Frequency Distribution Patterns of Precipitable Water in Iran

Introduction
Precipitable Water (PW) is highly variable in space and time, being one of the most important abundant greenhouse gases to play a crucial role in the study of climate change, hydrological cycle, energy budget, and numerical weather prediction. Knowledge about spatial and temporal variability of PW is important for understanding climatic processes along with monitoring drought conditions and desertification processes (Kaufman & Gao, 1992). It is, therefore, necessary to obtain the distribution condition of water vapor in the atmosphere and understand the effects of spatiotemporal variations of PW on regional, meso-micro scales as well as global climate change (Wang, 2013). PW has a very short life cycle in atmosphere. This rapid turnover, accompanied by temperature variations with altitude and geography, distance to sea, evapotranspiration, and moisture advection, causes an irregular PW distribution in the atmosphere, both horizontally and vertically. Thus this study aims at identifying the distribution patterns of PW in Iran and correlate these patterns with elevation and distance to sea.
Materials and Methods
The present research uses MODIS Aqua data (MYD05_L2. A V06) and selects the data with spatial resolution of 1 km (Near Infrared). The selected study period covers from 2002/07/04 to 2017/07/25 (5501 days), taken from NASA web site. These data are erroneous in the range between 5% and 10% (Kaufman & Gao, 2003). The spatial resolution of PW data are 1 km and the temporal resolution, twice per day. Afterwards, by using the functions, these data are converted from Level_2(swath data) to Level_3(grid data) and PW values interpolate on sinusoidal grid in 1800×2700 matrix with 1 km spatial resolution and daily temporal resolution. These data have been extracted for pixels within the political boundary of Iran and result in a matrix with 1884080 rows (locations) and 200 columns (PW classes). Then on the matrix’ base, frequency distribution is calculated in 1 mm intervals from 0-199 mm for every pixel (1884040×200). Finally, Principal Component Analysis (PCA) is performed, allowing the identification of frequency distribution patterns in Iran. The effects of altitude and distance to sea on these patterns are analyzed. The special program is developed and employed in MATLAB software for data analysis.
Results and Discussion
The spatial distribution of atmospheric humidity in Iran is controlled by the altitude, distance to sea, and moisture advection. Based on the results, the mean annual PW of the country is about 12 mm. PW is maximum near the southern and northern coasts of the country, with the highest and lowest PW value observed near the Oman Sea coast (31 mm) and the peak of Damavand (3 mm), respectively. Results from PCA show that 95% of spatial variation of PW can be explained through 4 components. Accordingly, local factors like distance to sea and altitude are the most important ones in spatial distribution of PW. The study of the correlation between distance from the sea and frequency distribution patterns of PW shows the effect of distance and proximity to the sea in the frequency distribution patterns. This is more evident in the first and second components. As expected, up to a distance of approximately 250 kilometers in the first component and 150 kilometers in the second, the amount of PW will gradually decline. From then on, the spatial pattern of PW is affected by altitude and morphology rather than by distance from sea and sea\land breeze. In the third component, due to the formation of a moisture convergence belt at approximately 11 and 4 km, respectively, on the south and north coasts, the amount of atmospheric moisture is maximum. Then from 11 to 66 kilometers due to the Alborz Range, which is a short distance from the Caspian Sea, the amount of PW is minimal. Minimal atmospheric humidity on the southern coast occurs approximately at 250 kilometers away from the sea. In the South Coast, moisture penetrates the country further away from the coast, as it is smoother than the North Coast. Thus, sea moisture enters through the straits of Kahnouj Area into Jazmourian Plain, distinguishing it from its surrounding areas in terms of moisture. Moisture in the Caspian Sea enters the Tarom Valley through Manjil Strait. The spatial distribution of moisture in the western, middle, and eastern Persian Gulf coasts does not have a similar pattern. This difference is because of factors like the dominance of sea-land breeze in the eastern areas of Bushehr and the presence of small firth and bays in the area that increase its atmospheric moisture, in comparison to the surrounding areas. The amount of moisture in the coasts of Oman Sea is clearly different from PW of the Persian Gulf. PW MODIS is also overestimated in places such as near beaches with high temperatures and humidity.
In addition to the altitude and distance to sea, the role of moisture advection should not be ignored. In the coastal region, the variability is caused by high temperature and moisture advection, whereas in areas far from coastline, it is the altitude that causes many spatial differences in moisture distribution.
Conclusion
Although Iran is bounded from the north and south to the sea, atmospheric moisture is very low in the country. According to the results from this paper, minimum and maximum difference of PW is about 27 mm. Thus, in a region above 3000 m from the sea level, PW falls below 6 mm, and the coasts of Oman Sea are above 26 mm, 60% of the time. This means that in spite of the great sources of water both in the south and the north, Iran’s atmosphere suffers from poor moisture. Topography acts as a barrier for moisture to enter inland regions from both north and south seas. Inland, the altitude plays a crucial role for frequency distribution of PW, while in the coastal regions, both moisture advection and temperature are culprits. In this way, moisture advection is an important factor to justify spatiotemporal variations of PW in Iran well. And it is this parameter that affects water budget.