Journal of Spatial Analysis

Climate change is one of the most significant threats facing the world today. One of the most important consequences of climate change is increasing frequency of climate hazards, mainly heat waves. This phenomena has a robust impacts on human and other ecosystems. The aim of this study is investigating changes of heat waves in historical (1980-2014) and projected (2040-2074) data in northern cost of Persian Gulf.

The focus here is on Mean daily maximum temperature and Fujibe index to extract heat waves. For this purpose 6 weather stations locating in north coast of Persian Gulf, Iran, are used (table 1).

Table1: weather stations

  In addition, 4 model ensemble outputs from the Coupled Model Intercomparison Project Phase 5 (CMIP5) are used to project future occurrence and severity of heat waves (2040 to 2070), under Representative Concentration Pathways 8.5 (RCP8.5), adopted by the Intergovernmental Panel on Climate Change for its Fifth Assessment Report (AR5) (table 2).

Table2: List of the AR5 CMIP5 Used Models

The output of models is downscaled using artificial neural network method (ANN). A feed-forward network of multi-layer perceptron with an input layer, a hidden layer and an output layer is used for this purpose. 73 percent (1980 – 2000) of the data is used for training and 27 percent (2000-2005) for testing ANN models. Root Mean Square Error (RMSE) is used as an indicator of the accuracy of Models.

RMSE=

Here   is the outputs of ANN models (downscaled data) and is the observation data.

Fujibe et all (2007) used an index based on Normalized Thermal Deviation (NTD) for extracting long-term changes of temperature extremes and day to day variability using following equations:

Where N is the number of days in the summation except missing values. Then nine-day running average was applied three times in order to filter out day-to-day irregularities.

=(i,j,n)=T(i,j,n)-T(I,j)

The departure from the climatic mean is given by

=

If NTD >2 and at least lasts for 2 days it determine as a heat wave.

Results

Table 3 shows the results of downscaling selected GCM models.

Table 4 compares the frequency of heat waves for GCMs and historical data.

For historical data, heat waves are more frequent in Abadan station than other stations. There is an increasing trend in the occurrence of heat waves in historical data and monthly frequency of heat waves show the highest amounts for summer.

For both historical and future data 2 days listening heat waves are more frequent.

Table 5 shows seasonal changes of heat waves for historical data and GCMs.

In recent years the frequency of heat waves is increasing in all studied stations. Coincide with Russia and Europe, the highest amounts of heat waves is occurred in 2010 in northern coast of Persian Gulf and this is adopted Esmaeilnezhad et all (2013), Gavidel (2015) and Azizi (2011).