| AFFILIATIONS | - Porter School of the Environment and Earth Sciences, Tel Aviv University, Tel Aviv, Israel
- Department of Geography and the Human Environment, Porter School of the Environment and Earth Sciences, Tel Aviv University, Tel Aviv, Israel
- Department of Natural Sciences, The Open University of Israel, Ra'anana, Israel
|
| ABSTRACT | The rainfall in the Mediterranean-climate region of Israel is confined almost to the period between October and April, 2/3 in the mid-winter months. The analysis is based on daily rainfall in 126 stations of the Israel Meteorological Service (representing the Mediterranean-climate region of Israel) for the period 1951-2014. A 'daily rainfall' average is calculated over the entire set of stations and the seasonal time-series is smoothed using a 27-days window, which is found optimal. In order to expose the modality of the individual seasons, each season is approximated by a series of Gaussians. Each Gaussian corresponds to one mode, characterized by its timing, amplitude and duration. The compatibility of the modeled season with the observed (smoothed) one is assessed by the correlation between them, the standard error and the variance, attempting to keep the number of Gaussians to minimum. The correlation between the modeled and the observed rainfall is above 0.9. Up to five modes are usually found, not confined to the mid-winter months. Significant modes were found for example in October (2000) and in April (1971). Moreover, a minimum in the middle of the winter was found in several seasons. An average course of 30 seasons, chosen randomly, as well as the average course performed for the entire set (62 seasons) indicate a minimum in the mid-season, with one peak in late December and another in the beginning of February. We hypothesize three factors that may explain this minimum: (1) the high contrast between the warm water of the Eastern Mediterranean and the overlying cold air masses responsible for the rains in Israel, reaching the maximum around December; (2) the maximum in the intensity of the Cyprus lows, which is expected to be in January-February, corresponding to the maximum of the polar-subtropical temperature gradient. The combination of the two factors above leaves the center of January with a secondary minimum. Another hypothesis (3) is that the Cyprus Lows' activity does not have a distinct peak, but their peak activity is spread over the entire mid-winter (DJF), so that occasional minima in the high season in various winters causes a secondary minimum to appear in the long-term mean. To examine the hypotheses, especially #2 and #3, the daily occurrence and intensity of the Cyprus lows will be analyzed. |