Baseline Environmental Study

1. Introduction

1.2 Scope of services Eight key areas of investigation were covered by the study: geomorphology, groundwater resources and condition, flora, fauna, farming systems, social anthropology, archaeology, and geographical information systems. The key findings for each of these areas are outlined in summary form below.

2. Geomorphology

2.1 Geomorphology provides a structure for describing and explaining the patterns and processes described by other facets of the study.

2.2 The region of the KNP is made up of carbonate limestones and sandstones that have been uplifted and compressed by the Cenozoic collision of the Indian subcontinent with Asia. The result is a series of NS trending anticlines and synclines cut by normal faults.

2.3 The KNP today is a basin-range desert, dominated by alternating mountains and plains. In EW cross-section there is a typical sequence of geomorphic units: the overlying sandstone units have been stripped off leaving cuestas of sandstone around a limestone core. The mountains are cut by deep gorges that pass onto a piedmont that is characterised by the following units: bedrock-cut pediments that are now dissected by alluvial fans that terminate in a plain, and finally a floodplain. The piedmont plain may be subdivided into plains cut in bedrock (pediments) and plains formed of alluvium (alluvial plains). The latter includes alluvial fans and base-level plains (such as river valley floors or playas (wind-eroded surfaces)).

2.4 Overall, nearly half of the study area consists of bedrock mountains with little sediment cover. Thirty percent is plains (of which one third (650 km 2 ) is not covered with stone and is available for agriculture), and twenty percent is alluvium (fans and floodplains).

2.5 In general the landscape of KNP is dominated by erosion rather than deposition. Over 1000m of rock has been eroded over the last 5 million years or so (a rapid denudation rate of 0.5mm y -1 ), but there is remarkably little deposition in the valleys (less than 10s of metres). Much of the material has been removed in solution. 2.6 Soils are either poorly developed, or absent from the great majority of the study area.

2.7 Several sites of geomorphic significance have been identified in the park including escarpments, gorges, the Khar knickpoint, and various other sites.

2.8 Erosion rates have almost certainly increased on the plains as a result of grazing pressure. Wind erosion of the surface of the plains unit is high, estimated to be 2- 4 mm per year.

3. Groundwater resources and condition

3.1 Groundwater discharging from springs or pumped from wells provides the only source of water to sustain wild and domestic animals, irrigation and human use in Khirthar National Park. This Chapter of the Baseline Study reports an assessment of groundwater resources in Khirthar National Park based on existing information, brief inspections in February/March and September 2000, and analyses of water samples taken during those periods.

3.2 Discharge of water from springs that were inspected is estimated to range from nil to 10 MLd -1 . Other areas of surface water that may be springs in Khirthar National Park have been identified by interpretation of satellite imagery.

3.3 Water from all but one sample from 15 springs and 20 wells that were sampled failed to meet World Health Organisation (WHO) guidelines for the bacteriological quality of drinking water. Water meeting the guidelines came from a well that was reported to be more than 80 m deep.

3.4 In general terms, the water samples were brackish to saline and slightly to strongly alkaline with ionic composition consistent with the regional geology. Some of the water samples had reported concentrations of sulphate, cadmium or lead in excess of common guidelines for drinking water. Water samples were not tested for diesel or its degradation products. However, diesel spillage from pumps was observed in several wells, and it is likely that there is a degree of contamination in these cases.

3.5 Stable isotope concentrations in springs and wells show that rainfall had entered the shallow aquifer with little loss by evaporation.

3.6 During the drought year of 2000, evapotranspiration from irrigated land in Khirthar National Park is estimated to have been around 19,000 ML, very much greater than estimated amounts drawn for drinking by people (around 800 MLy -1 ) and domestic animals (around 430 MLy -1 ), or taken by wild animals (around 30 MLy -1 ). When all irrigable land is in use, evapotranspiration from irrigated crops in Khirthar National Park is estimated to total around 180,000 MLy -1 .

3.7 Groundwater recharge to aquifers in the Taung Valley is estimated to average 20 mmy -1 , which is equivalent to around 7500 MLy -1 over the area of the valley below the 350 m contour. In the drought year 2000, evapotranspiration from irrigated crops in this valley is estimated to have been about 4700 ML, but in years of maximum irrigation it would be around 24,000 MLy -1 , reflecting a large increase in the area of irrigated land during good conditions. These figures are subject to errors of estimation, but the groundwater resources of the Taung Valley are probably over-developed.

3.8 A model of groundwater flow in the Taung Valley suggests that transmissivity of the uppermost 100 m of sediments is around 250 m 2 d -1 in the upper region of the valley, around 400 m 2 d -1 in the central region, and around 1500 m 2 d -1 in the lower regions. The average hydraulic conductivity of the aquifer material to an assumed depth of 100 m appears to be higher in all regions than earlier estimates based on the texture of strata exposed in wells and eroded gullies. If the rate of groundwater recharge in the valley has been over-estimated, then aquifer transmissivity would be closer to values inferred from the texture of exposed material.

3.9 Rainfall in a normal monsoon season is estimated to increase the average groundwater level in the Taung Valley by around 0.2 m. This seasonal increase of groundwater levels will be much greater near the bed of a river and negligible more than a few kilometres from any river.

3.10 The groundwater model indicates that continued irrigation of even the reduced area irrigated in 2000 would result in substantial decreases of groundwater level in the Taung Valley, particularly in upper and mid-valley regions. Irrigation of all land in the Taung Valley that was used for this purpose in 1994 would result in an average long-term declining trend of groundwater levels of around 0.4 my -1 .

3.11 Evapotranspiration from irrigated crops will result in increasing salinity of groundwater in Khirthar National Park. However, water that is drawn from the dug wells is unlikely to be affected to a measurable extent, and any increase of salinity of water drawn from springs or deeper boreholes is unlikely to be measurable except over a period of 20 years or more.

4. Flora

4.1 A total of 475 plant species were recorded for the Khirthar National Park from a sample of more than four hundred 900 m 2 quadrats surveyed during the periods February to April, and September to October, 2000. The vegetation survey is estimated to have recorded at least 67% of the likely park flora, so that a total flora in the vicinity of 700 species might be expected if a full census of species were able to be conducted.

4.2 Approximately 30% of species (n=133) are summer ephemerals. These species are present only as dormant, soil stored seeds for most of the year. They germinate following the summer monsoon rains, rapidly grow, flower, set seed and die, typically completing their life cycles by November. Grasses and herbs dominate the ephemeral flora.

4.3 A small number of species either new to science, or new to the region, were discovered during the field surveys: the small mountain-top herb Lepidagathis sp. is new to science. It is restricted to a single known location within the Khirthar National Park. Two woody shrubs (Pycnocycla sp. and Campylanthus sp.) of a rocky fan area on the eastern side of Khirthar Mountain are new records for Pakistan and Sindh Province respectively, and constitute major components of a rare and sensitive community type. Several species previously unrecorded from Sindh (e.g. Celtis australis, Pistacia chinensis) were found in canyons and on mountain tops. For a number of species further work is required to clarify their taxonomic status.

4.4 Mountain-tops and deep canyons represent particularly sensitive and important reservoirs of high species richness and rare species presence in this region. Canyons appear to have acted as refuges for species of more mesic environments, with a different sub-set of relictual species surviving in different canyons. Botanical exploration of canyons is likely to yield more new species for the park.

4.5 Species richness is correlated with altitude; the plains and valleys have low diversity on average, while the mountains have high diversity. This trend may reflect both natural (more mesic environmental conditions in the mountains) and human-related (human settlement, agricultural development and livestock grazing and browsing impacts in the lowlands) factors.

4.6 Investigation of the vegetation of captive breeding enclosures at Khar Centre revealed more plant species inside than outside, suggesting a possible negative human impact on species richness and vegetation condition in the park. Indices of wood harvesting and livestock grazing impacts on plants were also significantly higher outside the enclosures (which have been in existence for only 6 years). This evidence points to a very large and widespread human impact on the vegetation and habitat in the park. This impact is led most strongly by the grazing of goats and other domestic livestock and the cutting of vegetation for forage and other purposes.

4.7 Classification (Twinspan) and ordination (MDS) analysis of vegetation survey data identified 10 major vegetation communities within the park, including one human land-use community (irrigated orchard) and nine natural communities. The natural communities are: 1. Wetlands, 2. Riparian woodlands, 3. Sparse sandy riparian woodlands, 4. Plains woodlands, 5. Sparse stony shrublands, 6. Stony foothills woodland, 7. Rocky river bed shrublands, 8. Rocky slopes woodlands and 9. Cliffs, peaks and canyons.

4.8 Major differences in species composition exist between wetlands, riparian vegetation, vegetation of mountains versus plains, and for agricultural fields. Differences between assemblages within the mountains and plains respectively, are more subtle.

4.9 The community types vary considerably in their spatial cover and human impacts as well as in their relationships to key environmental variables such as proximity to water, substrate type, altitude, slope and geomorphology. There is a strong, but not perfect, correlation between geomorphic unit type and vegetation type. A vegetation map of the park has been prepared based on the classification of vegetation quadrats and the relationships between vegetation types and geomorphic units.

4.10 Size frequency distributions of stem diameters for woody species revealed a generally low level of recruitment of new individuals to species populations. This implies that populations of some species may decline as mature individuals die without replacement. Human impacts through stem and branch lopping (for fodder, fencing and fuel) and livestock browsing may be a major cause of this recruitment failure, especially in the valleys and plains.

4.11 Estimated annual rates of wood production are close to estimated rates of human usage of wood by residents within the park. Because wood biomass distribution varies markedly between vegetation types, it is likely that wood resources are being over-exploited in many of the valley and plains areas of the park. Riparian vegetation zones are more dense and likely more productive, but are very limited in their spatial extent. Further human population increase within the park threatens the long-term viability of the major woody species of the valleys and plains.

4.12 Invasive species are relatively few. Prosopis juliflora and Prosopis glandulosa are the major potential problem species at the present time. They are gradually increasing their abundance within the park, mostly from areas in the south-east and south-west, near centres of major human population. Other invasive species are closely associated with irrigated fields and show no evidence of spread into non-irrigated lands. No invasive species were recorded for disturbed sites such as road verges and pipelines.

5. Fauna

Part A: Fauna survey

5.1 The faunal component of the Khirthar National Park Baseline Environmental Study was established to determine the distribution and abundance of representative animal species in the park and relate those to other components of the park. Faunal field work was carried out by staff of the Sindh Wildlife Department, Zoological Survey Department and the University of Melbourne. It took place over five periods totaling 11 weeks from February to November 2000. It involved detailed faunal data collection at 194 sites throughout the park, together with incidental observation at other sites, and a helicopter-based aerial census of large ungulates. Data were entered on computer spreadsheets and analyzed statistically.

5.2 The faunal survey concentrated on terrestrial vertebrates as representative of the park fauna, with some additional data being collected on invertebrates. Seven different faunal habitat types were identified and used as a basis for documenting species distributions and abundance. Eight different sampling techniques were used to locate species and estimate their abundances.

5.3 277 vertebrate species were detected in the park during the survey: 203 species of birds, 36 species of mammal, 34 species of reptile and 3 species of amphibians, and distribution and abundance data collected on each.

5.4 The report sets out, for each vertebrate species, details of its distribution in the park as a whole and across the individual habitat types, together with a map showing where it was detected. For the more abundant species, habitat features significantly associated with each species are set out, including site characteristics, associated plant species and any associations with invertebrate taxa. There is also comment on the species status if it has been identified as either a Red List species or one scarce in Sindh or Pakistan as a whole.

5.5 Larger invertebrates were identified to the Class or Order Level only. Their distributions and abundances were related to the various habitat types, proving to be most abundant in croplands and sandy plains sites, and least abundant in wetlands. Some were associated with particular vertebrate species.

5.6 Of the vertebrate species listed for KNP, 6 species of birds, 8 species of mammal and one species of reptile appear on the IUCN Red List of internationally endangered species. An additional 25 species of birds and 7 species of mammals appear to be of conservation significance; some of these seem to have been lost from KNP in recent years.

5.7 There has been a long history of human occupation in KNP, over which much of the fauna may have adapted to the presence of human activity. Only in recent years, when the human population rose several-fold, have several vertebrate species with large home ranges declined greatly in numbers, or disappeared altogether.

5.8 Although the distributions and abundances of animal species vary according to the species, it is possible to identify common patterns between species. Multivariate analysis of the distribution and abundance data on the more abundant vertebrates identified five groupings potentially useful as a park management tool. The groupings are: wetland species, farmland species, species of rocky sites, domestic livestock and associated species, and a group of generalist species. The approximate composition of these groups is set out. However there is insufficient information to group species present at only a very small number of sites.

5.9 Highest vertebrate biodiversity in KNP is found in wetlands and croplands, and the lowest in mountain sites. However, from a conservation viewpoint, the wetland and rocky sites groupings contain the highest proportion of rare species, with a small number of generalists also rare. The farmland areas appear be of relatively little importance to many of the rarer species.

5.10 A helicopter survey conducted in November yielded estimates of the total populations of the three large indigenous ungulates in the park. The Sindh ibex population was estimated at 13,155 ± 2460, and concentrated on the Khirthar Range, with lower concentrations on Khambu and Dumbar and small numbers elsewhere; higher elevations appear critical to this species. The Sindh urial population was estimated at 10,425 ± 675 and concentrated on rocky sites with characteristic vegetation mainly near Khar and at Dumbar, with small numbers elsewhere. The chinkara population was estimated at 1060 ± 580 and concentrated in the lowlands, sharing much of its habitat with farming. The KNP populations of all three species are of conservation importance.

Part B: Management options for the Sindh Ibex

5.11 This section of the fauna report describes a habitat and stochastic metapopulation model of the Sindh Ibex (Capra aegagrus) in Khirthar National Park. Field observations and published information on preferred habitat were used together with ArcView GIS (Geographical Information Systems) software and a digital terrain model to develop a habitat model on the basis of elevation and slope. Patches were associated with an important food plant.

5.12 The habitat model was used as a basis for developing an explicit metapopulation model, including a transition-matrix population model within each patch, dispersal and dependence. The model was implemented using RAMAS/Metapop (Akcakaya 1999). The program was used to explore management options and to evaluate the importance of parameters and assumptions by sensitivity analysis.

5.13 Some of the suggested management alternatives included re-introduction of natural predators, stronger restrictions on illegal harvest, and legal harvest from the adjacent game reserve.

5.14 Under the model assumptions, the current Ibex population of about 13,000 is likely to remain generally stable over the next 20 years. However, plausible future events including increasing human population pressures and competition from domestic animals might precipitate important population declines, unless primary habitat is carefully protected.

6. Farming Systems

6.1 A study was conducted of the current farming systems in Khirthar National Park, from February to October 2000. The main objective of the study was to provide a broad description of the main types of farming practised by the park's inhabitants, in acknowledgement of the significant impact that might have occurred through a long history of occupation. The study also sought to speculate on the possible future directions of farming in the park and the implications of such changes for management policies for KNP.

6.2 Four separate field studies were conducted during the period, covering general farming practices, herders and flocks, dryland field histories and wells & pumps. One hundred and forty-seven individual case studies were completed. Numerous informal discussions were held with groups, individuals and officials. Official data was used, where available, together with published maps and aerial photographs. The Farming Systems teams visited many locations throughout the park, conducting village walks and field transects. Sixteen hours of video film were collected.

6.3 Three distinct types of land-use could be identified. Irrigated fields covered the smallest land area but were the most productive form of cropping. Most irrigation water is pumped from dug wells and drilled bores - powered by diesel engines - but camel-powered water wheels and natural springs remain important sources. Winter is the main season for irrigated crops: wheat (mainly for consumption) and onions (mainly for sale) occupy the greatest proportion. Cotton is the main irrigated summer crop but planted on a much smaller area. Many vegetable crops are grown throughout the year on small plots, for both consumption and sale.

6.4 The dryland fields cover very large areas of the park and have an extremely long history. Using a system of retaining banks and supply channels, they trap run-off water and sediments from nearby hill and mountain catchments, following monsoonal rainfall. The trapped water percolates to the sub-soil; once the surface has dried, it is tilled and seeded. Short-term shot summer crops—millet, sorghum, cluster bean and mung bean are most commonly planted, often as mixtures. Use of the dryland fields is entirely dependent upon rainfall in the catchment. Many areas of the park had not planted dryland crops since 1994, because of the prolonged drought.

6.5 Livestock are a vital component of most farms. On dryland-dependent farms, livestock may be the only source of income from year to year. Multi-purpose, animals are a source of food (usually milk), income, power and a store of wealth. Goats and sheep are the most numerous specie, probably followed by cattle, donkeys and camels. Livestock feeds include crop residues and browsing, plus lopped branches and shrubs. Many are herded on a nomadic basis, covering large areas of the park in the search for feed.

6.6 The main farming systems draw from all or some of these components. Families with access to irrigated lands, dryland fields and animals have the most diverse and stable production possibilities. A greater number of families have acccess only to dryland fields and animals: intermittent cropping of the drylands meant almost total dependence upon animals. Some remote communities are completely dependent upon animals and nomadic herding.

6.7 Trees are another vital component of all systems, providing firewood, animal feed, construction materials, shade, fruit and more.

6.8 Water is the dominant factor in all systems. Six years of drought (by February 2000) had resulted in many wells drying, greatly reduced areas of irrigation and no dryland crops planted in most regions. Many people had migrated from the park to the Indus Plain, taking some of the livestock. This migratory cycle is a long-established strategy for drought-prone areas such as Khirthar.

6.9 Most people lived extremely poor, hard and uncertain lives. Poverty was particularly pronounced in remote areas. Access to social and physical infrastructure was limited. Many believe that living in a National Park contributes to these circumstances. Most would like to see a clearer delineation between "land for the park" and "land for the people".

6.10 Water was the dominant topic of discussion—farmers wanted more water, bores, dams and pumps. They believed that this would increase the stability and viability of their farms, and boost the low levels of income and welfare.

6.11 It is estimated that the park now contains a human population ranging from 70,000- 100,000 people, depending upon climatic conditions. Heavy monsoon rains might see the figure rise occasionally to 150,000 people. Animal populations will vary with climate, from minima of 74,000 goats, 20,000 sheep, 13,000 cattle and 4,000 other specie after prolonged drought periods, to 110,000, 38,000, 17,000 and 5,000 respectively, following one or two years of good rainfall.

6.12 Irrigated areas are estimated to currently occupy a minimum of 3,500 ha, after a prolonged drought but might extend to 14,000 ha in favourable conditions. Dryland farming is believed to have affected at least one-third of the park's area at some time, although a much smaller area would have been cropped. Grazing livestock have probably foraged over half the park's land area (150,000 ha), much of it at a very low intensity of feeding.

6.13 It is predicted that the future will see a slow but continuous increase in human population and, inevitably, more animals and land. Water will come under even greater pressure than at present, because of the desire to increase irrigation areas fed by bores and wells. All such activity can only increase tensions between the conservation and human activities of park.

6.14 A number of recommendations are made on measures to improve farmers' welfare through the introduction of more efficient farm practices. Positive policy steps are needed to enable everyone to benefit from being residents of Pakistan's largest National Park.

7. Archaeology

7.1 A total of 78 archaeological sites were recorded during the two-week field survey conducted in September 2000. These sites range in age from approximately 5000 years BP (before present) to approximately 100 years BP.

7.2 Only sites with visible surface remains were recorded during the survey. No excavation or sub surface testing was carried out. Survey work was carried out in four defined districts of the park, with occasional recording of sites in other areas, as observed during transit from one area to another. The survey concentrated on the flatter, more open valley floors and along permanent water-courses. The arid tops of mountain ranges were not surveyed.

7.3 Types of site recorded include: lithic scatters, rock art sites, multi-period habitation sites, hydraulic works, monumental defensive architecture, single structures, and cemeteries.

7.4 The survey revealed traces of archaeological sites which are highly significant in local, regional and global contexts. These include an extremely important but visually unimpressive rock art site and the visually spectacular, but comparatively modern Rani Kot fort.

7.5 Issues regarding management of archaeological sites in the park must take into account: protection of sites from natural destructive processes, including wind and water erosion; damage caused by human activities such as pastoralism, agriculture, land clearance, and vandalism; access to significant sites for tourism, educational and recreation purposes; and preservation of antiquities for future excavation and archaeological examination.

8. Social anthropology

8.1 Settlement patterns in the park are characteristic of the rural population of Sindh Province. The majority of the population live in small hamlets (gothro/goth) comprising lineage based groups, sometimes clustered to form villages. Hamlets range from five to 50 households. The main distinction is between hamlets and towns (tandos), the latter having populations of over 1000 people and some services such as schools, medical clinics and stores.

8.2 The population has been underestimated previously and in-migration has increased. The fertility rate appears similar to the national (5.4). The available data indicate that between 54,000 and 70,000 are permanent residents. In-migration following monsoonal rains can increase the population to around 100,000. At 2.65% increase each year, by 2025 the population would be 140,000.

8.3 Housing in permanent settlements comprises small mud-brick dwellings around earthen courtyards with low fences, sometimes made of thornbush. Some settlements have houses made of brush and mud over a wooden frame. These dwellings are probably indicative of relatively recent settlement.

8.4 Government administration of the park area is from the town of Thano Bula Khan except for the small SW section that is in Malir District, where Gadap is the major town. Courts, land administration, educational and health services are located in the towns. Park inhabitants sell agricultural products at Thano Bula Khan markets and in Karachi.

8.5 Non-Government Organisation activities are limited, mainly concentrating on health provision.

8.6 Services are poor and in many places non-existent. Utilities such as reticulated water and electricity are not available within the park settlements. Roads are unsealed and unsuitable for standard vehicles. However there are numerous private buses and trucks that provide freight and passenger services. Inhabitants walk or use camels for local travel.

8.7 Fuel for cooking is mainly wood collected from the park. Families buy petroleum fuels for lamps, generators and machinery. Water availability varies but is from wells, sometimes pumped. There are no sewerage systems, waste is buried or dumped away from settlements.

8.8 Educational facilities comprise just over 100 primary schools for boys and eight for girls. In many places there are buildings but no teachers. There are 3800 registered male students and 310 female students enrolled according to official records. Absenteeism is high for both students and teachers, and parental commitment to education generally low. Koranic education is given in some villages but literacy rates are extremely low.

8.9 Health services vary greatly between towns, where they are basic and understaffed, and villages where they are sporadic and sometimes non-existent. Many medical posts in town centres were unfilled and scheduled trips to rural clinics were irregular. Basic medical facilities exist at the hospital in Thano Bula Khan and records show 12,900 people were treated in 1999. Major surgery cases go to Karachi. Immunization programs exist and official estimates suggest 90% of urban and 40% of rural children are vaccinated against TB, polio, tetanus, diphtheria and measles. There is a bias towards males in all medical treatments.

8.10 Maternal and Child Health in the rural areas are poor and all interviews endorsed the general findings for major studies by World Health Organisation of mothers and infants in rural Pakistan. Maternal deaths are common and medical attendance at births rare. Women depend on religious healers, patent medicines and herbs rather than seeking medical treatment. Very few women use family planning and sexual health programs are not available. Anaemia and malnutrition rates are reportedly higher among females than males because of differences in diet.

8.11 Local Government in the park area comprises Union Councils which have consisted of political appointees since 1993. The councils at present are primarily administrative bodies and manage service delivery and maintenance of state facilities and utilities. About half the park population is registered to vote, 20% are women.

8.12 Law and Justice issues are generally resolved according to traditional practices without recourse to police or the state judicial systems. Police report crime rates are low and former problems with dacoits have declined. Disputes are resolved at local level by male elders or by informal courts where Malik Asad, his representatives or other caste leaders preside.

8.13 Local Economy is based on cash-cropping with pastoral herding of goats and sheep also providing a crucial contribution to subsistence. Commerce is limited to store keepers in the larger villagers and most artisans and shopkeepers are based in the towns. Villagers are served by hawkers and sometimes travel to towns to obtain goods. In times of low productivity men and women migrate to other areas to seek work as labourers. The most significant group of wage labourers were employed by the Sindh Wildlife Department.

8.14 Patterns of Agricultural Production and Land Use reveal a range of entrepreneurial uses of land and other resources that are exclusively managed by men. The diversity and ingenuity of poor farmers attempting to maximise production and income reveals that people are aiming to expand agriculture in ways that have profound implications for park management. Ten case studies examine variation in land tenure and resource utilisation. Pastoral production has suffered because of the drought but remains an essential economic strategy.

8.15 Social Organisation retains many elements of a feudal system based on tribal divisions with a hierachy of male leadership over patrilineally defined castes or lineages. Major landowners have retained control over large areas of land in spite of land reforms and remain politically dominant over their groups. Family Structure is patrilineal with patrilocal residence and marriage between the children of brothers as the preferred system. Marriages are arranged and child betrothal is common. Polygyny occurs but is generally restricted to those who are relatively well-off. Families are large and there is no family planning.

8.16 Gender relations are characterised by a patriarchal system of authority whereby women have lower status than men and limited autonomy. Women are disadvantaged in respect of access to health care and education. The sexual division of labour is strict, with women undertaking all household tasks and childcare as well as some agricultural labour. Only men herd animals. Seclusion occurs among Baloch groups, but all villagers are protective of women and access to hamlets by strangers is thus restricted.

8.17 Culture is dominated by Islamic traditions although there is a minority group of Hindus, mainly living in towns. A large Hundu shrine is located at Mol. There are a number of votive shrines to Islamic saints in the park area which are culturally significant sites. Islamic life-cycle and religious ceremonies are observed with festivities.

8.18 A postscript has been included with some recommendations relating to a new management plan for the park. This emphasises three main areas of need that were consistently identified in interviews: transport infrastructure, education and health. The attitude of local inhabitants to KNP and their views of the ways it effects their lives are crucial elements in establishing a succesful management plan.

9. Geographical Information Systems

9.1 To address the needs of Component 1 fieldwork, a relatively course scale Geographic Information System (GIS) was established from 1:250,000 scale mapping. This course scale database has been replaced by a high resolution 1:50,000 scale database that represents the final Khirthar National Park Baseline Study GIS.

9.2 The Khirthar National Park Baseline Study GIS has been developed using Environmental Systems Research Institutes ArcView (Version 3.2 ) software. All data is stored as ArcView Shapefiles and can readily be converted to other spatial data formats.

9.3 Base mapping was derived from 1:50,000 scale maps provided by the Survey of Pakistan. This includes key data layers such as hydrology, topography and road networks. These maps have been converted to digital form using manual digitising techniques.

9.4 A high resolution 20 metre Digital Elevation Model (DEM) of the Khirthar National Park has been created. ANUDEM surface interpolation software was used to create the DEM. ANUDEM is a software program developed to produce accurate digital elevation models with sensible drainage properties from small to very large, topographic datasets. In addition to the DEM, a number of derived spatial data products have been created include slope, aspect and hillshade maps for use in park management and flow direction and flow accumulation maps for use in hydrologic analysis.

9.5 An important component of the study is the integration of satellite imagery into the GIS to support the research aims. The study integrated two satellite images for agricultural crop, vegetation and geomorphological mapping. These included a Landsat TM image from 1994 (30 metre resolution) and an IRS Panchromatic image captured in February 2000.

9.6 The development of accurate and representative climate surfaces for the KNP requires excellent spatial and temporal accuracy and completeness. Owing to the topographic variation in the KNP, localised temperature and rainfall variations are large. Unfortunately, the only reliable rainfall and temperature records were only available for a limited number of stations including Karachi Airport and Hyderabad. This is an inadequate spatial representation to build a representative rainfall and temperature surface for the KNP.

9.7 Owing to the large volumes of botanical and zoological data derived from the two field seasons, and the need to integrate this data into a GIS, and efficient and modular relational database management system (RDMBS) has been developed. A RDMBS has been developed using Microsoft Access and linking this dynamically to ArcView GIS.

9.8 Metadata is data about data. A detailed metadata dictionary has been completed for all spatial datasets and describes the key feature including its scale, custodian, source, accuracy and lineage of the data. The metadata dictionary will be a key component to ensure the long-term utility of the GIS data. The metadata dictionary is provided in Microsoft Word and is associated with each spatial dataset.

9.9 As a major component of the capacity building aims of the KNP Baseline study GIS software and computer hardware has been obtained to provide an in-country GIS capability to the Sindh Wildlife Department. Three copies of ArcView GIS 3.2 have been obtained from an Environmental Systems Research Institute Conservation Program grant. In addition, three PC's have been donated by Micropro Computing Australia.

9.10 The capacity building component of the KNP baseline study has resulted in two workshops being conducted. The first was held in February 2000 and introduced stakeholders to general GIS concepts. The second workshop is being held in February 2001 to introduce advanced GIS concepts.

10. Concluding remarks

10.1 Khirthar National Park houses a variety of physical, biological and cultural resources that have significant conservation value (see Figure 1).

10.2 A large, and growing, human population, large livestock populations, and extensive agricultural land use within the park is having inevitable negative impacts upon the flora (through wood harvesting, livestock grazing/browsing, and land clearance for agriculture), especially in lowland areas. Impacts also affect the fauna, through land and resource alienation, and landscape, mostly through wind erosion in the lowlands.

10.3 In order to safeguard the future of the natural and cultural resources of the park it is imperative that a park management plan be established and implemented. This plan should incorporate both conservation on the one hand, and community development on the other. Only through a process that improves the livelihoods of indigenous peoples within the park, and gives them a greater incentive to identify with the conservation goals of the park, will the flora, fauna and cultural heritage of the region be safeguarded for the benefit of present and future generations.

10.4 To be effective, it is likely that any such management plan would need to be implemented through a Board of Management that included a broad spectrum of stakeholder groups.

Back to Main