Tayeb Raziei

Regional Drought Modes in Iran Using the SPI: The Effect of Time Scale and Spatial Resolution

In the present paper, regional drought modes in Iran are identified applying the Principal Component Analysis (PCA) and Varimax rotation to the Standardized Precipitation Index (SPI) computed on different time scales. Data used include gridded monthly precipitation covering the period 1951–2007 retrieved from the Global Precipitation Climatology Centre (GPCC) archive with different spatial resolutions (2.5, 1 and 0.5° resolution). The objective of the study is twofold: (i) Investigate the stability of drought spatial modes as a function of the SPI time scales used for monitoring the different kinds of drought, (ii) Evaluate the impact of the spatial resolution of gridded data on drought regionalization. For the coarse spatial resolution of 2.5°, results show four drought modes of distinct variability, which remain quite stable when the SPI time scale is varied from 1- to 24-month. Differently, for higher spatial resolutions drought modes appear more sensitive to the index time scale and become less spatially homogeneous as the time scale is increased. Moreover, the number of identified modes (sub-regions) may reduce to three or two, but in all cases the most well defined sub-region appears to be the southern one. This suggests that both the spatial resolution of precipitation data and the time scale may affect drought regionalization, i.e. the number of drought modes and their spatial homogeneity.

Spatial patterns and temporal trends of precipitation in Iran

Spatial patterns of monthly, seasonal and annual precipitation over Iran and the corresponding long-term trends for the period 1951–2009 are investigated using the Global Precipitation Climatology Centre gridded dataset. Results suggest that the spatial patterns of annual, winter and spring precipitation and the associated coefficients of variation reflect the role of orography and latitudinal extent between central-southern arid and semi-arid regions and northern and western mountainous areas. It is also shown that precipitation occurrence is almost regularly distributed within the year in northern areas while it is more concentrated in a few months in southern Iran. The spatial distribution of Mann–Kendal trend test (Z statistics) for annual precipitation showed downward trend in north-western and south-eastern Iran, whereas western, central and north-eastern exhibited upward trend, though not statistically significant in most regions. Results for winter and autumn revealed upward trend in most parts of the country, with the exception of north-western and south-eastern where a downward trend is observed; in spring and summer, a downward trend seems to prevail in most of Iran. However, for all seasons the areas where the detected trend is statistically significant are limited to a few spot regions. The overall results suggest that the precipitation is decreasing in spring and summer and increasing in autumn and winter in most of Iran, i.e. less precipitation during the warm season with a consequent intensification of seasonality and dryness of the country. However, since the detected trends are often not statistically significant, any stringent conclusion cannot be done on the future tendencies.

Spatial patterns and temporal trends of daily precipitation indices in Iran

Spatial patterns of daily precipitation indices and their temporal trends over Iran are investigated using the APHRODITE gridded daily precipitation dataset for the period 1961–2004. The performance and limitations of the gridded dataset are checked against observations at ten rain-gauge stations that are representative of different climates in Iran. Results suggest that the spatial patterns of the indices reflect the role of orography and sea neighborhoods in differentiating central-southern arid and semi-arid regions from northern and western mountainous humid areas. It is also found that western Iran is impacted by the most extreme daily precipitation events occurring in the country, though the number of rainy days has its maximum in the Caspian Sea region. The time series of precipitation indices is checked for long-term trends using the least squares method and Mann-Kendall test. The maximum daily precipitation per year shows upward trends in most of Iran, though being statistically significant only in western regions. In the same regions, upward trends are also observed in the number of wet days and in the accumulated precipitation and intensity during wet days. Conversely, the contribution of precipitation events below the 75th percentile to the annual total precipitation is decreasing with time, suggesting that extreme events are responsible for the upward trend observed in the total annual precipitation and in the other indices. This tendency towards more severe/extreme precipitation events, if confirmed by other datasets and further analyses with longer records, would require the implementation of adequate water resources management plans in western Iran aimed at mitigating the increasing risk of intense precipitation and associated flash floods and soil erosion.

SPI Modes of Drought Spatial and Temporal Variability in Portugal: Comparing Observations, PT02 and GPCC Gridded Datasets

Regional drought modes in Portugal are identified applying the Principal Component Analysis (PCA) and Varimax rotation to the Standardized Precipitation Index (SPI) computed on various time scales using the three precipitation datasets covering the period 1950–2003: (i) The observation dataset composed of 193 rain-gauges distributed almost uniformly over the country, (ii) the PT02 high-resolution gridded dataset provided by the Portuguese Meteorological Institute, and (iii) the GPCC dataset with 0.5° spatial resolution. Results suggest that the three datasets well agree in identifying the principal drought modes, i.e. two sub-regions in northern and southern Portugal with independent climate variability. The two sub-regions appear stable when the SPI time scale is varied from 3- to 24-month, and the associated rotated principal component scores (RPCs) do not show any statistically significant linear trend. The degree of similarity between the rotated loadings or REOFs of different SPI time scales for the three used datasets was examined through the congruence coefficients, whose results show a good agreement between the three datasets in capturing the main Portuguese sub-regions. A third spatial mode in central-eastern Portugal was identified for SPI-24 in PT02, with the associated RPC characterized by a statistically significant downward trend. The stability of the identified sub-regions as a function of studied time period was also evaluated applying the same methodologies to a set of three different time windows and it was found that the southern sub-region is very stable but the northern and central-eastern sub-regions are very sensitive to the selected time window.

A precipitation regionalization and regime for Iran based on multivariate analysis

Monthly precipitation time series of 155 synoptic stations distributed over Iran, covering 1990–2014 time period, were used to identify areas with different precipitation time variability and regimes utilizing S-mode principal component analysis (PCA) and cluster analysis (CA) preceded by T-mode PCA, respectively. Taking into account the maximum loading values of the rotated components, the first approach revealed five sub-regions characterized by different precipitation time variability, while the second method delineated eight sub-regions featured with different precipitation regimes. The sub-regions identified by the two used methods, although partly overlapping, are different considering their areal extent and complement each other as they are useful for different purposes and applications. Northwestern Iran and the Caspian Sea area were found as the two most distinctive Iranian precipitation sub-regions considering both time variability and precipitation regime since they were well captured with relatively identical areas by the two used approaches. However, the areal extents of the other three sub-regions identified by the first approach were not coincident with the coverage of their counterpart sub-regions defined by the second approach. Results suggest that the precipitation sub-region identified by the two methods would not be necessarily the same, as the first method which accounts for the variance of the data grouped stations with similar temporal variability while the second one which considers a fixed climatology defined by the average over the period 1990–2014 clusters stations having a similar march of monthly precipitation.

Ground-Based GPS Measurements of Precipitable Water Vapor and Their Usefulness for Hydrological Applications

Half-hourly ground-based GPS measurements of precipitable water vapor (PWV) from January 2009 to December 2012 are analyzed to investigate their potential for hydrological applications at basin level. In particular, the usefulness of these high temporal resolution data for monitoring extreme weather conditions, such as floods and meteorological dry/wet spells, is discussed. Two sample GPS stations in U.S. from the SoumiNet network are considered that have rather continuous data for the last four years and a few missing values. Results suggest that: (i) A flood event is characterized by an anomalous increase of PWV accompanied by a sudden lowering of surface pressure; (ii) Precipitable water tendency (DPWV) becomes increasingly small moving from half-hour to monthly time scale, but not negligible compared with both the moisture flux divergence div(Q) and the imbalance between actual evapotranspiration and precipitation (E–P), especially during spring and fall; (iii) GPS observations, jointly with other meteorological data, can provide an accurate estimate of the imbalance (E–P) that is of interest for drought assessment, and of the terrestrial water storage rate of change that is known to be difficult to measure; (iv) the availability of on-site precipitation observations allows the computation of precipitation efficiency, which is a key parameter for estimating the water availability in a given area and monitoring dry/wet spells. It appears that for a comprehensive monitoring of a river basin, a GPS network that encloses the area of concern, equipped with meteorological ground sensors, is suitable and desirable.

An analysis of daily and monthly precipitation seasonality and regimes in Iran and the associated changes in 1951–2014

Daily and monthly total precipitation of 155 synoptic stations with relatively regular distribution over Iran, covering the 1990–2014 period, were used to investigate the spatial pattern of precipitation seasonality and regimes over Iran, using a set of precipitation seasonality indices. The results suggest a strong agreement between the indices computed at monthly time scale. The result also shows a latitudinal decreasing gradient from the lower index values in the north to the highest values in the south of Iran, suggesting a strong negative relationship between the latitude and the indices. A weak but statistically significant association was also found between the indices and the longitude, showing a gradual west-east contrast between the mountainous western Iran and the central-eastern lowlands and deserts of the country. The spatial patterns of the indices well agree in revealing different precipitation regimes in Iran, in spite of the observed discrepancies in their areal extent of the regions identified. All the indices characterized northern Iran by a precipitation regime having a moderate seasonality, while the mountainous areas of the western and northern Iran are featured by a marked precipitation regime possessing a longer dry season. However, the most seasonal precipitation regime with the longest dry period describes the southern country and some spot areas of the central-eastern Iran. The spatial distribution of the seasonal precipitation regimes and the month and season of maximum precipitation amounts across Iran was also identified, suggesting that from the 24 possible precipitation regimes over the globe, eight were found in Iran, from which a precipitation regime with the highest precipitation amount in winter, followed by autumn, spring, and summer characterized most parts of the country. January and JFM were also found as the month and season of maximum precipitation in a majority of stations distributed over Iran, respectively. The precipitation concentration index (DPCI) computed using daily precipitation data ranges between 0.56 and 0.76 across the country; nonetheless, the values between 0.64 and 0.70 characterized a majority of stations distributed over most parts of Iran. Contrarily to the indices computed at monthly time scale, the DPCI does not show a clear latitudinal pattern over the country. The Mann–Kendal trend test and the Sen slope estimator were applied to the computed indices relative to 16 stations with the longest and complete precipitation records during 1951–2014 time period. The indices time series showed no significant trend in the majority of the stations, indicating that the precipitation regimes of the studied stations did not change over 1951–2014 period.