**3. Heat wave events and synoptic patterns**

A strong relationship exists between large scale circulation patterns and regional surface variables such as surface pressure, dynamical rainfall, wind and temperature (Tymvios et al., 2007, 2008, 2010a; Xoplaki et al., 2003). As a consequence, synoptic upper air charts at certain levels comprise a valuable tool for the operational weather forecaster to qualitatively predict occurrences of weather phenomena observed on the ground (e.g., heavy rainfall; see Tymvios et al., 2010a). The height pattern at 500 hPa is often used for this purpose. In order to take advantage of these semi-empirical methods and to simplify the statistical processing, stochastic downscaling methods are often applied to the actual weather patterns in order to generate clusters of synoptic cases with similar characteristics. Weather type classifications are simple, discrete characterizations of the atmospheric conditions and they are commonly used in atmospheric sciences. For a review of various classifications, including their applica‐ tions, refer to Key & Crane (1986), El-Kadi & Smithoson (1992), Hewitson & Crane (1996) and Cannon & Whitfield (2002).

Heat waves have a distinct impact on society through increased mortality, change in the en‐ ergy consumption profile and the diversification of social behavior. The severity of the heat events may include the local climatological characteristics, the community design and the individual tolerance to heat. Both the frequency of appearance and the intensity of heat waves are increasing in the Mediterranean area (Founda & Giannakopoulos, 2009).

The eastern Mediterranean climate is characterized by the succession of a single rainy sea‐ son (November to mid-March) and a single longer dry season (mid-March to October). This generalization is modified by the influence of maritime factors, yielding cooler summers and warmer winters in most of the coastal and low-lying areas. Visibility is generally very good. However, during spring and early summer, the atmosphere is quite hazy, with dust trans‐ ferred by the prevailing south-easterly to southwesterly winds from the Saharan and Arabi‐ an deserts, usually associated with the development of desert depressions (Michaelides et al., 1999). The influence of synoptic types on the urban heat island has been investigated by Mihalakakou et al. (2002) who have also adopted a neural network approach.

The definition for a heat wave recommended by the World Meteorological Organization is "when the daily maximum temperature of more than five consecutive days exceeds the maximum temperature normal by 5°C". Nevertheless, in most countries, the definition of ex‐ treme heat events is based on the potential for hot weather conditions to result in an unac‐ ceptable level of adverse health effects, including increased mortality. Also, a threshold in maximum temperature is in practical use in many countries.

These periods of abnormally and uncomfortably hot and (usually) humid weather are very common in the eastern Mediterranean during summer and early autumn. Expert examina‐ tion of the synoptic patterns on upper air charts may reveal the potential for a heat wave event. In this respect, the research presented here attempts to identify height patterns favor‐ able for heat events by using a neural network classification method, namely, Kohonen's Self Organizing Maps (see Kohonen, 1990).

#### **3.1. Data**

**Figure 3.** Urban areas of Cyprus shown in red

4 Remote Sensing of Environment: Integrated Approaches

and Cannon & Whitfield (2002).

**3. Heat wave events and synoptic patterns**

A strong relationship exists between large scale circulation patterns and regional surface variables such as surface pressure, dynamical rainfall, wind and temperature (Tymvios et al., 2007, 2008, 2010a; Xoplaki et al., 2003). As a consequence, synoptic upper air charts at certain levels comprise a valuable tool for the operational weather forecaster to qualitatively predict occurrences of weather phenomena observed on the ground (e.g., heavy rainfall; see Tymvios et al., 2010a). The height pattern at 500 hPa is often used for this purpose. In order to take advantage of these semi-empirical methods and to simplify the statistical processing, stochastic downscaling methods are often applied to the actual weather patterns in order to generate clusters of synoptic cases with similar characteristics. Weather type classifications are simple, discrete characterizations of the atmospheric conditions and they are commonly used in atmospheric sciences. For a review of various classifications, including their applica‐ tions, refer to Key & Crane (1986), El-Kadi & Smithoson (1992), Hewitson & Crane (1996)

Heat waves have a distinct impact on society through increased mortality, change in the en‐ ergy consumption profile and the diversification of social behavior. The severity of the heat events may include the local climatological characteristics, the community design and the individual tolerance to heat. Both the frequency of appearance and the intensity of heat

The eastern Mediterranean climate is characterized by the succession of a single rainy sea‐ son (November to mid-March) and a single longer dry season (mid-March to October). This

waves are increasing in the Mediterranean area (Founda & Giannakopoulos, 2009).

As an indication of a possible heat event, the maximum temperature of Nicosia station in Cyprus was chosen. This station is located within the urban area of the city of Nicosia (35°17'N, 33°35'E, 170m, see Fig. 4) and equipped with traditional instrumentation was op‐ erational from 1957 until 2001, when it was upgraded to an automatic station. The database used in this study comprises the maximum and minimum temperature records from this station. The maximum monthly temperature measurements are presented in Fig. 5. Also, for the classification of synoptic patterns, the ERA40 reanalysis for the period of 1958 to 2000 (covering roughly the ERA40 time window) were utilized.

The temperatures database was checked for consistency and homogeneity against measure‐ ments from nearby stations while the maximum temperatures were also checked for normal distribution fitting.

#### **3.2. Methodology**

The maximum daily temperature at Nicosia station was checked against the climatological monthly average value of the period 1961-1990. If the difference was 5°C or more, then the period was characterized as "possible heat event". If the subsequent days were also positive against this temperature test for more than three days, then the period was considered as heat event. The heat events were checked against the weather classification patterns in order to identify a connection among particular patterns and heat events. The same procedure was adopted for a difference of 3°C, since events with a 5°C difference are rare even during summer. Special care was taken when checking the last and the first day of the month whereby daily maximum temperature values were subtracted from the average climatologi‐ cal value of the two subsequent months.

Details of the Self Organizing Maps methodology used for the classification have been pre‐ sented in Michaelides et al. (2010). The 36-Cluster classification adopted also in the present

Satellite and Ground Measurements for Studying the Urban Heat Island Effect in Cyprus

http://dx.doi.org/10.5772/39313

7

The distribution of the heat events in consecutive days for 3°C and 5°C difference is illustrat‐ ed in Fig. 6. It is clearly evidenced that more than 75% of the events last for 3 to 5 days. Most of the identified heat events occur in the transition periods (i.e., Spring and Autumn). This finding is also supported by the findings in Fig. 5, where the larger variation (the area be‐ tween 25th and 75th percentile) of the average of the maximum temperatures is given for the same periods. With the exception of the periods 12 to 21 July 1978 (10 days) and 2 to 14 July 2000 (13 days), all incidents lasting more than 10 days for this station occurred in October,

Clusters 5 and 34 share most of the heat event occurrences. They are both transition period clusters with similar characteristics, exhibiting a distinctive upper level ridge over the east‐ ern Mediterranean and a deep low to the west of this ridge; Cluster 5 belongs to the cold period and Cluster 34 to the warm period. An example of a Cluster 5 member is illustrated

When these clusters appear during early Spring and late Autumn, the heat events last from 8 to 15 days, while when they appear just before or after Summer (May and Sep‐ tember) they last around 5 days. Summertime appearances of heat events are equally shared between Clusters 12, 19, 24 and 36, all characterized by warm and dry conditions

**Figure 6.** Distribution of the heat events in consecutive days for 3°C and 5°C difference

study has been recently demonstrated by Tymvios et al. (2010b).

**3.3. Results**

in Fig. 7.

November, March, April and May.

(Michaelides et al., 2010).

**Figure 4.** Location of ground stations used

**Figure 5.** Box and Whiskers plot of the maximum temperatures in Nicosia (1958 - 2000)

Details of the Self Organizing Maps methodology used for the classification have been pre‐ sented in Michaelides et al. (2010). The 36-Cluster classification adopted also in the present study has been recently demonstrated by Tymvios et al. (2010b).

#### **3.3. Results**

summer. Special care was taken when checking the last and the first day of the month whereby daily maximum temperature values were subtracted from the average climatologi‐

cal value of the two subsequent months.

6 Remote Sensing of Environment: Integrated Approaches

**Figure 4.** Location of ground stations used

**Figure 5.** Box and Whiskers plot of the maximum temperatures in Nicosia (1958 - 2000)

The distribution of the heat events in consecutive days for 3°C and 5°C difference is illustrat‐ ed in Fig. 6. It is clearly evidenced that more than 75% of the events last for 3 to 5 days. Most of the identified heat events occur in the transition periods (i.e., Spring and Autumn). This finding is also supported by the findings in Fig. 5, where the larger variation (the area be‐ tween 25th and 75th percentile) of the average of the maximum temperatures is given for the same periods. With the exception of the periods 12 to 21 July 1978 (10 days) and 2 to 14 July 2000 (13 days), all incidents lasting more than 10 days for this station occurred in October, November, March, April and May.

Clusters 5 and 34 share most of the heat event occurrences. They are both transition period clusters with similar characteristics, exhibiting a distinctive upper level ridge over the east‐ ern Mediterranean and a deep low to the west of this ridge; Cluster 5 belongs to the cold period and Cluster 34 to the warm period. An example of a Cluster 5 member is illustrated in Fig. 7.

When these clusters appear during early Spring and late Autumn, the heat events last from 8 to 15 days, while when they appear just before or after Summer (May and Sep‐ tember) they last around 5 days. Summertime appearances of heat events are equally shared between Clusters 12, 19, 24 and 36, all characterized by warm and dry conditions (Michaelides et al., 2010).

**Figure 6.** Distribution of the heat events in consecutive days for 3°C and 5°C difference

**4. Satellite estimates of temperature versus ground measurements**

is considered as the network's output.

2005b) and others.

Coll et al., 2009).

in the study.

**4.2. Methodology**

used was reduced to twelve, as shown in Table 1.

**4.1. Data**

In this Section, a methodology is presented in which the temperature estimates from the MODIS sensor onboard the Terra Satellite is contrasted with ground measurements. The methodology consists of a neural network approach in which measurements on the ground are used as input to the neural network, whereas, the temperature estimate from the satellite

Satellite and Ground Measurements for Studying the Urban Heat Island Effect in Cyprus

http://dx.doi.org/10.5772/39313

9

The neural network methodology adopted has successfully been implemented before in tackling several climatological problems in Cyprus: the prediction of maximum daily total solar irradiance (Kalogirou et al., 2002), the prediction of the daily average solar radiation (Tymvios et al., 2002, 2005a), the modeling of photosynthetic radiation (Tymvios et al.,

For the needs of this research, data from MODIS onboard the Terra satellite have been used. More specifically, the level-3 product MOD11A1 (version 5) for the period 2000-2009 was exploited, at a resolution of about 1km by 1km (0.927km). Using the available Land Surface Temperature (LST) fields derived from MODIS, a time series was established corresponding to the position of ground stations. Wan & Dozier (1996) have developed the Generalized Split Window (GSW) algorithm for the retrieval of LST, using the thermal (infrared) channels of MODIS and under different atmospheric condi‐ tions (see also, Wan, 1999, 2008). This algorithm retrieves LST on the basis of emissivi‐ ties in bands 31 and 32 of MODIS. The accuracy in estimating LST was found to be better than 1K, whereas in most cases it was better than 0.5K (Hulley & Hook, 2009;

The data base for the surface measurements used in this research consist of the hourly re‐ corded temperature at each of the automatic meteorological stations of the network operat‐ ed by the Cyprus Meteorological Service (see Fig. 4), in the period 2000-2007. Based on these data, the maximum temperature recorded in the time period 1100 – 1300 UTC (local stand‐ ard time=UTC+2 hours) was considered as the day's maximum and was subsequently used

The training of the neural network implemented requires that there are no missing data in the time series used, because the data are used in groups and are therefore inter-dependent. Therefore, the estimated LST (by the neural network implemented) is based on the data of a whole day and missing values result in the rejection of that day. Following quality control based on the above constraint, the number of automatic stations that were subsequently

Artificial Neural networks (ANN) are small autonomous computational units (algorithms) with inter-connections which, to a large extent, resemble the functioning of natural compu‐

**Figure 7.** The height pattern at 500hPa (Cluster 5) from 24 November 1962

Although it appears that some Clusters are associated with heat events over Cyprus, the connection between heat events and atmospheric circulation at 500hPa did not give definite results that any of these patterns dominate heat event occurrences (as it was possible to demonstrate in previous studies on rainfall and extreme rainfall events). There might be two reasons for this inadequacy. The first is that the window that was chosen for the classifica‐ tion does not include the synoptic patterns that influence the area sufficiently; the second reason is that, although upper air patterns at 500hPa contribute significantly to the evolution of certain surface features (such as dynamical or extreme rainfall), such an association is not so clear for the temperature field. In the search for associations of the temperature fields with synoptic patterns in the Mediterranean, it is important to consider also the lower parts of the atmosphere.

Future research concerning the connection of the weather classification patterns will be fo‐ cused into a new, much larger window that will include Northern Africa and the Middle East and a combination of classification of patterns at lower levels of the atmosphere (e.g., 850hPa, 700hPa).
