainstem, which passes areas of river that have large cross-sectional area and volume, especially through more remote areas (Directorate of Land Reclamation Punjab, at depths greater than 1 m, presumably to reduce the risk of becoming G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 37 isolated in small pools and to maximise foraging opportunities. Removal study on the Mekong River found that construction of all planned tribu- of water from the river exacerbates and concentrates anthropogenic tary dams there would have catastrophic impacts on fish biodiversity threats, for example, by increasing the concentration of nutrients and and on the floodplain's fish productivity, far greater than the combined pollutants, and forcing the dolphins to congregate in deep pools that impact of six upper mainstem dams on the lower Mekong River itself are also important areas for fishing, thereby increasing the chances of (Ziv et al., 2012). New dam construction will likely continue for some negative interactions with humans (Braulik et al., 2012b; Chaudhary time on the upper Indus and the consequences for people, fisheries et al., 2012; Kelkar et al., 2010). and dolphins located downstream are unclear. The statement, ‘In a land where it seldom rains, a river is like gold’ Fortunately, further fragmentation of Indus dolphin habitat by con- (Albinia, 2008), could not be more appropriate anywhere than in struction of new barrages is unlikely, as the current network is adequate South Asia. Access to water is one of the most politically charged issues for distributing water to the present agricultural area. If there is a need in Pakistan. The vast majority of the nation's water comes from the to irrigate additional land, it is likely that new canals rather than entirely Indus River, and the river passes through neighbouring India prior to new barrages will be constructed. entering Pakistan, which makes river discharge a sensitive matter. The finite surface water resources are under great pressure from a large and rapidly growing Pakistani population and an expanding economic 5.4. Population dynamics and downstream migratory attrition and agricultural sector. New dams, barrages, river linking projects, and hydropower developments are planned, many are already under con- For most of the year barrage gates are virtually closed to ensure that struction, and there is constant demand for more cultivable land. Per water is diverted into canals. Consequently only a small opening is pres- capita water availability has dropped to one of the lowest levels world- ent at the base of the gate and it would be difficult for a dolphin to phys- wide, and at present there is little culture of water conservation (Asian ically pass through this. Barrage gates are fully open and the river flow Development Bank, 2010). The projected demand for water is predicted unimpeded to allow passage of high floods, which generally last for a to outstrip availability before 2025 (Siddiqi and Tahir-Kheli, 2004). few days or weeks, and for three weeks at each barrage during low Consequently, the future of Indus dolphins is tied to much broader na- flows when the canals are closed for maintenance. During these brief tional issues of governance, security, poverty alleviation and water periods the flow through the barrage is still extremely rapid (greater management. than around 5 km/h in the dry season and much greater in the flood) and turbulent, but there is no physical obstruction preventing dolphins 5.3. Climate change from passing through a barrage (see Fig. 3). It has been suggested that dolphins sometimes move through bar- One response of cetaceans to climate change is that their ranges may rage gates and between subpopulations (Reeves et al., 1991). The only shift to remain within preferred climatic conditions (MacLeod, 2009). solid evidence of this is from a single radio-tracked dolphin released Species and populations such as many of the river dolphins that have re- above Sukkur barrage during canal closure that traversed the barrage stricted geographical distributions, with little or no opportunity for in both an up- and downstream direction several times. The barrage range expansion, are expected to be especially vulnerable. Under differ- gates were eventually closed, leaving the animal below the barrage, ent climate change scenarios, Indus river discharge was predicted to in- trapped in a new subpopulation downstream from its origin (WWF- crease dramatically (N90%) by 2050 (Palmer et al., 2008). This positive Pakistan, unpublished). This example from a single dolphin shows news is tempered by the fact that because of the great discrepancy be- that movement through Sukkur barrage is possible, but as each barrage tween water availability and withdrawals for human use, the Indus has a very different design and operation, as well as different dolphin was still predicted to remain one of the most water-stressed basins densities in adjacent portions of the river, this does not prove that (Kundzewicz et al., 2008; Palmer et al., 2008). The models suggest cross-barrage movement is a regular occurrence at Sukkur or that dol- that climate change alone is unlikely to spell doom for Indus dolphins phins can move through other barrages. and, if better water conservation practises are adopted in the future, However, the potential implications of such movements are great. water supplies may in fact increase, signifying a positive change for For example, if animals do regularly move through barrages and they the dolphins. are more likely to move downstream than upstream, the result would It is possible that, if global temperatures rise and a greater propor- be the gradual attrition of upstream subpopulations and the augmenta- tion of Indus flow is derived from rainfall as opposed to glacial melt, tion of those downstream. Migration of this type could dramatically de- river water temperatures will rise. Water temperatures in the mainstem plete upstream subpopulations over time, especially as many of these of the Indus River vary from approximately 5 °C in mid-winter to at least are already very small, potentially leading to their gradual extirpation 33 °C in early summer, an annual temperature range of almost 30 °C (Reeves et al., 1991). (Braulik, unpublished). Indus dolphins have evolved the capability to Several aspects lend weight to the downstream migratory attrition cope with large temperature fluctuations, so they may be more resilient hypothesis: to climate change-driven increases in water temperature than species with more uniform habitat. However it is still possible that the effects a) Many dolphin subpopulations located in the upper reaches of rivers of climate change on the dolphins' prey or broader ecosystem will have been extirpated over the last century and it is those down- prove detrimental to them. stream that have persisted (Braulik et al., 2014a). b) Each subsequent downstream subpopulation, except the last, is larg- 5.3.1. Dams and barrages er than the preceding one, despite a continually diminishing river Pakistan depends almost completely on irrigated agriculture for em- flow (Braulik, 2006). The exception to this is the small subpopula- ployment, to feed its large and booming population (185 Million, 1.6% tion furthest downstream between Sukkur and Kotri barrages that growth per annum in 2014) and as the major source of its international persists despite severely depleted river discharge rendering the hab- exports. Large dams on the Indus River are often advocated as the itat marginal. It is possible that this subpopulation is augmented by, means to satisfy demand for electricity and stimulate economic growth. or consists solely of, migrants from the upstream subpopulation Pakistan has plans to add 10,000 MW through five projects by the year (Guddu to Sukkur). 2016, and another 14 projects totalling about 21,000 MW are under c) Dolphins enter irrigation canals through gates that are very similar study for construction by 2025. Meanwhile, India has many dams to barrage gates. Once dolphins enter canals they are usually unable under construction or planned on the Indus tributaries where they to travel back upstream through the canal gates and return to the flow through Indian Territory prior to entering Pakistan. A recent Indus River. As dolphins are known to pass downstream through 38 G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41

Fig. 3. Photographs showing 1. a gate at Taunsa barrage which is virtually closed and 2. a gate at Guddu barrage which is fully open because it is canal closure and the barrage is not di- verting water.

canal gates regularly, it seems likely that they also pass downstream effectiveness of existing strategies. A functioning stranding network through similar barrage gates. would also allow the collection of other important life history and health data that are at present almost completely lacking. Such a net- work would be the first of its type in Pakistan and given the lack of vet- The magnitude of dolphin migration at each barrage would likely erinary capacity, infrastructure (power, freezers etc.) and technical vary based upon differences in engineering design, operational cycle, di- expertise, it would require considerable investment to get off the version capacity and location as well as dolphin density in each subpop- ground. Clearly it would be important to begin with a basic system ulation. Barrage permeability would determine subpopulation that could be expanded over time. immigration and emigration rates, and therefore whether migration re- sults in a net attrition or augmentation of that subpopulation. For exam- ple, if the downstream migration rate at a barrage is high, the 6.2. Protected areas (PAs) subpopulation upstream would suffer rapid attrition. Alternatively, if the downstream migration rate at a barrage were low, the upstream The Sindh Dolphin Reserve is one of very few freshwater PAs desig- subpopulation would contribute few migrants downstream and may in- nated specifically to protect river dolphins. The Reserve was established stead exhibit its own net increase from upstream immigrants. Sukkur in response to a specific threat, that of dolphin hunting, which at that barrage diverts more water than other Indus barrages and its gates are time was determined to be the major threat to the subspecies. The ap- therefore lowered, or closed, for a larger part of the year. High dolphin parently consistent increase in dolphin abundance since reserve estab- abundance between Guddu and Sukkur barrages may therefore be the lishment indicates that it was very effective at combating that threat result of high immigration through Guddu barrage and low emigration (Braulik et al., 2012a). However, other than the ban on dolphin hunting through Sukkur barrage, resulting in an overall augmentation of the there are few restrictions on human activities within the Reserve—very Guddu to Sukkur subpopulation by downstream migration (Braulik, limited management of fishing activity, pollutant discharges and vessel 2006). traffic. As human populations increase and new threats emerge, it is likely that the Reserve as presently designed and managed will become 6. Conservation options and research priorities less effective at conserving dolphins. Managers need to find ways to adapt and respond to changing conditions if the Reserve is to continue 6.1. Mortality monitoring providing dolphins with meaningful protection. Meanwhile, the section of river from Taunsa to Guddu barrage in The demonstrated link between extirpation of Indus dolphin sub- Punjab was also declared a PA for dolphins in 2014, and there is interest populations and low river discharge is important for macro-level in gazetting other stretches of river in KPK Province. Obviously, the sim- conservation strategies. However, there is still very little information ple creation of PAs does not guarantee the long-term survival of vital on what the immediate or proximal causes of death of individual ecosystems or endangered species without carefully considered and im- Indus dolphins are. In many threatened or endangered long-lived mam- plemented management. Among a total of 25 PAs in northern Pakistan, mals, variation in mortality (particularly adult mortality) is a primary 16 lack baseline information, 22 do not have any management plan, and determinant of population growth. Therefore, describing the causes 19 are without any management infrastructure (Nawaz, 2007). As such, and rates of mortality is an important component of conservation re- they are ineffective at protecting the environment. Given that national search to guide conservation actions. An important step is to establish governance is very weak, and large parts of the river are outside of gov- a national stranding network so that animal deaths, associated bio- ernment control and instead run by tribal landlords, legislated PAs will logical information, and determination of causes of death can be docu- not be effective without community involvement. Many of the people mented systematically. This would involve reporting of deaths to a who live by the river are among the very poorest communities and designated authority, systematic collection of data that are stored in a they have relied on the river for their livelihoods for generations. There- central location, collection and analysis of tissue samples using standard fore community-based conservation is likely to be the most effective protocols, and training of responders in necropsy techniques. Until such conservation strategy whether inside or outside a formal PA framework. a system is in place, it will not be easy to identify or respond to peaks in A system of community- and ecosystem-based management, with mortality that may be cause for concern. Without information on the in- zoned PAs that include highly protected reserves in critical areas as cidence of Indus dolphin deaths and their causes it is not possible to de- well as buffer zones that allow human uses such as carefully managed sign conservation measures to reduce mortality, or easily monitor the tourism and fishing, would be ideal. This has been successfully G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 39 implemented for river dolphins in the Sundarbans of Bangladesh (WCS/ Translocation programmes typically have varied goals that include Bangladesh Cetacean Diversity Project, 2014). bolstering genetic heterogeneity of small populations, establishing sat- Freshwater PAs have characteristics that are quite different to those ellite populations to reduce the risk of species loss due to catastrophes, in other ecosystems because they tend to be greatly affected by activi- and speeding recovery of species after habitat restoration (Carpenter ties that occur outside their boundaries, as water arrives from upstream et al., 1989). If upstream Indus dolphin subpopulations are declining or runs off from nearby terrestrial areas. Practises such as dam building due to downstream migratory attrition and individuals are concentrat- or diverting water for agriculture can occur outside PA boundaries and ing between Guddu and Sukkur barrages, then translocating dolphins yet have major negative consequences for habitat within the PA. As out of this high-density subpopulation to supplement the low numbers such, the dolphin PAs in Pakistan are likely to be most effective for con- in upstream areas may be sufficient to prevent the extirpation of several servation when attempting to manage localised threats such as fishing, very small subpopulations. hunting, vessel traffic or point sources of pollution, whereas for other The subpopulation between Guddu and Sukkur barrage is relatively important threats such as depleted river flows, dams and distant pollut- large (estimated as 1289 in 2006), occurs at high density (10.35 km−1), ant sources a river basin-wide approach will be needed. appears to have been increasing over the last 30 years (Braulik et al., 2012a) and could likely sustain the removal of a modest number of in- 6.3. Downstream migration dividuals per year. In general, translocation success is highest when an- imals are wild-caught and originate from a high-density, increasing Discussion of whether or not dolphins move through irrigation bar- population (Carpenter et al., 1989). rages and in what directions and numbers permeates most of the ques- Capturing dolphins in the wide, fast-flowing river would be fraught tions surrounding the conservation of Indus dolphins. Obtaining a with difficulty and danger to both dolphins and capture team and would greater understanding of this issue is of high priority because, depend- be ill-advised. However, dolphins become trapped in irrigation canals ing on how many individuals move and in which predominant direc- every year and are rescued and returned to the river (Waqas et al., tion, the persistence of upstream subpopulations may be threatened, 2012). These animals would die if they were not captured and therefore and there may be a continual loss of dolphins to areas of marginal hab- are ideal candidates for translocation. In general the dolphin rescue pro- itat downstream. It is not possible to measure or count the number of gramme in Pakistan demonstrates that these animals are fairly robust dolphins that move through a barrage visually unless individuals actual- during transport, the greatest risk of mortality being during capture ly surface inside a gate which has seldom been seen. Tracking dolphins (Waqas et al., 2012). However it is important to note that we know using radio tags may show that single animals have traversed a barrage, nothing of survival after release as this has not been monitored. Animals but tags can't be deployed on sufficient animals to be able to quantify captured and transported to overseas dolphinaria in Japan, Switzerland population-scale movements. Satellite and GSM tags aren't feasible to and the USA in the 1960's did not survive at all well (Herald et al., 1969; use on these river dolphins because they spend insufficient time at the Kasuya, 1972; Reeves and Brownell, 1989). surface (about 1 s) to obtain or transmit a signal. Barrages are politically A translocation programme would need to be well-researched and sensitive structures so it is difficult to get permission from the govern- carefully planned prior to implementation. It would be necessary to ment to attach any sort of acoustic or electronic device to them. Anchor- consider animal welfare, impacts on the founder subpopulation, habitat ing fixed passive acoustic recorders in, and on either side of, the and threats for the receiving subpopulation, and how to achieve long- barrages was attempted (Braulik, unpublished) and the rapidly flowing term monitoring of released individuals and the whole population. At water and large amount of floating and submerged debris in the river a minimum the following components would need to be investigated meant that anchor lines had to be cleaned hourly to prevent fouling thoroughly: and loss of the expensive devices. It appears theoretically possible to use acoustic tracking to record animals moving through a barrage, but • Habitat. Without good habitat quality and low threat levels in the it would be necessary to have the devices inside the mouth of every range of the receiving subpopulation, translocation has little chance gate—again logistically difficult and expensive. The turbidity of the of success regardless of how many organisms are released or how water precludes the use of underwater video. It is possible that high- well they are prepared for release (Carpenter et al., 1989). Before an resolution, multi-beam acoustic imaging such as that used for counting Indus dolphin translocation programme could be considered serious- migrating salmon could be used for quantifying dolphin movement ly, detailed studies of habitat and threats in the receiving environment (Rawding and Liermann, 2011). would be required. Previous studies have shown that channel geome- As it is not feasible to collect biopsies from these dolphins with cur- try and river morphology are significantly different and more suitable rently available methods, the use of genetics to provide information on for dolphins between Guddu and Sukkur barrages than in the habitats such things as abundance, sex ratio, reproductive success or changes in of subpopulations upstream (single sinuous deep channel, versus genetic diversity over time will be challenging, if even possible. Rescued shallow braided channels) (Braulik et al., 2012b). However, river dis- animals are a source of genetic material, but these all originate only charge is greater in upstream areas, and it was shown, based on phys- from the largest sub-population and it is those upstream that are of ical characteristics, that these upstream subpopulations should be greatest concern. able to persist for several centuries (Braulik et al., 2014a). Preliminary observations indicate threat levels to be similar in all areas, but pollu- 6.4. Translocation tion is likely to be greatest downstream. • Monitoring individual survival. It is not possible to observe or photo- Dolphin subpopulations in the Indus River system are being slowly graph Indus dolphin individuals well enough in the field to allow for extirpated primarily from the upstream portions of their range their identification using either natural or man-made marks. Dead an- (Braulik et al., 2014a). There appears to be a domino effect with subpop- imals are seldom found because they wash downstream with the cur- ulations upstream disappearing prior to those immediately down- rent. Therefore survival of released animals would need to be stream. The Jinnah-Chashma subpopulation has only a handful of accomplished by radio tracking. animals and is on the brink of extirpation, while the Chashma-Taunsa • Monitoring subpopulations. Regular population monitoring to identi- subpopulation is estimated at 96 individuals (Noureen, 2013) and the fy changes in subpopulation abundance would be essential. A translo- Beas River subpopulation consists of only approximately 10 animals cation programme would be determined to be successful if the (Behera et al., 2008). The possibility has been raised of translocating founder supopulation was not severely affected by the removals, Indus dolphins from the high-density, largest subpopulation (Guddu and the small recipient subpopulations upstream either increased in to Sukkur) to a subpopulation with low abundance. size or were at least maintained at current levels. 40 G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41

Indus dolphins are not yet so threatened that the loss of a few indi- sinus), or the Mekong Irrawaddy dolphins (Orcaella brevirostris){IUCN, viduals would outweigh the advantages of maintaining additional sub- 2015 #3109}. However it remains one of the world's most endangered populations. If conducted cautiously and with careful planning, such a mammals and its long-term prognosis is not at all good. Despite a bold management action is worthy of serious consideration because great increase in knowledge about Indus dolphins over the last 20 the long-term conservation benefits of establishing or maintaining ad- years, there are still many important unanswered questions that are ditional Indus dolphin subpopulations may well be worth the risks. vital to address if the species is to be conserved effectively (Table 2). Do dolphins move through barrages and thus between subpopulations? 6.5. River management Are upstream subpopulations declining in abundance? How much water is enough to sustain a dolphin population? How is the new legis- Irrigation barrages on the Indus River will not be removed and advo- lation concerning fisheries and protected areas affecting the dolphin cating removal as a dolphin conservation strategy would be unrealistic population? Of top research priority is the need to understand and and counter-productive. If Indus dolphins are to persist, they must do quantify movement of dolphins through barrages and the implications so within the current configuration of fragmented habitat. The strategy of this on population dynamics. Although challenging to assess, there that is most likely to be successful for Indus dolphins is managing the are realistic options for mitigation of dolphin movement through bar- Indus River so that the natural hydrological regime is at least partially rages that could be implemented and help solve the problem. By con- restored or maintained. Determination of the environmental flows or trast the other high-priority threat, limited dry-season river flows, is a the quantity and quality of water required for ecosystem preservation far-reaching issue that will require political solutions, as well as social and resource protection is an integral part of modern river basin man- interventions and ecological understanding; it will require investment agement, and essentially means that enough water is left in a river to and influence that is way beyond species conservation. ensure downstream environmental, social and economic benefits It is important that we do not allow ourselves to become paralysed (Dyson et al., 2003). It requires negotiations among stakeholders to rec- by the lack of certainty about some threats, and that the search for oncile the different interests that compete for the use of water, especial- more information does not become the sole focus. It is essential that ly in river basins such as the Indus where competition is already fierce. It concrete conservation actions be taken immediately. These should requires that relationships among flow, suitable habitat and species are focus on what is known: that declining river flows threaten the dol- understood and described by linking the physical properties of river phins, small upstream subpopulations are especially vulnerable to extir- stretches, e.g. depth and flow velocity, at different measured or pation and need rigorous protection, and fisheries-related mortality is modelled flows, with the physical conditions that key animal or plant an increasing problem. With reference to the extinction of the baiji, species require (Tharme, 2003). Describing the flow needs of river dol- the Scientific Committee of the International Whaling Commission phins could be a large, expensive and complicated exercise. It would (IWC, 2008) stated that “despite extensive scientific discourse for have to include specific evaluations of river discharge, hydrology and more than two decades, little effort was made to implement any real dolphin habitat use over a range of flows and seasons and ideally conservation measures for this species. In hindsight, the extinction of would include information on dolphin life stages, reproduction and for- this species is not surprising; species cannot be expected to save them- aging, all information that is currently incomplete or lacking entirely. selves”. Conservation and research action is urgently needed to prevent Hydrology–ecology relationships frequently exhibit non-linear re- the Indus dolphin from succumbing to the same fate as the baiji. sponses to flow, and thresholds based on those responses can be instru- fl mental for managing environmental ow programmes (Shafroth et al., Acknowledgements 2010). Given that extraction of water from the Indus River is one of the primary threats to Indus dolphins, and anticipating that some kind Reviews of earlier versions of this manuscript were provided by fl of environmental ow study will be conducted in the future, it would Simon Northridge and Phil Hammond. I thank Andy Read and Vincent be prudent to begin gathering data on Indus dolphin habitat use so Janik for the original suggestion to write this manuscript. Valuable that they are available as input to such a study. funding for preparation of the paper was provided by the US Marine Water scarcity is experienced by almost all Pakistanis. As stated by Mammal Commission (E4047595). the World Bank (2005), “the survival of a modern and growing Pakistan is threatened by [lack of] water. The facts are stark”. There is frequent tension between the provinces over the allocation of river References water, with Sindh, located downstream, being especially sensitive to Adams, A.L., 1867. Wanderings of a Naturalist in India: The Western Himalayas and Cash- its vulnerable position (Shah, 2009). Despite this tension and the atten- mere. Edmonston and Douglas, Edinburgh, UK. dant publicity, there is almost no concept of water conservation in Ahmad, M.F., 1999. The wildlife of the Indus River and allied areas. In: Meadows, A., Meadows, P.S. (Eds.), The Indus River: Biodiversity, Resources, Humankind. Oxford Pakistan; the focus is almost entirely on obtaining and capturing more University Press, Pakistan, pp. 3–11. water from the rivers and from ground water. The irrigation system is Albinia, A., 2008. Empires of the Indus. The Story of a River. John Murray, London. extremely inefficient, water delivery is unreliable and inequitable, and Anderson, J., 1879. Anatomical and Zoological Researches: Comprising an Account of the Zoological Results of the two Expeditions to Western Yunnan in 1868 and 1875 and a crop yields per cubic metre of water are below international standards Monograph of the two Cetacean Genera Platanista and Orcella. Bernard Quaritch, and those of neighbouring countries (World Bank, 2005). Forty percent Piccadilly, London. of the water diverted from the Indus basin in Pakistan is lost in convey- Anon., 1846. The Sikh's and the Punjab. United Service Magazine 214. Anon., 2011. Fishermen Caught in Contract net. In Dawn, Lahore. ance and in the late 1980s it was estimated that improvements in sup- Asian Development Bank, 2010. Asian Development Outlook 2010: The Future of Growth 3 ply efficiency could save 14.8 billion m /yr. of water. Canal lining is one in Asia. Asian Development Bank, Mandaluyong City, Philippines. such improvement (World Commission on Dams, 2000). Water con- Behera, S.K., Nawab, A., Rajkumar, B., 2008. Preliminary investigations confirming the oc- servation and good stewardship would go a long way to improving currence of Indus River dolphin Platanista gangetica minor in River Beas, Punjab, India. J. Bombay Nat. Hist. Soc. 105, 90–126. the water resources situation in Pakistan, and this would have far- Bhaagat, H.B., 2002. Status, population abundance, strandings and rescues of Indus blind reaching benefits for society as well as for river ecosystems and Indus dolphin (Platanista minor) in River Indus (Pakistan). Tiger Paper 29, 9–12. dolphins. Bhatti, M.U., Pilleri, G., 1982. Status of the Indus dolphin population (Platanista indi BLYTH, 1859) between Sukkur and guddu barrages in 1979–1980. Investig. Cetacea 13, 245–262. 7. Conclusions Blanford, W.T., 1901. The distribution of vertebrate animals in India, Ceylon and Burma. Philos.Trans.R.Soc.Lond.B194,335–436. Braulik, G.T., 2002. Entrapment of Indus dolphins (Platanista minor)inirrigationcanals: The conservation status of the Indus River dolphin is not yet as dire incidence, implications and solutions. International Whaling Commission, Scientific as that faced by the baiji (Lipotes vexillifer), the vaquita (Phocoena Committee Document SC/52/SM9, Cambridge. UK. G.T. Braulik et al. / Biological Conservation 192 (2015) 30–41 41

Braulik, G.T., 2006. Status assessment of the Indus River dolphin, Platanista gangetica Moazzam, M., 1999. Fisheries of the Indus estuarine system. In: Meadows, A., Meadows, minor,March–April 2001. Biol. Conserv. 129, 579–590. P.S. (Eds.), The Indus River: Biodiversity, Resources, Humankind. Oxford University Braulik, G.T., Arshad, M., Noureen, U., Northridge, S., 2014b. Habitat fragmentation and Press, Pakistan, pp. 62–72. species extirpation in freshwater ecosystems: causes of range decline of the Indus Nawaz, M.A., 2007. Status of the brown bear in Pakistan. Ursus 18, 89–100. River Dolphin (Platanista gangetica minor). PLoS ONE. Noureen, U., 2013. Indus River Dolphin (Platanista gangetica minor) Abundance Estima- Braulik, G.T., Barnett, R., Odon, V., Islas-Villanueva, V., Hoelzel, A.R., Graves, J.A., 2014c. tions Between Chashma and Sukkur Barrages, in the Indus River, Pakistan, In Depart- One species or two? Vicariance, lineage divergence and low mtDNA diversity in geo- ment of Plant Sciences. Quaid-e-Azam University, p. 50. graphically isolated populations of South Asian River dolphin. J. Mamm. Evol. 1–10. Ohsumi, S., 1979. Interspecies relationships among some biological parameters in ceta- Braulik, G.T., Bhatti, Z.I., Ehsan, T., Hussain, B., Khan, A.R., Khan, A., Khan, U., Kundi, K., ceans and estimation of the natural mortality coefficient of the southern Hemisphere Rajput, R., Reichert, A.P., Northridge, S.P., Bhaagat, H.B., Garstang, R., 2012a. Robust minke whale. Report for the International Whaling Commission 29, 397– 406. abundance estimate for endangered river dolphin subspecies in South Asia. Endan- Pakistan Wetlands Programme/WWF-Pakistan, 2008. Illegal trade in freshwater turtle ger. Species Res. 17, 201–215. parts: preliminary findings of turtle trade in major barrages and head-works of Pun- Braulik, G.T., Reichert, A.P., Ehsan, T., Khan, S., Northridge, S.P., Alexander, J.S., Garstang, R., jab, Islamabad. 2012b. Habitat use by a freshwater dolphin in the low-water season. Aquat. Conserv. Palmer, M.A., Liermann, C.A.R., Nilsson, C., Florke, M., Alcamo, J., Lake, P.S., Bond, N., 2008. Mar. Freshwat. Ecosyst. 22, 533–546. Climate change and the world's river basins: anticipating management options. Braulik, G.T., Smith, B.D., Chaudhry, A.A., 2012a. Platanista gangetica ssp. minor, In The Front. Ecol. Environ. 6, 81–89. IUCN Red List of Threatened Species. Version 2014.3. http://www.iucnredlist.org Pilleri, G., 1972. Field observations carried out on the Indus dolphin Platanista indi in the (Downloaded on 25 November 2014). winter of 1972. Investig. Cetacea 4, 23–29. Brownell Jr., R.L., 1984. Review of reproduction in Platanistid dolphins. Report of the Pilleri, G., Gihr, M., Purves, P.E., Zbinden, K., Kraus, C., 1976. On the behaviour, bioacoustics International Whaling Commission, 149–158. and functional morphology of the Indus River dolphin (Platanista indi,BLYTH,1859). Burnes, A., 1835. Travels into Bokhara. Being the Account of a Journey from India to Investig. Cetacea 6, 13–70. Cabool, Tartary and PersiaAlso Narrative of a Voyage on the Indus from the Sea to Pilleri, G., Pilleri, O., 1987. Indus and Ganges river dolphins Platanista indi (Blyth, 1859) Lahore Carey. Hart & Co., Philadelphia, Baltimore. and Platanista gangetica (Roxburgh, 1801). Investig. Cetacea 20, 2–33. Burnes, A., 1842. Cabool: Being a Personal Narrative of a Journey to and Residence in That Pilleri, G., Zbinden, K., 1973–74. Size and ecology of the dolphin population (Platanista City, in the Years 1836, 7 and 8. John Murray, London. indi) between Sukkur and Guddu Barrages, Indus River. Investig. Cetacea 5, 59–70. Butt, J.A., 1977. Food of the freshwater dolphin (Platanista minor Owen) from the River Qadir, A., Malik, R.N., Husain, S.Z., 2007. Spatio-temporal variations in water quality of Indus, Dera Ismail Khan, Pakistan. Biologia 23, 91–94. nullah Aik-tributary of the river Chenab, Pakistan. Environ. Monit. Assess. 1–17. Carpenter, J.W., Griffith, B., Reed, C., Scott, J.M., 1989. Translocation as a species conserva- Rawding, D., Liermann, M., 2011. Comparison of DIDSON sonar based estimates of chi- tion tool: status and strategy. Science 245, 477–480. nook salmon escapement with other methods in the Coweeman River, Chaudhary, S., Dey, S., Dey, S., Sagar, V., Nair, T., Kelkar, N., 2012. River dolphin distribution Washington. J. Acoust. Soc. Am. 129, 2692. in regulated river systems: implications for dry-season flow regimes in the Gangetic Reeves, R.R., 1998. Conservation status of the Indus River dolphin in Pakistan. IBI Rep. 8, basin. Aquat. Conserv. Mar. Freshwat. Ecosyst. 22, 11–25. 1–13. Directorate of Land Reclamation Punjab, 2007. Surface Water Quality Monitoring Plan. Ir- Reeves, R.R., Brownell, J.R.L., 1989. Susu Platanista gangetica (Roxburgh, 1801) and rigation and Power Department, Government of Punjab, Lahore, p. 27. Platanista minor Owen, 1853. In: Ridgway, S.H., Harrison, R.J. (Eds.), Handbook of Ma- Federal Flood Commission, 2010. Flood Daily Report. Government of Pakistan, Ministry of rine MammalsThe First Book of Dolphins vol. 4. Academic Press Ltd., London, Water and Power. pp. 69–99. Flow. In: Dyson, M., Bergkamp, G., Scanlon, J. (Eds.), The Essentials of Environmental Reeves, R.R., Chaudhry, A.A., 1998. Status of the Indus River dolphin Platanista minor.Oryx Flows. IUCN, Gland, Switzerland and Cambridge, UK. 32, 35–44. Hall, J.H.W., 1848. Scenes in a Soldier's Life: Being a Connected Narrative of the Principal Reeves, R.R., Chaudhry, A.A., Khalid, U., 1991. Competing for water on the Indus plain: is Military Events in Scinde, Beeloochistan, and Affghanistan, During 1839, 1840, 1841, there a future for Pakistan's river dolphins? Environ. Conserv. 18, 341–349. 1842 and 1843. B & C Chalmers, London. Reeves, R.R., McClellan, K., Werner, T.B., 2013. Marine mammal bycatch in gillnet and Herald, E.S., Brownell, J.R.L., Frye, F.L., Morris, E.J., Evans, W.E., Scott, A.B., 1969. Blind river other entangling net fisheries, 1990 to 2011. Endanger. Species Res. 20, 71–97. dolphin: first side-swimming cetacean. Science 166, 1408–1410. Roberts, T.J., 1997. The Mammals of Pakistan. Revised edition. Oxford University Press, Isaac, N.J.B., Turvey, S.T., Collen, B., Waterman, C., Baillie, J.E., 2007. Mammals on the London & Tonbridge. EDGE: conservation priorities based on threat and phylogeny. PLoS ONE 2, 3. Shafroth, P.B., Wilcox, A.C., Lytle, D.A., Hickey, J.T., Anderson, D.C., Beauchamp, V.B., Islam, B.N., Talbot, G.B., 1968. Fluvial migration, spawning, and fecundity of Indus River Hautzinger, A., McMullen, L.E., Warner, A., 2010. Ecosystem effects of environmental Hilsa, Hilsa Ilisha. Trans. Am. Fish. Soc. 97, 350–355. flows—modeling and experimental floods in a dryland river. Freshw. Biol. 55, 68.–85 IUCN, 2011. Pakistan Water Gateway: Gateway to Water Information About Pakistan. Shah, S., 2009. Thal Canal to Cause Decline in Sindh Agri Output (In The News). IWC, 2008. Report of the scientific committee. J. Cetacean Res. Manage. 1, 1–80. Siddiqi, T.A., Tahir-Kheli, S., 2004. Water needs in South Asia: closing the demand–supply Jabbar, R., 2005. Pakistan Fishermen in Sindh Protest Against a Contract System In The gap. Global Environment and Energy in the 21st Century. East–West Center, Honolu- News. lu, Hawaii. Jerdon, T.C., 1874. The Mammals of India; a Natural History of All Animals Known to In- Smith, B.D., Reeves, R.R., 2000. Survey methods for population assessment of Asian river habit Continental India. Wheldon, London, UK. dolphins. In: Reeves, R.R., Smith, B.D., Kasuya, T. (Eds.), Biology and Conservation of Kannan, K., Senthilkumar, K., Sinha, R.K., 1997. Sources and accumulation of butyltin com- Freshwater Cetaceans in Asia. IUCN, Gland, Switzerland & Cambridge, U.K., pounds in Ganges River dolphin, Platanista gangetica. Appl. Organomet. Chem. 11, pp. 97–115. 223–230. Tharme, R.E., 2003. A global perspective on environmental flow assessment: emerging Kannan, K., Sinha, R.K., Tanabe, S., Ichihashi, H., Tatsukawa, R., 1993. Heavy metals and or- trends in the development and application of environmental flow methodologies ganochlorines residues from Ganges River dolphins from India. Mar. Pollut. Bull. 24, for rivers. River Res. Appl. 19, 397–441. 159–162. Toosy, A.H., Khan, U., Mahmood, R., Bhagat, H.B., 2009. First tagging with a radio- Kasuya, T., 1972. Some informations on the growth of the Ganges dolphin with a com- transmitter of a rescued Indus River dolphin near Sukkur barrage, Pakistan. Wildl. ment on the Indus dolphin. Sci. Rep. Whales Res. Inst. 24, 87–108. Middle East 3, 6. Kasuya, T., Haque, A.K.M.A., 1972. Some informations on distribution and seasonal move- Waqas, U., Malik, M.I., Khokhar, L.A., 2012. Conservation of Indus River dolphin (Platanista ment of the Ganges dolphin. Sci. Rep. Whales Res. Inst. 24, 109–115. gangetica minor) in the Indus River system, Pakistan: an overview. Rec. Zool. Surv. Kelkar, N., Krishnaswamy, J., Choudhary, S., Sutaria, D., 2010. Coexistence of fisheries with Pak. 21, 82–85. river dolphin conservation. Conserv. Biol. 24, 1130–1140. WCS/Bangladesh Cetacean Diversity Project, 2014. Research on Freshwater Dolphin Ecol- Khan, H., 1947. A fishery survey of River Indus. J. Bombay Nat. Hist. Soc. 46, 529–535. ogy and Human Activities in Three Wildlife Sanctuaries in the Eastern Sundarbans Khan, U., 2005. Cetacean Conservation: A Case Study of the Indus Dolphin Rescue Project: Mangrove Forest, Bangladesh In Background Document Prepared by the Wildlife A Collaboration for Rescuing Stranded River Dolphin from Sindh to Punjab. WWF- Conservation. Society's Bangladesh Cetacean Diversity Project, Khulna, Bangladesh. Pakistan, Lahore, p. 18. World Bank, 2005. Pakistan Country Water Resources Assistance Strategy. Agriculture Kundzewicz, Z.W., Mata, L.J., Arnell, N.W., Doll, P., Jimenez, B., Miller, K., Oki, T., Sen, Z., and Rural Development Unit, South Asia Region. World Bank, p. 144. Shiklomanov, I., 2008. The implications of projected climate change for freshwater re- World Commission on Dams, 2000. Dams and Development. A New Framework for sources and their management. Hydrol. Sci. J. 53, 3–10. Decision-making. Earthscan Publications Ltd., London and Sterling. Lowis, R.M., 1915. Gharial, Gavialis gangeticus, and porpoise, Platanista gangetica,catching WWF-Pakistan, 2011. Report on Indus River Dolphin Mortality: Analysis of Dead Dolphin in the Indus. J. Bombay Nat. Hist. Soc. 23, 779. Samples for Pesticides. WWF-Pakistan, Lahore. MacLagan, R., 1885. The rivers of the Punjab. Proc. R. Geogr. Soc. Mon. Rec. Geogr. 7, Xiong, Y., Brandley, M.C., Xu, S., Zhou, K., Yang, G., 2009. Seven new dolphin mitochondrial 705–719. genomes and a time-calibrated phylogeny of whales. BMC Evol. Biol. 9. http://dx.doi. MacLeod, C.D., 2009. Global climate change, range changes and potential implications for org/10.1186/1471-2148-1189-1120. the conservation of marine cetaceans: a review and synthesis. Endanger. Species Res. Ziv, G., Baran, E., Nam, S., Rodríguez-Iturbe, I., Levin, S.A., 2012. Trading-off fish biodiver- 7, 125–136. sity, food security, and hydropower in the Mekong River basin. PNAS http://dx.doi. McNair, J.F.A., 1908. Oral Tradition from the Indus. R. Gosden, Brighton. org/10.1073/pnas.1201423109.