Ls Land Issue 26 Hawaiian Breeze
Bonifacia Cramm
Local sea-breeze and land-wind regimes in Hawaii meet and interact with the prevailing trade wind giving rise to cloud lines of distinctive character. These cloud systems are sources of rain and are sufficiently frequent to be important influences in local microclimates. Four different types of interaction are described.
The types of interaction are primarily determined by the height and size of mountain barriers. High barriers may split the trade wind into lateral currents flowing around the mountain while low barriers allow the trade wind air to blow over the top of the mountain.
Measurements of local surface-pressure gradients and pilot-balloon data taken adjacent to the zones of interaction provide bases for a description of the vertical and horizontal circulations set up by the meeting of trade wind and sea breeze.
Sea- and land-breeze circulations are frequent phenomena over the Hawaiian Islands (Leopold 1949; Eber 1957; Lavoie 1967; Mendonca 1969; Garrett 1980; Schroeder 1981; Chen and Nash 1994). These thermally driven diurnal circulations may contribute to rainfall by reinforcing trade wind orographic lifting (Garrett 1980), generating areas of low-level convergence when interacting with the prevailing trade winds (Leopold 1949; Takahashi 1977; and others), or generating orographic lifting of upslope flow in areas not exposed to trade winds (Yang and Chen 2003a). The cloud and precipitation patterns developed by these diurnal circulations are important local weather phenomena, and can affect temperatures, humidity, rainfall, and the surface airflow (Chen and Wang 1994; Wang and Chen 1995; Carbone et al. 1995). Such local weather must be taken into account during the forecast process.
Weather forecasting over the northwestern part of the Big Island is also challenging due to large variations in local climate ranging from humid tropical conditions on the windward side of the Kohala Mountains to semiarid desert terrain on the lee side of the Waimea Saddle with large rainfall gradients within a short distance (Giambelluca et al. 1986). The vegetation cover changes from tropical forest on the windward side to sparsely vegetated grazing land (lava rocks) along the upslopes (lowlands) on the lee side. Notable spatial and temporal variations in local circulations and weather are observed in this area under different trade wind regimes (Schroeder 1981; Yang and Chen 2003a). Even though trade winds are persistent during the summer months, there are still large day-to-day variations in the summer trade wind weather over the Hawaiian Islands (Ramage and Schroeder 1999; Frye and Chen 2001). The global model output from the National Centers for Environmental Prediction (NCEP) is too coarse to resolve these local weather and circulations under different large-scale settings. Wang et al. (1998) applied the NCEP hydrostatic Regional Spectral Model (RSM; Juang and Kanamitsu 1994) with 10-km resolution to simulate the island weather. They noted that the 10-km RSM showed significant improvements over the NCEP global model in forecasting the local weather. However, a 10-km grid is still inadequate to resolve the complex island terrain and local circulations over the Hawaiian island chain. Large discrepancies were noted between the RSM simulations and observations in terms of rainfall distributions, rainfall rate, and low-level winds (Wang et al. 1998). High-resolution mesoscale models are needed to realistically represent the local terrain and surface conditions, to resolve local weather conditions such as the Waimea winds and the sea breezes, and to provide better numerical guidance for forecasters in the region.
The RSM and MSM utilize the two-layer soil model of Mahrt and Pan (1984) and Pan and Mahrt (1987) with modifications based on Pan (1990). There are 13 vegetation categories and 16 soil categories used in the LSM. Vegetation fraction ranges from 0% to 100%. The vegetation types and vegetation fraction for the MSM domains were compiled from the U.S. Geological Survey (USGS) 1:100, 000-scale land use land cover level II data for Hawaii (USGS 1986). The soil types were also compiled from the USGS data with reference to the soil surveys in Hawaii (Foote et al. 1972; Sato et al. 1973).
In our compiled datasets, the vegetation type near the coast over the northwestern part of the Big Island is bare soil with scattered broadleaf shrubs dominating uplands. The vegetation fraction is zero near the coast and rather small (
The boundary layer physics employs the Medium-Range Forecast (MRF) model scheme with nonlocal diffusion for the mixed layer developed by Hong and Pan (1996). This scheme is strongly coupled to the surface layer physics. In the scheme, the turbulent diffusivity coefficients are calculated from a prescribed profile shape as a function of boundary layer height and scale parameters derived from similarity requirements. Above the mixed layer, the local diffusion approach is applied to account for free atmospheric diffusion.
June 1978 typified early summer trade wind weather. Surface wind steadiness (defined as resultant speed/mean speed) exceeded 90% at first-order stations over the Hawaiian Islands (Lihue on the island of Kauai, Honolulu on the island of Oahu, and Kahului on the island of Maui) (Schroeder 1981).
At 0200 HST 25 June, as the TUTT cell drifted farther north, the subtropical high north of the Hawaiian Islands weakened (not shown). As a result, the trade wind speed continued to decrease. Weaker winds were blowing along the transect on 25 June (Fig. 9a) than on 24 June. The sea breezes extended upslope of Puu Nui where it merged with the anabatic flow of Mauna Kea (Fig. 9a). Showers developed along the slopes, within 5 km of Puu Nui.
The subtropical high located to the northeast of the Hawaiian Islands sustained moderate trade winds over the island chain on 26 June (Fig. 10). The trade winds along the transect for this day were slightly stronger (Fig. 11a) than 25 June (Fig. 9a). Longer lasting sea breezes appeared in midmorning and persisted through the afternoon (Fig. 11a). The sea-breeze front advanced beyond Lava Flow. Cumuli developed along the sea-breeze front but did not produce rain.
The simulated sea breezes persist between 1000 HST and 1800 HST and advance past Waikoloa Office (Fig. 11b), in good agreement with observations (Fig. 11a). The simulated surface air temperature distributions are also consistent with observations in that the 27.5C isotherm extends past Puu Nui between 1000 HST and 1500 HST.
The remnants of Hurricane Carlotta completely disrupted local circulations as she passed the Big Island on the night of 27 June (Fig. 13). Heavy rains fell on the islands of Kauai and Oahu with light showers over the Big Island. Onshore winds blew at Beach between 2200 HST 27 June and 0300 HST 28 June (Schroeder 1981). Weak southeasterly synoptic flow prevailed behind Carlotta. Since the transect was sheltered from the southeasterlies by Mauna Kea and Mauna Loa (see Fig. 2b), anabatic winds developed unhindered, reaching the Waimea airport by 0900 HST 28 June (Fig. 14a).
Observations from Schroeder (1981) showed that convective showers occurred within or adjacent to the transect on three of the six observation days. In two instances (25 and 28 June), weak showers developed from shallow clouds, which formed upslope in anabatic currents and drifted downslope as the anabatic winds subsided and the trade winds returned. On 26 June, cumuli developed along the sea-breeze front but did not produce rain.
Soil properties and insolation (i.e., cloudiness) regulate surface heating, which in turn regulates the characteristics of sea breezes (Schroeder 1981). More cloudiness would result in less surface heating, which in turn would inhibit further development of sea breezes. Several modeling studies (McCumber 1980; Garrett 1982; Yamada 1982) have shown that sharp horizontal changes in the character and type of the vegetation cover would induce mesoscale circulations. Observational studies by Segal et al. (1988) also showed noticeable air temperature differences due to the effect of vegetation cover under the same environmental conditions. There are no prior studies on the effects of lava rocks on sea-breeze behavior over the Hawaiian Islands using high-resolution models.
Our sensitivity tests (Fig. 17) show that by reducing surface albedo, thermal conductivity, and heat capacity at the same time by 90%, the model simulates a diurnal skin temperature range of about 30C at Lava Flow, close to the observed diurnal range. A similar reduction in any of surface albedo, thermal conductivity, and heat capacity or any two of them combined would produce a diurnal skin temperature range that is less than 30C (Fig. 17). It is important to note that the 90% reduction in soil parameters is rather arbitrary since the exact values for surface albedo, thermal conductivity, and heat capacity for lava rocks are unknown.
Earlier studies on sea breezes (e.g., Estoque 1962; Neumann and Mahrer 1971; Lambert 1974) indicated that greater horizontal resolution is necessary to achieve more accurate prediction. Case et al. (2002) performed verifications of high-resolution Regional Atmospheric Modeling System (RAMS) forecasts over east-central Florida during the 1999 and 2000 summer months. They noted that the skill scores of the RAMS sea-breeze predictions over central Florida improved when using 1.25-km grid spacing compared to 5-km grid spacing. Analysis of the individual sea-breeze cases in New England (Colby 2004) also showed that, with higher-resolution grids, the model was able to resolve realistic details in the sea-breeze circulation that were missed by coarser-resolution grids.
b37509886e