MM5 simulations of diurnal winds and moisture transport in the Mt. Everest area of the Nepal Himalayas: Some initial findings

As part of an ongoing study of the relationship between daytime valley winds, moisture transport and thunderstorm initiation at elevations above 4000m asl in the Mt. Everest area of the Nepal Himalayas, a first run of the NCAR/ PSU MM5 Mesoscale model was performed, yielding 60 hours of ½ hourly simulations for the period 12UTC 9 May to 00UTC 12 May 1996 and including two full diurnal cycles of valley winds. Using triple- nested interactive domains with 27, 9 and 3km resolution, default parameters, such as land use, were left in tact. The computer-smoothed topography left the N-S oriented Dudh Kosi- Tughla Khola- Imja Khola valley system with a 7000+ meters ridge surrounding the valley on all sides, except at its southern entrance. Elevation rise from south to north is 4400m; the valley side walls on the southwest and entire east section are extremely steep.

A combined City College- Tribhuvan University team of researchers made weather observations and measurements during this period at the Hotel Everest View, 3800m asl, and at the Italian weather research station Pyramid, 5050m asl. The Nepal DHM provided additional weather data from their 4355m asl station at Dingboche.

A strong valley wind and weaker mountain wind were simulated. For example, a south-to-north cross-section along the model's valley axis from 3000 to 7000m indicated vigorous daytime valley flow at speeds of 10 to 15 m/sec from the south along the entire axis, up to just above 5000m, with duration several hours longer at the lower elevations. Nocturnal wind speeds appeared to be affected by elevation; below 5000m daytime and nighttime wind speeds were identical, above 5000m daytime wind speeds were twice that of the nocturnal flows. Depth of the nighttime flows varied from 100 to 200m at 3500m to ~ 25m at 5000m. At 6000m, northerly nighttime drainage flows appeared to be completely blocked by westerly winds. The duration of the transition from daytime to nighttime winds varied from less than an hour at 6000m to almost 6 hours at 4000m.

In spite of the low, 3km resolution, there was remarkable agreement and disagreement between our measured wind data at HEV and that of the model. For example, daytime wind speeds, 10 to 12 m/sec, and direction, from the south, were identical. Nocturnal wind speeds of up to 10 and 12 m/sec were almost five times faster than the observed 3 m/sec.

Temperature and moisture comparisons between the MM5 results and, for example, those from the Pyramid did not fare so well. MM5 surface temperatures were consistently 4C colder, did not exhibit the expected nighttime cooling and missed the typical, steep rise, ~ 8C of the observed surface temperature between 0700 and 0800hrs. Except for a 6 hour period during the day, the model consistently underestimated moisture at the surface.

Severe thunderstorms occurred in the entire northeast and east areas of Nepal on May 10 and 11, as well as in the Everest area. The model predicted the observed thunderstorms but not at the same times and locations.

These findings indicate a potential that the MM5, perhaps with improved topography and land-use parameterizations, could become a useful tool for predicting severe pre-monsoon weather in the region of Mt. Everest.