The remote sensing study provides the first sea breeze cloud frequency composites over the Iberian Mediterranean area and the isle of Mallorca (an area bounded between 35º32’N and 42º48’N and 5º56’W and 4º42’E and shown in Fig. 1), both in Spain. The aim of this study is to offer quantitative and high quality composites of frequently occurring sea breeze convergence zones in relation to the regional scale atmospheric circulation, the shape of the coastline and the small-scale geographic features. Sea breeze fronts are common phenomena from May to October in the Iberian Mediterranean area and the isle of Mallorca. During the summer dry season, this is typically the only source of precipitation in this area, bringing an average of 100-125 mm yearly to inland areas. Severe and moderate thunderstorms events can occur under sea breeze situations, even though the forecast is for mostly clear skies. Even though this study looks at one period of data, it captures representative cloud frequency patterns for different background wind regimes applicable on regional and local scales. Information from this one time period can be used to tailor further long-term studies that utilize cloud frequencies.
FIG. 1. Map of the study area showing locations named in the text. The white circle represents the rawinsonde upper air sounding location at Murcia.
Daytime Advanced Very High Resolution Radiometer (AVHRR) data from the U.S. National Oceanic and Atmospheric Administration (NOAA) polar orbiting satellites were used for deriving sea breeze cloud frequency composites. AVHRR full horizontal resolution data (1.1 km) from NOAA-17 morning (900 – 1200 UTC) and NOAA-16 afternoon (1200 – 1500 UTC) orbits were collected from the High Resolution Picture Transmission (HRPT) receiving ground station placed at National Institute for Aerospace Technology (INTA, Mas Palomas, Canary Isles, Spain; http://www.inta.es/index.asp), for the warm 6-month period May-October 2004
The cloud frequency composites presented here only looked at pure convective development associated with the sea breeze. A set of three mixed objective-manual criteria were applied to detect true sea breeze boundaries and reject active synoptic-scale or mesoscale disturbances. The first criteria focused on selection of favourable synoptic environments using Jenkinson and Collison weather type classification: i.e., anticyclonic circulation, thermal lows over the Iberian Peninsula or weak surface pressure gradients. The second criteria of temperatures =20 C in the planetary boundary layer ensured enough surface heating to provide convection. Under the third criteria, large-scale flows (mean layer vector wind between 1000 to 700 hPa) <13 m s-1 allowed for the sea breeze to develop. Those days which met criteria 1 to 3 were visually screened. Images which showed a line of larger Cu and Cb clouds at the leading edge of the sea breeze on a cloudless sky were selected. Convection embedded in multilevel cloud cover was rejected: i.e. the manual selection visually screened out high, medium and low-clouds cases where we could not detect a characteristic line of Cu and Cb clouds associated with the sea breeze front development. A set of 69 and 76 images from NOAA-17 and NOAA-16 respectively were used.
The daytime over land algorithm which was used to derive sea breeze cloud composites consists of four spectral tests applied to each pixel. The constant thresholds have been successfully tested to be functional during the 6-month study period May-October. The algorithm discretizes all AVHRR data into three groups: cloud-free, cloudy and snow-ice. The final cloud determination is obtained by subtracting snow-ice pixels from cloudy ones.
The high-resolution composites aided in identifying the location of five preferential Sea Breeze Convergence Zones (SBCZ) in relation to the shape of coastline and orographic effects. We found five preferential areas along the sea breeze front with higher convective activity as shown by increased cloud frequency over surrounding areas. The regions of increased cloud frequencies (with >30% or more maximum cloud frequency) are referred to as convective hot spots and are shown in Figs. 2a and b and are as follows:
The SBCZ1 corresponds to the eastern Pyrenees and Prepyrenees ranges (Vall de Núria and Ripollés mountains) in northeastern IP with afternoon maximum cloud frequencies ranging from 50.0% to 60.0%.
The SBCZ2 includes the eastern region of the Iberian system mountains and has afternoon maximum cloud frequencies above 35.0% for the Maestrazgo mountain range and 30.0% for the Puertos de Beceite mountain range. Unlike the general minima of rainfall in summertime over the Mediterranean basin, the SBCZ1 and SBCZ2 have local maxima of precipitation at this season.
The SBCZ3 is a small region which contains the Prebetic mountain ranges (1000-1600 meters) in Alicante. It has afternoon maximum cloud frequency amounts above 30.0%.
The SBCZ4 is collocated with the Betic system mountains (Nevada, Filabres and Baza mountain ranges; 2000-3500 meters) with afternoon maximum cloud frequency values above 35.0%.
The SBCZ5 corresponds to the isle of Mallorca in the Balearic archipelago. Sea breeze flows from the SW (Bay of Palma) and NE (Bay of Alcúdia) coastlines produce a convective area with afternoon maximum cloud frequency amounts around 30.0% near the center of the island.
FIG. 2. Sea breeze cloud frequency composites for (a) morning (left, NOAA-17) and (b) afternoon (right, NOAA-16) orbits during the six-month study period: May-October 2004. The number of images averaged is shown in the lower-left corner of each image (n = sample size). The thunderstorm symbols indicate location of the five prominent SBCZ. Note that color table ranges in frequency from 0 to 60% to highlight SBCZ.
Composites from May to October show a marked inter-monthly cycle on sea breeze cloudiness over the study area. May (Fig. 3a) is the cloudiest month. June (Fig. 3b) represents the next cloudiest month. The anticyclonic subsidence associated with the Azores high pressure system produces an overall low mean cloud frequency in July (Fig. 3c). A secondary peak of mean cloud frequency is found in August (Fig. 3d). A gradual decrease in frequency is found in September (Fig. 3e), whereas the sea breeze front is barely discernible in October (Fig. 3f) due to fewer hours of sunshine and consequently weak local circulations along the Mediterranean coast.
FIG. 3. As in Fig. 2, except for morning and afternoon composites during (a) May; (b) June; (c) July; (d) August; (e) September, and (f) October.
Regimes 1 to 4 (NE-E and SE-S; hereafter onshore flows) and 5 to 8 (SW-W and NW-N; hereafter offshore flows) were grouped in order to contrast the impact of wind direction (onshore vs. offshore) which theoretically express different sea breeze cloud patterns.
Under onshore synoptic flows (Figs. 4a and 4b) clouds are widespread and there is higher cloud cover on mountains away from the coast (e.g., Iberian system mountains, Lower plateau and Central mountain ranges). Onshore flows develop low cloud amounts in the morning (Fig. 4a), whereas sea breeze fronts and inland convection grow for the afternoon composites (Fig. 4b).
In contrast, under offshore background synoptic flows (Figs. 4c and 4d), sea breeze convergences are distinctly visible as thin lines of clouds and located parallel to and close to the eastern shore of the IP. offshore background flows develop high amounts of sea breeze cloudiness during the morning orbits (Fig. 4c) and, therefore, small differences between morning and afternoon (Fig. 4d) mean cloud amounts were detected.
FIG. 4. As in Fig. 2, except for morning and afternoon composites under (a, b) 1 to 4 (NE to S; onshore) and (c, d) 5 to 8 (SW to N; offshore) large-scale regimes summarized in Table 2.
It was found that light to moderate (=5.1 m s-1) winds aloft result in more clouds at the leading edge of sea breezes. In contrast, strong synoptic-scale (>5.1 m s-1) flows weaken boundary layer convergence.
FIG. 5. As in Fig. 2, except for morning and afternoon composites under (a, b) light to moderate (1 and 3 regimes) and (c, d) strong (2 and 4 regimes) onshore large-scale flows.
FIG. 6. As in Fig. 2, except for morning and afternoon composites under (a, b) light to moderate (5 and 7 regimes) and (c, d) strong (6 and 8 regimes) offshore large-scale flows.
The results from this satellite meteorology study could have practical applications for many including those that forecast the weather, those that use the forecast for making decisions related to energy use, fishing, recreation or agriculture activities as well as for estimating pollution or issuing warnings for heavy rain or flash flooding.