Spatiotemporal variations of white sky albedo upper of average in Iran

The main sources of albedo change are variations in snow cover, variations in soil moisture, droughts, and variations in vegetation phenology, forest fires, and land use/ cover changes directly related to human activities, such as deforestation, irrigation, and urbanization. Forests obtain lower albedo values than shrubs, dry crops, grasslands, and bare soils. As a result, the conversion of forests to these land cover types leads to increases in surface albedo. This potentially has local and regional feedback, since an increase in surface albedo leads to a reduction in net radiation, turbulent heat fluxes, convective clouds, and precipitation, leading to a drier atmosphere Furthermore, black carbon decreases the surface albedo when deposited on snow and glaciers because it is incorporated in snowflakes, darkening snow and ice surfaces and increasing surface melt. Aerosols like dust transferred into the atmosphere and transported by the wind into the mountains where it settles on snow and glaciers, reducing albedo and leading to enhanced warming at higher elevations. It is noted that even though precipitation is the main driver of variations in soil moisture, its impact on albedo is controlled by evaporation, soil type, irradiation, vegetation, and topography. The present paper aims to evaluate the spatiotemporal variations of white sky albedo in Iran. For this, daily Albedo datasets from Moderate Resolution Imaging Spectroradiometer (MODIS) from onboard Aqua and Terra (MCD43A3v006) were applied from 2000 to 2019 with a spatial resolution of 500 × 500 m. MODIS provides black-sky albedo for direct and white-sky albedo for isotropic diffuse radiation at local solar noontime. For this, daily white sky albedo datasets from Moderate Resolution Imaging Spectroradiometer (MODIS) from onboard Aqua and Terra (MCD43A3v006) were applied. One of the main applications of the principal component analysis (PCA) is climatic zoning which is a method of determining environmental changes in temporal dimensions. A plethora of studies have been conducted using principal component analysis in the field of climatology but little has been done in relation to the albedo variation. To the best of the authors’ knowledge, this study uses a technique that has not been applied in scientific texts related to Modis albedo data. The questions that we will address in this study include: what is the temporal-spatial behavior of white sky albedo in Iran? How many components explain the variation of white Sky Albedo? What factors will distinguish white sky albedo in Iran?

In this investigation, daily white sky albedo datasets from Moderate Resolution Imaging Spectroradiometer (MODIS) from onboard Aqua and Terra (MCD43A3v006) were applied for the period of 2000-02-24 to 2019-06-03 (7040 days) with a spatial resolution of 500 × 500 m. Among the various MODIS datasets, white sky albedo datasets were extracted. The daily white sky albedo was averaged over the 19-year period for each pixel inside the border of Iran. The size of this array was 7541502 pixel *12month. Long-term monthly and seasonal means were also calculated by the available time series data. In the next step, the PCA method was applied to analyze the spatio-temporal variations of albedo in Iran. PCA is a method to reduce the number of the data and convert them into several finite components so that these few components explain the largest amount of the variance. This procedure is searching for the variable with the largest amount of the variance in space (PCA was invented in 1901 by Karl Pearson , and it was later developed by Harold Hotelling in the 1930s. In this method, initial variables are converted into n principal components each being a linear combination of the variables. In this way, the first principal component has the largest possible variance, and the components afterward explain a smaller percentage of the variance. Principal component analysis leads to the analysis of space-time array into two time-array and space-array. In this case, it is possible to identify what important spatial patterns the primary data have and at what time periods each of these patterns has been active or inactive. Because the principal components are finite, the temporal and spatial patterns introduced by the first component are more important than the temporal and spatial patterns of subsequent components.

The long-term average of Iran's white sky albedo was calculated; The results showed that the average albedo of spring, summer, autumn and winter in Iran 14.99%, 16.06% , 15.53%, and 19.58%, respectively. The evaluated long-term mean white sky albedo for each season showed that the highest value had occurred in winter. The dramatic increase in this value was placed along the Zagros, Alborz, Sahand, and Sabalan Mountains which exceeds 90 to 100 percent in some places. In the next step, the temporal-spatial variations of white sky albedo values in Iran were analyzed using principal component analysis, and the results showed that the three main components are able to explain 97.7% of the data variation. The first component explains more than 73%, of the total changes, the second component more than 20.8% and finally the third component explains more than 3.9% of the changes.

Spatial analysis revealed that the values which are higher than the mean are places in highlands and mountainous regions of Iran, such as the Zagros and Alborz Mountains, Sabalan, Sahand mountains and Zard Kuh-e Bakhtiari, which are associated with snow cover Therefore, the first component was named as snow cover as the maximum variance of albedo was explained by snow cover. The spatial analysis of the second component revealed that higher values were placed in small areas across Iran including, Hajaligholi desert Gavkhuni wetland, Qom salt lake, Sirjan salt lake parts of Loot desert. In the second component, most of the cell's scores upper of average in Iran corresponded to areas covered with salt. As the maximum variance is explained with salt cover, therefore, it can be named as the salt land. Spatial analysis indicated that in very limited parts of Alborz, Zagros, Alam-Kuh Mountain , Sahand, and Sabalan mountains Kino Mountain values are mostly positive which is related to the glaciers (regions with appropriate conditions to keep the snow cover in most of the year) and is the origin of the seasonal or permanent rivers Therefore, according to the cell scores (upper of average in Iran) in the third component, it was found that these cells corresponded to the , so it was named as the glacier component.