To divide the Himalaya into natural regions

The sites referred to are given in Fig. 7.6.
To divide the Himalaya into natural regions is itself difficult because the ecology is complex, as Schweinfurth (1984) has discussed.
However, Karan and Iijima (1985) have suggested a tripartite subdivision of the region which is depicted in Fig. 7.7.
Each zone is currently experiencing a range of pressures which are listed in Table 7.4.
As Ives (1987) has pointed out and the issue of environmental degradation in the Himalaya is extremely complex and has been extensively debated during the 1980s, giving rise to what he describes as the Himalaya Environmental Degradation Theory (HEDT).
This invokes the operation of vicious circles, involving deforestation for arable cultivation   and fuelwood underpinned by rapidly increasing population pressure; as land is degraded by accelerated erosion, even steeper slopes are cultivated and more forest is cleared.
This view is espoused by Myers (1986), who believes that deforestation in the Himalaya has been responsible for flood disasters in the lower reaches of the Ganges and Brahmaputra and the higher incidence of flooding that has occurred since 1940; apparently flooding now affects some 10 6 ha of land as compared with 6 × 10 6 ha in the early 1950s.
Such a theory is, according to Ives (1987) and too simplistic not least because much of the data on deforestation and fuelwood consumption on which it is based are unsound.
For example, M Thompson et al.(1986) have highlighted the fact that estimates of the latter vary by a factor of 67!
That such discrepancies arise is due to compilation of data at a local scale and its subsequent incautious use to derive estimates at a regional scale.
The comments made in Fig. 7.7, concerning land-use pressures in the Himalayan   zones and thus represent a generalised overview which may be grossly inaccurate at the local scale.
Similarly, management and development plans may not be generally appropriate and should be tailored to alleviate local pressures within an integrated regional framework.
Deforestation still remains a crucial issue in the Himalaya but the temporal and spatial pattern of its occurrence is, in the light of more recent research, very different from that hitherto envisaged.
For example, Myers (1986) states that between 1947 and 1980 Nepal's forest cover declined from 57 per cent to 23 per cent , but Mahat et al.'s (1986a, b, 1987a, b) work in the Middle Hills of Nepal (Sindhu Palchok and Kabhri Palanchok districts) suggests a different spatial and temporal pattern.
Their data indicate that serious deforestation was occurring more than 200 years ago and was most severe between 1890 and 1930.
More recently and reforestation has occurred naturally on abandoned farm terraces where grazing pressures have been low.
Mahat et al.also show that the area of forested land has not changed very much since 1950, although forest quality has declined as grazing pressure has increased.
Using records of oral history from these two districts, Gilmour (1988) has shown that tree cover has actually increased on private farmland over the last 20 years and while there is still an overall loss of forest it is not as acute as the HEDT suggests.
Similarly, Byers's (1987) work in the Mount Everest National Park indicates that there has been less forest removal than envisaged and that shrubs/grassland and forested slopes below 4000 m OD are stable.
Nevertheless and soil erosion is an appreciable problem in some Alpine pastures of the park due to overgrazing in a transhumance system, fuelwood collection and periglacial processes.
There are also variations in the temporal and spatial patterns of deforestation in the Indian Himalaya; in Uttar Pradesh, for example, Tejwani (1987a) reports that forest losses have been extensive and that much of what remains is degraded.
In the Dehra Dun Valley some 16 per cent of the forest cover has disappeared during the last 100 years (J F Richards 1987).
Moreover, Tucker (1987) has pointed out that the first period of large-scale deforestation in northern India occurred in the 1850s and 1860s as the British colonisation of India gained momentum and railways were constructed to gain access to the upper Ganges and Indus plains.
Not only is the extent of deforestation variable but the reasons why it is occurring also vary from region to region.
Mahat et al.(1986a, b, 1987a, b) believe that in the Middle Mountain region of Nepal the loss of forest is due mainly to the increased requirements for arable land rather than to the need for fuelwood.
As Ives (1987) comments, findings such as these have important implications for the formulation of mitigating plans; the introduction of more efficient wood-burning stoves, for example, while reducing the need for fuelwood does not provide a substitute for arable land.
In fact, Hrabovszky and Miyan (1987) advocate land-use intensification rather than clearance of new land as the most effective way to increase food production in the Nepal Himalaya.
This and they argue, could be achieved, at least partially, by increasing the area of irrigated land especially in the Terai plains of southern Nepal.