Biodiversity Of Punjab Shiwaliks Environmental Sciences Essay

There is an increasing awareness that biodiversity is not only intimately interconnected with long term health and vigor of the biosphere as an indicator of global environment but also as a regulator of ecosystem functioning. Tropical communities are often worse susceptible to loss of biological diversity than temperate communities because tropical species are occur in lower densities and are less widely distributed and often have weaker dispersal capabilities. Increasing human intervention and excessive exploitation of resources have resulted in great changes and provide alarming signals of accelerated biodiversity loss. The conventional species level approach for biodiversity management has major limitations. A major change in the understanding the priorities of biodiversity conservation and management has resulted in a policy shift from conservation of single species to their habitats through interactive network of species at landscape level is considered important (Orians, 1993; Edward et al., 1994).

Biodiversity can be represented at various levels of organization, like alleles or genotypes within a population, species or ecosystem across a landscape or even a planet. There is a growing need for better understanding of the biodiversity distribution pattern vis-à-vis human interventions. India have very varied environmental conditions and are among the countries with high biodiversity. The flowering plants, which provide maximum direct benefit to mankind, comprise about 19395 taxa, which is about 7% of the described species in the world (Karthikeyan, 2000). The vast stretches of coastal belt in south and high mountains species of Himalayas in the north provide a complex environmental set up or niches for plant and animals. It has resulted in the formation micro-endemic centres of plants in the country. The great heights and complex terrain of the Himalayan region, which change frequently, provide endless microclimatic conditions suitable to the species to grow and evolve. The Himalayas have more than 5 micro-endemic centres. Shiwalik ranges are one such mirco-endemic centre. About 125 wild relatives of crop plants have been reported in the western Himalayas and Shiwaliks ranges are part of these. These ranges are among the youngest hills in the world and are active as well as fragile. Shiwalik hills of Punjab state have rich biological diversity and act as major natural green belt of the state. The state is well known for agricultural produce and industries. But the degradational activity coupled with the destruction of forests in the name of developmental activities has altered the natural landscape of the region to a great extent. It is facing threat for its very existence again from mankind in the name development. Because of these increased anthropogenic activities, as a result of population explosion and change in land use practices, the natural landscape has been modified which has resulted in fragmentation of forests with poor species composition. Hence the resulting landscape mosaic is a mixture of natural and human managed patches that vary in size, shape and arrangement. Now it is realized that we must move from the conservation of single species to scales beyond individual sites and levels of organization. Understanding landscape spatial pattern is important since it contains all levels of the biological hierarchy, from ecosystems to species and genes, which are targeted for biodiversity conservation.

Remote sensing technology is being widely used world over for the quick assessment of the Earth resources. Because it being a cost effective and repetitive in nature with synoptic coverages technology, has endless application potentials. The technology assumes significance in terms of conservation of natural resources as it provides ‘bird’s eye view’ of the ecosystems or landscapes or regions and their status, because conservation is now not limited to a plant or species but much beyond that. Vegetation is composed of several communities and these communities may occur in patches of small to large size and in various shapes. Vegetation composition of these patches is governed by several environmental factors, and there exists a relationship of the biological richness with area of the forest and the influence of biotic factors. Ecologists have established the relationship among these environmental and biotic factors with the biodiversity. Patches can be characterized based on their size, shape, location, area etc. in through geospatial modeling in Geographic Information Systems (GIS). Thus various factors can be simultaneously considered and processed in GIS.

In this report, vegetation type map derived from satellite data was considered as prime input for landscape ecological analysis of forest ecosystem. Geographic Information System (GIS) is used to derive landscape indices such as fragmentation, porosity, patchiness, patch density, interspersion and juxtaposition, which depict landscape characteristics. These indices were integrated with biotic pressure zones to depict disturbance gradient in the study area. Phytosociological data collected from field sampling was analyzed to derive species richness, biodiversity value and ecosystem uniqueness of various forest types. Ancillary databases such as proximity from the roads/villages, terrain complexity etc. were derived in GIS domain. All above inputs were integrated in systematic manner by assigning relevant weightages to derive maps showing disturbance gradient and biological richness. The resultant maps highlight areas that are biologically rich.

2. Study Area

2.1 Physical & Climatic Status

The study was carried out in the Shiwalik hills of Punjab state having geographical area of 9448.97 km2 and situated in north western part of the state. It lies between latitude 30o 34′ 10″.82″ and 32o 33′ 02.96″ N and longitude 74o 50′ 30.30″ and 76o 52′ 51.26″ E. It is broadly divided into sub-mountainous Himalayas and the eastern and western alluvial plains. The important rivers draining the region are the Sutlej and the Beas. The average annual rainfall varies between 400 to 600 mm and the mean annual temperature ranges from 22.500 C to 25.00 C.

2.2 Geology

The composition of the Shiwalik deposits shows they are nothing else than the alluvial detritus derived from the sub aerial waste of mountains, swept down by their numerous rivers and streams and deposited at their foot. This process was very much like what the existing river systems of the Himalayas are doing at present day on their emerging to the plains of Punjab. An important difference is that the former alluvial deposits now making up the Shiwalik systems have been involved in the latest Himalayan systems of upheavals, by which they have been folded and elevated into their outermost foot-hills, although the oldest alluvium of many parts of northern India serves to bridge the gap between the newest Shiwaliks and the present alluvium.

2.3 Lithology

The Shiwalik system is a great thickness of detrital rocks, such as coarsely bedded sandstones, sand-rock, clay and conglomerates, measuring between 4,500 and 5,200 m in thickness. The bulk of formation is very closely similar to the materials constituting the modern alluvia of rivers. The lithology of the Shiwliks suggests their origin; they are chiefly the water-worn debris of the granitic core of the central Himalaya, deposited in the long and broad valley of the Shiwaliks. The upper coarse conglomerates are the alluvial fans or talus-cones at the emergence of the mountain streams; the great thickness of the clay and sand represents the silts and finer sediments of the river laid down in the lower plain. The weathering of the Shiwalik rocks has been proceeding at an extraordinally rapid rate since their deposition, and strictly abrupt forms of topography have been evolved in this comparatively brief period. Gigantic escarpments and dip-slopes separated by broad longitudinal strike valleys and intersected by deep meandering ravines of the transverse streams-surface features, which are the most common elements of Shiwalik topography. The strike is remarkably constant in a Northwest-Southeast direction, with only brief local swerves, while it is almost always in strict elevations.

3. Land Use Pattern

The recorded forest area (including dry deciduous scrub) of Shiwalik hills of Punjab state is 1599.42 km2, which constitutes 16.93% total geographic area. The major forest types are dry deciduous, moist deciduous, dry deciduous scrub and coniferous forest. The non-forest area such as agriculture, grassland, water body, canals, settlements, riverbed and barren land contributes 7849.55 km2 which constitutes 83.07% of the total geographical area.

3.1 Vegetation type

(a) Northern dry mixed deciduous forest (5B/C2)

In most localities the tree canopy has been seriously broken by human activities, results in scattered tree and small shrubs. This forest is dominated by Acacia catechu, Anogeissus latifolia with the association of Lannea coromandalica, Aegle marmelos, Ehretia laevis, Mallotus philippensis, Nyctanthus arbor-tritis, Dendrocalamus strictus etc.

(b) Dry bamboo brakes (5E9)

Only one species Dendrocalamus strictus occurs and forms relatively low brakes with a sprinkling of the tree and shrubs of dry deciduous forest such as Anogeissus latifolia, Lannea coromandelica etc.

(c) Dry deciduous scrub (DS1)

A low broken soil cover of shrubby growth 3 to 6 m high including some tree species reduced to similar conditions, usually many stemmed from the base. The community is dominated by Woodfordia fruiticosa, Carrissa opaca, and Nyctanthus arbor-tristis with the some association Dodonaea viscosa, Aegle mormelos, Cassia fistula, and Acacia catechu.

(d) Khair-Sissu forests (5/1S2)

Dalbergia sissoo predominates in this association. The canopy is open associated with Acacia catechu. The older woods have more or less definite under storey, which is mainly composed of young species and few species of Tamarix dioca, Acacia fernesiana, Cannabis sativa and grasses like Saccharum spontaneum, Erianthus munja etc.

(e) Shiwalik chirpine forest (9/C1)

The pine stands singly or in groups with a scattered with lower deciduous tree story on the ridge and side slopes. There is usually a fairly continuous growth of xerophytic shrubs occurs as under-storey vegetation. Pinus roxburghii is associated with Mallotus philippensis, Pyrus pashia, Syzygium cuminii, Albizzia chinensis, Acacia catechu, and Terminalia chebula.

(f) Subtropical Euphorbia scrub (9/C1/DS2)

Euphorbia royleana forms consociations sometimes of considerable extent. Their distribution is related to edaphic factors, notably dry rocky ridges, where biotic pressure has been high.

4. Approach

Biodiversity characterization satellite remotely sensed data is being used for deriving vegetation cover type map. The vegetation type thus derived represents the habitats and their surroundings. These patches of the different forest types occur randomly as per the existing environmental conditions. Using landscape ecological principals these can be analyzed and quantified. Some of these parameters are fragmentation, porosity, juxtapositions, interspersion etc. and indices are derived to show their characteristics. Proximity of forests to road and villages and their impact is established. All these layers are overlaid to obtain disturbance index. Disturbance index image is the important for characterizing and identifying least or no disturbed areas (Fig. 1).

Ground observations are taken through stratified random sampling in all the forest types. Their economic uses have been found from literature. Total Importance Value (TIV) of each plant is established for its value for food, fuel wood, charcoal, timber, medicine etc. Ecosystem uniqueness is established from the list of the species found during survey based on its representativeness, species

Fig 1: Approach for Biodiversity Characterization at Landscape Level Using

Remote Sensing and GIS (after Roy et al., 1998)

endemism etc. Biological richness is obtained after integrating species richness, biological value, disturbance index, ecosystem uniqueness, terrain complexity.

5. Materials and Methodology

5.1 Materials

Following satellite remote sensing data have been used for land cover and land use classification of the area (Table 1).

Table 1: Details of satellite data used

Satellite ID

Path

Row

Date

IRS-1D LISS III

12 Oct. 2000

IRS-1D LISS III

03 Nov. 2000

IRS-1D LISS III

03 Nov. 2000

IRS-1D LISS III

31 Oct. 2000

IRS-1D LISS III

31 Oct. 2000

5.2 Ancillary data

Survey of India topo sheets on 1:50,000 have been used. Relevant literature on flora has been consulted.

5.2 Vegetation Cover type mapping

5.3.1 Preprocessing of satellite data

IRS-1D LISS-III data (Oct, 2000 and March, 2001) were used to prepare vegetation cover type map. A total five scenes were loaded and each scene was rectified with respect to 1:50,000 scale SOI toposheet (Total 32 toposheets were used, geometrically corrected and mosaicked to a single image). A second order transformation was followed. Average root mean square error within one pixel was maintained while preparing transformation model. Lambert Conformal Conic projection (LCC) was used during rectification of image (Fig. 2). Each rectified

Fig. 2 FCC

scene was subjected to radiometric correction before mosaicing it to a single mosaic image. After extraction of required area from this single mosaic image, it was subjected to Supervised Maximum likelihood classification using the ground truth information collected during the fieldwork.

5.3.2 Ground truthing

Reconnaissance survey was carried out in the area to acquire the knowledge of the vegetation and other broad land uses. During reconnaissance information on the correlation of image elements with that of ground features was also obtained. Interpretation key was formulated and classification was performed.

5.3.3 Vegetation Classification

Based on the a priori knowledge supervised classification method was followed. Training sites were selected and processed and the features with high classification accuracy were extracted. Using binary image the remaining area was extracted and unsupervised classification method was performed. The classified output was finally subjected hybrid classification approach to prepare vegetation type map of the study area showing various vegetation types. The accuracy of the image was evaluated using field knowledge and the ground truth information. Along with the different types of forest, other general lands use / land cover classes were also classified so as to understand the landscape of the region. Following forest classes have been delineated and a comparison with Champion and Seth’s (1968) classifications scheme is given below (Table 2).

List of the land cover and land use classes identified on the satellite data.

Moist deciduous forest

Dry deciduous forest

Dry deciduous scrub (Lantana)

Coniferous forest

Grass land

Plantation/Avenue trees

Agriculture

Water body

Settlement

Riverbed/Barren land

Table 2:Vegetation classes compared With Champion & Seth’s Classification:

Satellite based classification of vegetation types

Champion and Seth Classification (1968)

Moist deciduous forest

Dry bamboo brakes

Khair-sissu forest

Dry deciduous forest

Northern dry mixed deciduous forest

Dry deciduous forest

Dry deciduous scrub

Subtropical Euphorbiascrub

Dry deciduous scrub

Pine

Shiwalik chirpine forest

5.3 Field data

One of the most important components of biodiversity characterization is the information on plants or species richness. Information on plant species is further processed for evaluating other parameters like Total Importance Value, Ecosystem Uniqueness, Biodiversity value etc. There it is important that well distributed enough sample are taken for information on species occurrence. Classified vegetation cover type has been used for finding the sample size. Sampling intensity of 0.021 % has been done. Higher sample intensity is adopted (than recommended) in view of variability in the area.

5.4.1 Sampling design

Stratified Random sampling approach was followed and numbers of sample points were distributed to its probability proportional to its size. Field data was collected from 74 sample points of 20×20 m size during October 2000 (Table 3). The sample plots of 20x20m were used for tree species and nested approach and nested approach has been followed for laying sample plots of 10x10m for shrubs and 1x1m plot (five plots) for herbaceous layer.

The data was collected on following parameters in each of the sample plot.

Description of ecosystem and forest type including phenology

Species name and number of individuals for every species

Girth at breast height in centimeters

Economic importance such as grazing, medicinal, human food, fuel, timber, charcoal and other uses such as industrial use, rope making, tanning leather etc. This importance value was collected by interviewing local people interviewing wherever possible and from literature.

The data was analyzed for deriving various indices indicating biodiversity value, species richness and ecosystem uniqueness for different forest types.

Table 3: The distribution of sample points in each of the forest types

found in all the Shiwalik Hills of Punjab state

Forest Type

No. of sample plots

Moist deciduous

Dry deciduous

Deciduous scrub

Pine forest

Total

74

Database of all the species collected was created in MS Excel for further processing, details have been provided in the section on phytosociological analysis.

6. Database Creation in GIS

In geo-spatial analysis integration of spatial and non-spatial data or vector data is important. As discussed earlier the biotic disturbance had played a very significant role in existing biodiversity of toady. Database in GIS domain provides opportunity to analyze their impact zones. Therefore, the following maps were digitized from ancillary sources for their integration:

Village locations point and polygon features)

Road and railway networks

Contour lines with 20m intervals

The road and settlement maps were used as input in further analysis for deriving disturbance gradient. Contour map was used to prepare digital terrain model using ERDAS IMAGINE 8.4 software.

7. Landscape Analysis

The approaches for biodiversity characterization discussed in the literature contains several broad categories such as genetically based approach, species based approach, ecosystem based approach, and integrative approach. The approach adopted in present study is integrative method that includes significance of ecological, social, and cultural factors to the biodiversity in addition to biological factors. The approach used for the study focuses on following aspects:

Rapid assessment for monitoring of biodiversity loss and/or gain

Mapping of biological richness to understand its spatial nature, that helps in planning and execution

The customized package “Bio_CAP” is developed at Indian Institute of Remote Sensing, Dehradun to carry out multi-criteria spatial analysis. The satellite data provide key input i.e. vegetation type, which is used for deriving several landscape indices (fragmentation, porosity, patchiness, interspersion, and juxtaposition) depicting status of forest ecosystem. Ancillary database on roads and settlements has been used to prepare proximity buffer map and contour map is used to prepare terrain complexity map. All these parameters were integrated together with the field data on species richness, ecosystem uniqueness and biodiversity value. The ultimate result is the map that depicts areas categorized as per disturbance index.

7.1 Landscape analysis using Bio_CAP

The field data, vegetation type map and ancillary GIS data (roads, village locations, and contour) were analyzed using Bio_CAP (Biodiversity Characterization Programme) a customized package to prepare various indices depicting landscape characteristics.

Fragmentation has been the major cause of biodiversity loss and has been measured as a number of forest and non-forest patches in per unit area.

Patchiness is a measure of the density of patches of all forest types or number of clusters in a given mask or area.

Porosity is a measure of number of patches or density of patches within a particular type vegetation, normally primary vegetation type(s), regardless of patch size.

Interspersion is a count of dissimilar neighbor pixels (feature) with respect to central pixel (feature) of a particular grid or measurement of the spatial intermixing of the vegetation types.

Juxtaposition is a measure of proximity and adjacency of two or more vegetation types. Higher weightage is given to the classes who share or are likely to share more boundary.

Human influence zone is dependent on socio-economic set-up of the area/region and proximity the forest resources and can vary from 0.5 km to 5 km or even more. Proximity buffer or zone of influence from the roads and human settlements has been prepared.

7.2 Disturbance Index:

Disturbance has direct impact on the occurrence of plants and animals or biodiversity. Disturbance regimes provide an insight into the impacted area and its degree of impact under various vegetation or other natural resources. Disturbance Index has been considered here as a function of fragmentation, porosity, patchiness, interspersion, juxtaposition and influence zones or distance for the source. The analysis has been performed in the customized GIS package called Bio-CAP for this purpose.

7.3 Biological Richness

Biological forms of any area reflect the environmental conditions supporting the growth and evolution. Recently emphasis has been to look at the ecosystems (micro-climatic variations) or landscape diversity and its utility for conservation rather than one species. Therefore, the biological richness here has been considered as a function of ecosystem uniqueness, biodiversity value, species richness, Terrain complexity (computed through DTM by determining variance in DTM values) and most importantly disturbance index. The details of these parameters have been discussed elsewhere (Roy et al., 1999).

8. Observation Highlights

8.1 Vegetation Classification

Hybrid approach has been followed to do the digital classification of the data set (Fig. 3). Table 4 summarizes the results of classification in the region. Forest types viz., moist deciduous, dry deciduous, pine and dry deciduous scrub together constitute about 1404.06 Km², which is about 14.65 per cent of the total geographical area of the region. Non-forest classes such as agriculture, plantation, riverbed, barren land, settlement, grassland and water body makes up about 85.35 per cent of the geographical area. Amongst the forests, dry deciduous forest has

Fig.3 classified map

wide distribution in the region from Chandigarh to Pathankot and covers an area of 775.85 Km². Dry deciduous scrub is the next dominant forest type of the region which is mainly constituted with Lantana scrub distributed throughout the region and then followed by moist deciduous forest found in Dhar, Pathankot, Dholba, Talwara, Nangal, Noorpur in a fragmented patches. And coniferous forest covers an area of 6.51 Km² and is localized on the higher ridges or side slopes in the northern part of the Pathankot district.

Table 4:Area under different Land cover / land use classes in the region

Land use / cover class

Area in km2

Moist deciduous forest

276.46

Dry deciduous forest

775.85

Dry deciduous scrub (Lantana)

345.24

Coniferous forest

6.51

Grass land

38.24

Plantation/Avenue trees

211.04

Agriculture

7443.22

Water body

78.94

Settlement

178.46

Riverbed/Barren land

238.19

Total

9592.15

9. Phyto-sociological analysis

Phytosociological analysis was carried out to understand the floristic and vegetation pattern in the region. For the phytosociological analysis the vegetation types were grouped into five major types. Based on the species area curves developed initially an optimized field plots size of 20×20 m was adopted uniformly for all the types. In each plot all the species names; height, girth, herbs and shrubs were recorded. The phytosociological database was created and computed the basic structural parameters viz. frequency, basal area and density. Utilizing these parameters the importance value index (IVI) was calculated for all the types (Annexure-II).

Fig.4: Graphical presentation of Land use/land cover types and their

Distribution.

9.1 Species richness (Shannon-Weaver Index)

Species richness can be described as the number of the species in a sample or habitat per unit area. Higher the value greater the species richness. Species richness (Shannon Weaver Index) was measured using importance value index (Table 5, 6). Dry deciduous forest shows highest diversity (3.5361) with total number of 363 species followed by moist deciduous (3.0959) with moist deciduous forests then dry deciduous scrub (2.2666) having 77 species and coniferous forest shows least diversity (1.6207) with 58 species (Fig. 5).

Table 5: Biodiversity status in the Shiwalik hills of Punjab state

Forest Type

No. of

Families

No. of Species

Total no. of

Species

Total Importance Value

Trees

Shrubs

Herbs

Moist deciduous

161

10.46

Dry deciduous

141

147

363

10.53

Deciduous scrub

9.75

Pine

8.01

Table 6: Forest type wise Shannon Weaver Index in the Shiwalik hills of

Punjab state

Forest Type

Average Basal Area (m2)

Shannon Weaver

Index SWI (H’)

Moist deciduous

38.412

3.0959

Dry deciduous

22.948

3.5361

Deciduous scrub

8.133

2.2666

Pine

6.53

1.6207

9.2 Economically Important Species

Economically Important plants are the species, which have social and economic value. In the Shiwalik hills of Punjab state 240 economically important plants were recorded. The total importance value (TIV) for each species was calculated considering 10 important uses. The parameters considered are (1). Food (2). Fuel (3). Fodder (4). Fiber (5). Timber (6). Medicinal (7). Oil (8). Gums/Resins (9). Tannin and (10). Others. These 10 parameters have been weighted for a scale of 1 to 10 wherein maximum value represents highest economic value. Maximum total importance value was observed in dry deciduous type (TIV of 10.53) and coniferous type has the least value (TIV of 8.01) (Table 7). Some of the economically important species are Acacia catechu, Achyranthes aspera, Adhatoda zeylanica, Ageratum conyzoides, Azadiracta indica, Cannabis sativa, Moringa oleifera, etc. List of 232 economically important species and their uses were given in the Annexure II.

Fig 5: Species distribution in each habit type

Table 7: Total Importance Value of different vegetation types

Forest Type

TIV %

Moist deciduous

10.46

Dry deciduous

10.53

Deciduous scrub

9.75

Pine

8.01

9.3 Medicinal Plants

About 132 medicinally important species have been recorded during the sampling. Medicinal usage of these plants occurring in Shiwalik hills of Punjab have been compiled from available literature sources. Some of the important medicinaly important plants are Adhatoda zeylanica, Aegle marmelos, Azadirachta indica, Bacopa monnieri, Cordia dichotoma, Terminlaia chebula, Terminalia bellirica etc. and the list has bee furnished (Annexure.IV). Table 8 provide overview of the per cent utilizable species.

Table 8: Percent utilizable species for Total Importance value (TIV)

Forest type

USES

Food

Fuel

Fodder

Fiber

Timber

Medicinal

Oil

Gums/ Resin

Tannin

Others

11.15

0.79

9.09

9.8

10.2

31.66

5.02

3.5

5.18

9.83

12.7

2.36

6.94

7.91

2.6

32.1

0.65

0.86

1.3

8.45

10.05

2.21

9.06

9.31

9.55

31.61

6.37

3.18

6.61

8.09

8.52

20.78

2.35

Total

33.9

5.36

33.61

27.02

22.35

116.15

12.04

7.54

13.09

28.72

9.4 Ecological importance:

The species recorded during the field data collection were screened for their uniqueness with help of RED DATA BOOK (Anonymous, 1987, 1988, 1990) and many other references. All the species recorded were abundant in nature. However, two species were found to be rare, viz., Delphinium danudatum Wall. ex HK.f.Th and Peristylus constrictus (Lindl.) Lindl.

10. Results of Landscape analysis

10.1 Fragmentation

Increasing anthropogenic activities created discontinuity in the natural vegetation cover gets fragmented and the class becomes porous. Thus, fragmentation analysis of a land use class is an important landscape characteristic, which defines the status of that class in the present day context. Among all the characteristics of the landscape, fragmentation is more significant. Hence, the results of fragmentation are being discussed. For analysis of fragmentation in the forest, the vegetation map was reclassified as forest and non-forest classes, which resulted in a new spatial layer. A grid cell of 250 x 250m was derived which gets convoluted with the spatial data layer with a criteria of deriving number of forest patches within the grid cell. Output values were normalized between 0-10. For preparation of fragmentation map, the normalized values were rescaled between 0 and 1 to define four fragmentation classes. The fragmentation value up to 3 was given the probabilistic weightage of 0.2 and was considered as non-fragmented. Like wise a value up to 5 was assigned the weightage of 0.4 and was considered least fragmented; a class value up to 8 was assigned a weightage of 0.6 and more than 8 as 0.8 and was considered highly fragmented. Thus the out put map consisted of four fragmentation classes. Figure 6 shows the fragmentation status in the region.

Forest fragmentation is an a indicator of anthropogenic activities showing the extent of degree of the isolation of patches from the contiguous area along with the habitat loss. Fragmentation map stratified over the vegetation types, all the forests show significant fragmentation. Dry deciduous scrub showed relatively higher degree of fragmentation (40.20 per cent) fallowed by Dry deciduous forest (5.69 per cent). Moist deciduous forest showed least degree of fragmentation (Table 9 and Fig. 7).

Fig 6 Fragmentation map

10.2 Disturbance index

Disturbance index (DI) or the biotic interference is due to human activities, interacting with the forest. It is assumed that the villages and transportation network act as catalyst as the point source of disturbance. Therefore proximity from transport network and settlements is considered as one factor for deriving the disturbance in the forest. The proximity buffer from transport network and settlements were generated. Three proximity buffers were generated from the roads and villages, each of 500 meters, using the BIO-CAP package. Other landscape indices like fragmentation, porosity, interspersion, and juxtaposition were also estimated using the customized software package.

Porosity is calculated to estimate the number of patches in a forest type. The out put statistics were sampled to a scale of 1 – 10. In order to introduce uniformity in the map output, assigning probabilistic weightages to generate four classes of porosity rescaled it. The porosity level up to 1 was given a lower value of 0.3: a porosity value up to 4 as 0.5; porosity level up to 7 as 0.7 and the porosity level up to 10 was assigned a value of 0.9.

Interspersion, which is a measure of intermixing of the vegetation type, is calculated by convoluting 3 x 3 window on vegetation type. The number of dissimilar pixels is counted. It represents landscape diversity and dispersal ability of the species. Higher interspersion indicated less dispersal ability of the vegetation type represented by the central pixel. Here again a weightage of 0.1 was assigned to interspersion level up to 2; 0.5 up to level 3; 0.7 up to 5 and 0.9 to interspersion level 6.

Fig 7 Disturbance map

Juxtaposition measures the proximity of habitats and relative importance of adjacency. It is taken to determine proportion of edge shared by central path with the adjacent patches. Relative importance of adjacency is taken care by assigning weightages to possible associations. The common rules to assign weightages are (1) natural forest adjacent to natural forest – Highest weightage;(2) Natural forest adjacent to agriculture and other man made features-Lowest weightage and (3) natural forest adjacent to disturbed forest-Medium weightage. All the above indices were integrated to prepare the map depicting the disturbance prevailing in the region (Figure 8).

Figure 8 shows the levels of disturbance in the forest area of the region. The disturbance levels were grouped in four classes. Class1 defines low disturbance, class 2 medium disturbances, class 3 high disturbance and class 4 as very high disturbance region. Deciduous scrub showed higher degree of disturbance (50.77 per cent and 16.79 per cent in high and very high categories respectively).Pine forest showed least disturbance (11.04 per cent and 8.71 per cent in high and very high categories respectively) followed by moist deciduous forest (17.98 per cent and 14.27 per cent in high and very high categories respectively (Table 10 and Fig. 9).

10.3 Biological richness

Biological richness (BR) map generated for the Shiwalik hills of Punjab state, is a function of disturbance index, terrain complexity, species richness, biological value and ecosystem uniqueness (Fig. 10). The four classes defined for the purpose are class 1 (Low); class 2 (Medium); class 3 (High) and class 4 (Very high). In order to generate the biological richness map, the above inputs were assigned probabilistic weightages. In the context of landscape characterization, forests of Shiwalik hills of Punjab state are moderately rich in some fragmented packets. Deciduous forest showed high degree of richness (55.09 per cent and 12.86 per cent in high and very high categories respectively) followed by moist deciduous forest (17. 92 in high and 16.19 in very high categories. Deciduous scrub shows least richness (13.96 per cent in high and 1.61 per cent in very high categories) as compared to pine forest (Table 11 and Fig. 11).

11. Discussion and Conclusion

Biodiversity of Punjab Shiwaliks is facing threat due to various developmental activities. Satellite remote sensing data provide much needed qualitative information on the land cover and land use pattern of the region. Geospatial information is crucial for their management for long-term benefits. Efforts directed towards conservation of one or a few species have limitations and therefore it is essential to consider biodiversity conservation in its entirety i.e. at landscape level. Disturbance is one of the major factors for biodiversity loss, and the forest patches represent biodiversity depending on the existing environmental conditions and associations. Landscape principles like fragmentation, porosity, interspersion, juxtaposition etc. allow us to establish the disturbance gradient. The analysis done through remote sensing and GIS and phytogeographical analysis revealed that majority of the area is highly disturbed and therefore it is matter of great concern for biodiversity conservation. Shiwalik hills are faced with increasing anthropogenic activities and forest landscape are invaded by humans in the name of development. Large areas are under scrub and need protection to avoid further site degradation. The region has a large number of medicinal plants. Biorich areas are restricted to certain areas and fragmented pockets. The Moist deciduous forests were found to appear with highest biological richness then followed by dry deciduous forest. A lantana scrub invaded the porous landscape. These biodiversity ‘refugia’ in the unique and fragile ecosystem of Shiwaliks need to be protected on priority for future prosperity.

Fig 8 BR map

Table 9:Fragmentation status of the vegetation of Shiwalik Hills of Punjab state

Vegetation type

Low

Medium

High

Very high

Moist deciduous

226.74

(81.93)

41.74

(15.08)

7.58

(2.74)

0.68

(0.25)

Dry deciduous

604.11

(77.82)

125.14

(16.12)

43.73

(5.63)

3.36

(0.43)

Deciduous scrub

153.88

(44.51)

37.31

(10.79)

15.54

(4.49)

139

(40.20)

Pine

4.98

(76.15)

1.29

(19.72)

0.23

(0.00)

0.04

(0.61)

Numbers in parentheses denote values in percentage

Fig 9: Forest type wise Fragmentation status of Vegetation of Shiwalik Hills of

Punjab state

Table 10:Degree of disturbance of the vegetation

Vegetation type

Low

Medium

High

Very high

Moist deciduous forest

40.88

(14.78)

146.49

(52.97)

49.73

(17.98)

39.46

(14.27)

Dry deciduous forest

57.58

(7.43)

193.93

(25.01)

393.63

(50.77)

130.15

(16.79)

Dry deciduous scrub

16.91

(4.85)

39.13

(11.23)

163.91

(47.04)

128.47

(36.87)

Pine

0.91

(14.15)

4.25

(66.01)

0.71

(11.04)

0.56

(8.71)

* Numbers in parentheses denote values in percentage

Fig. 10: Degree of disturbance of the various vegetation cover types

Table 11: Status of Biological richness

Vegetation type

Low

Medium

High

Very high

Moist deciduous forest

46.54

(16.81)

42.19

(15.24)

152.53

(55.09)

35.59

(12.86)

Dry deciduous forest

77.34

(9.97)

434.41

(56.00)

139.02

(17.92)

124.95

(16.11)

Dry deciduous scrub

235.17

(68.20)

55.95

(16.23)

48.13

(13.96)

5.56

(1.61)

Pine

3.25

(52.76)

1.6

(25.97)

1.03

(16.72)

0.28

(4.55)

* Numbers in parentheses denote values in percentage

Fig. 11 Forest type wise status of Biological richness

Order Now