Momentum balance of the near-surface flow over orography

During two field campaigns on a hill on the Isle of Arran, Scotland, all significant terms in the turbulent, stratified Bernoulli equation were measured at near-surface sites across a hill. This has enabled the two causes of acceleration at the hill crest or in the lee of the hill, and the deceleration upwind and in the lee of the hill, to be quantified. As a consequence, a momentum budget for near-surface streamlines has been obtained. The causes of flow speed variation are (a) pressure gradients (measured using microbarographs) and (b) vertical turbulent momentum transport (measured using sonic anemometers and three-axis propellor anemometers).

The observations were made on the hill Tighvein (height 458 m) on the Isle of Arran in SW Scotland. Eight microbarographs with wind, temperature and humidity recorded as 10 minute averages were deployed along an east-west transect across the hill. The total length of the transect was about 6 km. At four sites turbulent stresses were measured at 8 m and 15 m using propeller anemometers and 20 Hz sampling. The experiments were run for 5 weeks during October and November 1997 and 4 weeks during and March 1998. Additional data were provided using two-daily radiosondes launched from west of the hill and from a light aircraft (UMIST).

Quantitative analysis of the pressure gradient, turbulent momentum flux and near-surface wind reveals, as expected, essentially two types of flow regime: a regime with neglibible stratification in which the maximum speed-up is at the hill crest and a stratified (but still high Froude number) regime where the maximum speed-up transfers to the lee of the crest. For the part of the hill above half-height, the following conclusions can be drawn:

1. For the neutrally stratified case: (a) The flow accelerates approaching the hill crest and both the pressure gradient and the turbulent momentum flux divergence cause this. Generally, the pressure gradient force is the larger contribution. (b) The flow decelerates in the lee. The pressure gradient causes this whilst the turbulent momentum flux divergence still acts to accelerate the flow (but is generally smaller in magnitude than the pressure gradient force).

2. For the stably stratified case: (a) The flow accelerates approaching the hill crest and both the pressure gradient and the turbulent momentum flux divergence cause this. The pressure gradient force always dominates. (b) There is further acceleration in the immediate lee of the hill caused almost entirely by the pressure gradient. The turbulent momentum flux divergence is negligible.

A similar approach has been used to investigate Foehn flow through the Brenner pass and down the Wipp valley during the GAP project within the Mesoscale Alpine Programme. Preliminary analysis of the GAP data will be presented.