Spatial Analysis on Climatic Variations in Jordan using GIS

The country can be divided into three parts namely the Jordan valley Wadi Araba, the highlands and the plateau. Jordan is a country that undergoes climatic variability which is unequivocal occurring because of various atmospheric conditions that are related to nation-specific characteristics (Freiwan, M. and Kadioǧlu, M. Trends that are inconsistent in daily precipitation, maximum air temperature, mean air temperature, minimum air temperature, and relative air humidity were investigated by Man-Kendall Rank. In this study, linear regression analysis tests using long-term and historical data collected on a daily and monthly basis from close to 143 stationed points. Several kinds of literature exist on the variability of climatic conditions in Jordan that have been conducted over the last decade (Al-Houri, 2014; Al-Qinna et al, 2011; Bani-Domi, 2005; Freiwan & Kadioglu, 2006; Ghanem, 2013).

Most of these studies have indicated large spatial variability, especially in precipitation with significant yet unclear changes and trends at station levels (MoEnv, 1999; MoEnv, 2006; MoEnv, 2013). One of the most preferred and best methods that have been proven to reduce the prediction errors is the cokriging method encompassing well-defined variables such as elevation (Goovaerts, 2000; Diodato, 2005). The multivariate geostatistical technique consisting of digital elevation model into a spatial prediction of climatic variables tend to produce more accurate results (Weng et al, 2011). In order to achieve a fine resolution map for the climatic variables across Jordan, spatial interpolation was designed to utilize the ordinary cokriging technique between the climatic variable and the altitude. A recent study which explores 17 infilling methods (10) encompassing Kriging for daily precipitation and temperature.

The results from this study showed that Kriging is a better deterministic approach for prediction of temperature than other methods. The linear regression residuals occurring between precipitation and elevation proved to produce better results when incorporated in the original Kriging method. Furthermore, the utilization of elevation in stochastic methods for better results in the interpolation of variables has been largely recommended. Kriging method is aimed at achieving a reduction in errors and variance and therefore has a strong advantage over other estimation methods (Clark, 1979). Spatial cokriging mapping done in Jordan indicated that the relationship between climatic variables and auxiliary variables ranged between moderate and strong therefore coinciding with the studies by (Freiwan and Kadioglu, 2008). The country seemed to have an oriented spatial distribution in relative humidity.

The trend analysis showed that there are statistically insignificant trends in mean rainfall in the wider areas of Jordan. Inverse distance weighting was used in Jordan in the estimation of rainfall through observations of various rain gauge stations and the data that was collected processed using ArcGIS software utilizing IDW algorithm. Prediction of rainfall can be done using Artificial Neuron Networks through data processing in Mat lab. g. , Simonton and Osborn 1980; Tung1983) have experimented with variations in power, examining its effects on the spatial distribution of information from precipitation observations. Temperatures changes are known to have considerable effects on human health. Various spatial interpolation methods include IDW, Kriging, and Cokriging which have been explained in this paper. Of these techniques Kriging has been used widely in the estimation of temperatures (Bolstad et al, 1998; Brown & Comrie, 2002) and has been found to be a valid method with low bias by researchers as compared to other methods (Yang et al.