Temporal and spatial trend detection of maximum air temperature in Iran during 1960–2005

• Published in: Global and planetary change Vol. 111; pp. 97 - 110
• Main Authors: Kousari, Mohammad Reza, Ahani, Hossein, Hendi-zadeh, Razieh
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.12.2013; Elsevier
• Subjects: air temperature; Arid, subarid, mediterranean climates; autocorrelation; Bgi / Prodig; climate change; Climatology; Climatology. Bioclimatology. Climate change; Clusters; cold; Earth, ocean, space; emissions factor; Exact sciences and technology; External geophysics; greenhouse gas emissions; Iran; Land use; land use change; Meteorology; Physical geography; Seasons; Slopes; Spatial distribution; T max; Time; Time series; time series analysis; trend; Trends; Iran; Middle East; Asia; Iran; trend; T max; climate change; 1970-1980; trend-surface analysis; time variations; 1960-1970; Atmospheric temperature; 1990-2000; 2000-2010; Maximum temperature; spatial variations; 1980-1990; Temperature; Space time; Climatic trend; Air temperature; Time series
• ISSN: 0921-8181; 1872-6364
• DOI: 10.1016/j.gloplacha.2013.08.011

Trends of maximum air temperature (T max) were investigated in three time scales including annual, seasonal, and monthly time series in 32 synoptic stations in the whole of Iran during 1960–2005. First, nonparametric Mann–Kendall test after removal of the lag-1 serial correlation component from the T max time series was used for trend detection and spatial distribution of various trends was mapped. Second, Sen's slope estimator was used to determine the median slope of positive or negative T max trends. Third, 10-year moving average low-pass filter was applied to facilitate the trend analysis and the smoothed time series derived from the mentioned filter were clustered in three clusters for each time series and then were plotted to show their spatial distribution patterns in Iran. Results showed that there are considerable significant positive trends of T max in warm months including April, June, July, August and September and warm seasons. These trends can be found in an annual time scale which indicated almost 50% positive trends. However, cold months and seasons did not exhibit a remarkable significant trend. Although it was rather difficult to detect particular spatial distribution of significant trends, some parts in west, north-east and south-east and central regions of the country showed more positive trends. In an annual time scale, Kermanshah located in west regions indicates most change at (+) 0.41°C per decade. On the one hand, many clusters of normalized and filtered T max time series revealed the increasing trend after 1970 which has dramatically risen after around 1990. It is in accordance with many other findings for temperature time series from different countries and therefore, it can be generated from simultaneous changes in a bigger scale than regional one. On the other hand, the concentration of increasing trends of T max in warm seasons and their accordance to plants growing season in Iran can raise the importance of the role of frequent reported land use changes during past decades. Generally, the more sophisticated and comprehensive researches are needed to determine the role of different factors such as the emission of greenhouse gases and land use changes influencing temperature trends in Iran. •Focusing on T max trend, spatially and temporally in Iran.•Implementation of non-parametric Mann-Kendall test and Sen’s slop estimator.•Application of smoothed and clustered time series to facilitate trend analysis.•The frequent trends in Iran were observed in warm months and seasons than cold ones.•The trends of T max may refer to wide landuse changes in Iran during past decades.