Hydrological ecosystem services of the Shivapuri National Park

The Shivapuri National Park to the north of Kathmandu is a protected area of 13,000 ha. WaterWorld version 2.91 was used to examine the current distribution of hydrological ecosystem services and the potential impacts of a scenario for land use change.

Baseline hydrology

Elevation in the park varies from 1300 to 3600 masl with precipitation from 1600-2400 and mean annual temperature from 12 to 18C. Wind driven rainfall in the park has a mean 2200 mm/yr whilst actual evapotranspiration has a mean of 310 mm/yr but varies significantly with slope, aspect and tree cover. Fog inputs are only around 200 mm/yr (8.4% of total precipitation) on average for the park. Water balance is thus 2000 mm/yr on average within the park, compared with 1600 for the wider region (meaning 100 km around the park as shown in Fig 1.) and 1500 for the Kathmandu metropolitan area.

Mean water resource stress (measured as the annual average % of demand not supplied) is low for much of the region but high for the parts of the Nepalese and Indian plains and of Kathmandu, see here. The Kathmandu Valley has significant cropland and this significantly increases the human footprint on water quality. Shivapuri provides significant hydrological ecosystem services in diluting poor quality agricultural runoff with high quality forest runoff, benfiting many in Kathmandu.

Land use change scenario

We used the WaterWorld land use change model to project recent deforestation rates for 40 years into the future for the entire tile, allowing deforestation at recent rates but only outside of recognised Important Bird Areas (IBAs), replacing with cropland. The scenario thus assumes that IBAs are sustained and protected through ecosystem service related fiunance This leads to a tile-wide change from mean 25% tree cover to mean 18% tree cover (see fig 1) with most deforestation in the areas that have lost most forest recently - the foothill forests of Nepal. Cropland increases from 12% to 26% over the tile.

Figure 1 Tree cover (%) for baseline (left) and land use scenario (right).

The scenario leads to a decreases in evapotranspiration of -67mm/yr on average in the areas in which forest has been lost, and decreases fog inputs by an average of 37 mm in the areas deforested, so increases the water balance by an average of 77 mm/yr in some 8.5 % of areas and decreases it by 22 mm/yr in the remaining 6% of areas. The areas with the greatest increases are in the lowlands and the greatest decreases are in the uplands. Runoff decreases in some areas (where the change in fog inputs is greater than the change in ET), see figure 2.

Figure 2 Areas in which runoff decreases (% of baseline runoff)

Clearly even rivers that are in neighbouring valleys can show opposite signs of change according to the slope, aspect, exposure and other charcateristics of the areas deforested (see fig. 3).

Figure 3 Areas in which runoff increases (% of baseline runoff).

Impact on mean water resource stress is no change over 96% of the tile and a decrease of -3.5% over the remaining 4% of the tile. The agricultural extension leads to an increase in the human footprint on water quality of 13% over the tile comprising a 46% increase over 27% of the area and no change over 73% of the area. Clearly the lack of agriculturalisation in Shivapuri means the rivers flowing into Kathmandu from the north are protected from significant change in water quality. This is not the case for rivers flowing from the south.

Figure 4 Change in human footprint on water quality (%)