No description available.

Extreme climate responses such as floods or droughts pose multi-dimensional hazards to critical infrastructure and the most vulnerable sectors of society; these hazards may increase under climate change. The extreme rainfall events driving these responses may arrive non-uniformly in time, clustering on intra- and inter-annual scales; yet the dependent relationship between events is often ignored. This thesis examines extreme daily rainfall within year clustering to determine whether changes in their temporal pattern are apparent in observational records. It then identifies the key atmospheric variables which drive event frequency and intensity, before testing hypotheses related to clustering. Extreme rainfall regions were developed from the station maxima of a comprehensive new set of 223 daily rainfall observations, spanning the period 1856-2009. The observations are contained within 14 regions which represent the distinctive seasonal clustering, orographic and atmospheric variations in UK extreme rainfall. Significant increases in annual maxima and associated return frequencies over the period 1961-2009 were observed from a Generalized Extreme Value (GEV) analysis. Increases in spring, autumn and winter maxima and their estimated return frequencies were also found. Estimates from summer maxima were variable across the country but indicated an increase in the highest intensity events. Extreme rainfall seasonal clustering and the dependence on sea surface temperatures (SST), air temperature range and the North Atlantic Oscillation (NAO) were represented in flexible GEV and Poisson parameter estimates using Vector Generalized Additive Models. There is a strong negative correlation with air temperature range, reflecting heightened event intensity and probability when the diurnal temperature range is at its lowest. Event frequency is positively correlated with SST for all regions; event magnitude is dependent on either SST or NAO with a north-south divide. While the timing of events has not changed substantially, event probability has increased - resulting in greater within-year clustering. Climate projections indicate increasing SST and decreasing temperature range; this extreme rainfall model corroborates projected increases in event intensity and frequency.

Rainfall is a key observable of the weather, of importance to society. Rain gauges only provide point measurements; for areal rainfall information a radar may be used. Radars measuring just reflectivity (Z) may suffer large errors in derived rainrates due to drop size distribution (DSD) variations. The introduction of dual-polarisation radars to operational networks should lead to improvements in rainfall estimation, although these radars suffer high noise levels in polarisation parameters (~ 0.7 dB in differential reflectivity [ZDR]).

Water is essential for life. In the Arabian Peninsula (AP), rainfall is irregular, infrequent and low. Climate studies show that the AP receives between less than SOmm per year and more than 2S0mm per year depending upon location. Therefore rainfall monitoring and modelling are very important in optimising the use of this scarce resource. Monitoring rainfall using satellite observations is an alternative method which can solve the problem of the inadequate rainfall monitoring by ground-based methods (raingauge and radar). In this project, rainfall variability over the study area was shown using raingauges observations. The Tropical Applications of Meteorology using SATellite (TAMSAT) approach has been tried to estimate rainfall in the Arabian Peninsula. It depends on the use of cold cloud duration based only on thermal infra- red imagery. Then, the rainfall estimates were utilized to evaluate the rainfall ERA- Interim reanalysis dataset over the study area, and the tele-connection between the rainfall variability over the southwest mountainous region of the AP and the southwesterly monsoon was derived using the ERA-Interim data. The results show that the AP has extreme temporal/spatial variation of rainfall. Most of the' study area receives rainfall between October and May while rainfall occurs in the southwest region (SWAP) throughout the year by getting two rainy seasons (winter and summer). It was found that the TAMSAT approach performs well over the SWAP during summer which led to conducting an evaluation of rainfall ERA-Interim date only over this region. The evaluation has shown that the rainfall reanalysis data captures the summer intra/inter-annual rainfall pattern while its rainfall values are overestimated specifically for September. The analysis of the connection between rainfall variability over the SWAP and the southwesterly monsoon gave indication of the monsoon influences. The main finding is the relationship between weak (break)/strong (active) monsoon and the increase/decrease of rainfall over the study area in terms of providing perceptible water which enhances the creation of rainy clouds.

The improvement and further development of a laboratory catchment and rainfall simulator is described. The rainfall simulator enabled the reproduction of rainfall of any intensity between 0 and 120mm/hr, with properties similar to those of natural rainfall. The rainfall simulator was subsequently linked to a process control computer, which could control the apparatus to produce either a single or a series of rainstorms. Storms with uniform or varying rainfall intensities could be simulated, with the resulting runoff and percolation being measured and then-recorded by the computer. A description is given of the hardware and computer programmes which were developed to operate the equipment. An experimental programme was conducted on the apparatus to examine the relationships between intensity, runoff and infiltration, and other hydrological properties of a catchment. Particular attention was given to a soil's infiltration capacity, which was found to vary with rainfall intensity. This lid to the derivstion of an infiltration equation in which rainfall intensity is a variable. A study of the rate of recovery of a soil's infiltration capacity enabled an empirical formula to be established which enables a soil's infiltration capacity to be determined at a time after rainfall has ceased.

Climate change is a huge environmental issue that our society currently faces. This thesis develops and tests two bias correction methods for regional climate simulations of precipitation and runoff . Biases in the soil water physics are corrected by including new physics in the soil moisture parameterisation and the regional model inputs are corrected statistically. Case studies are performed on the Olifants River basin in the Limpopo region of South Africa using the Weather Research and Forecasting (WRF) regional climate model. Accurate knowledge of water availability in this water-stressed region is of great importance for adaptation and future water policy development. The concept of tightly bound water, in which a reservoir of soil water is held stationary within small soil pores but is still available for evapotranspiration, is parameterised for the first time within the land surface scheme of a regional atmosphere-land surface model. Results of a WRF simulation forced by re- analysis show that the standard NOAH land surface scheme over-estimates mean annual runoff by 120% with respect to observations, despite rainfall and atmospheric conditions similar to observed. Use of the tightly bound water scheme within the NOAH model reduces this bias to 22%. Simulations with the WRF model forced with 1980s and 2040s CCSM3.0 general circulation model data show that the tightly bound water scheme significantly reduces runoff in different climates. The new scheme projects a 10% decrease in runoff by the 2040s compared to a 4% decrease projected by the standard model. A new quantile-mapping bias-correction of inputs to regional climate models is proposed. Linear correction and quantile-mapping methods are implemented to correct CCSM3.0 data using re-analysis. Simulation results show a significant difference between the correction methods. The results indicate that the quantile-mapping correction method could be developed to help produce more accurate regional climate predictions for impact studies.

The central theme of this research project was the investigation of the spatial variation of rainfall over a small area. The rainfall was measured over a 369 ha basin in East Devon by a network of forty-five inexpensive gauges installed at 30.5 cm. The accuracy of the rain gauges was evaluated and the systematic gauge error amounted to 6%. The gauge sampling error, or random error, of rainfall measurement was found to be approximately 4%. The rain gauge network revealed patterns of rainfall that varied considerably from storm event to storm event. The spatial variability of rainfall of the storms as expressed by the coefficient of variation exceeded 4.5% on all occasions and 10% on eighteen occasions. The variability was not simply due to systematic or random errors, even though the amounts of rainfall involved tended to be low; the range in rainfall (decile range) was no more than 4.0 mm for forty-two storm events. The storm rainfall amounts were related to various meteorological and topographical factors. The storms of convectional origin had a greater spatial variability than storms experiencing no convection and more clearly-defined rainfall gradients. Storms of Tropical air mass origin had a lower variability than storms of Arctic or Polar origin whilst Arctic storms had the greatest variability for low rainfall amounts. There was a tendency for storm rainfall to increase downwind particularly with the wind from a southerly direction, probably as a result of interaction with the topography. The degree of explanation of the spatial variation of storm rainfall afforded by the topographical variables of altitude, exposure and vegetation influence was low.

A new stochastic rainfall field model is described which employs the Turning Bands Method (TBM) to transform a unidimensional Gaussian process, generated by the fractional differencing process along a line, into a multidimensional space-time Gaussian process with a specified space-time correlation structure. Transformations are applied to give the rainfall process a non-Gaussian and non-stationary structure. A correction factor is introduced into the model to take account of the effect of topography on rainfall. The model has been applied to the small upland Wye catchment in mid Wales (area 10.55 km2) and shown to reproduce satisfactorily the statistics and correlation structure of observed hourly point rainfall. As an extension to the rainfall field model, a new technique of conditional simulation has been used to generate the rainfall fields. The conditionally simulated rainfall fields reproduce exactly the observed point rainfalls at measurement points and likely realizations of rainfall fields between points. Rainfall fields generated by the above mentioned rainfall field model and the conditional simulation technique are fed directly into the Systeme Hydrologique European (SHE) model and the sensitivity of runoff prediction errors to (i) level of space-time correlation (ii) sampling of rainfall with different schemes in space and (iii) antecedent conditions are explored. It is found that in case of Wye catchment the errors deriving from sampling procedure used are generally small when rainfall fields were based on observed correlation structure. Sensitivity of errors to different correlation levels give the impression that errors increase with a decrease in correlation level. Further it is noticed that this trend of errors is more pronounced in `dry' catchment conditions as compared to `wet' catchment conditions. Overall the results for the small Wye catchment illustrate that the catchment acts as a smoother of the spatially distributed rainfall input at this spatial scale and for the rainfall regime in question. However, the results imply that for the typical raingauge densities encountered for larger catchments, significant errors may occur.

Η παρούσα εργασία διαπραγματεύεται την επεξεργασία ετήσιων και εποχικών χρονοσειρών βροχόπτωσης από 12 σταθμούς της Δυτικής Ελλάδας για την περίοδο 1975-2004. Επιπλέον για τους 8 από τους συνολικά 12 σταθμούς όπου υπήρχε η δυνατότητα, η επεξεργασία αφορά μια περίοδο 50 ετών (1956-2005). Χρησιμοποιώντας ως κλιματικό δείκτη το ετήσιο βροχομετρικό ύψος προκύπτει ότι το σύνολο των 12 σταθμών χαρακτηρίζεται εν γένει από έναν συνδυασμό ημίυγρου ή υγρού κλιματικού τύπου. Χρησιμοποιώντας τον μη-παραμετρικό έλεγχο των Mann-Kendall για την εξακρίβωση παρουσίας τάσεων σε βάθος χρόνου, για την περίοδο 1975-2004 δεν διαφαίνεται η ύπαρξη κάποιας σημαντικής τάσης εκτός από τις ετήσιες βροχοπτώσεις του Πύργου που εμφανίζουν μία σημαντικά αρνητική τάση. Την περίοδο από το 1956-2005 προκύπτουν σημαντικά αρνητικές τάσεις τόσο σε εποχική βάση (κυρίως την άνοιξη) όσο και σε ετήσια για τους μισούς από τους οκτώ σταθμούς που εξετάστηκαν. Η Γάμμα κατανομή είναι εκείνη που περιγράφει καλύτερα το φυσικό μέγεθος ύψος βροχόπτωσης και στην περίπτωση μας προσδιορίζονται ανά σταθμό και για την περίοδο 1975-2004 (σε εποχική και ετήσια βάση), οι παράμετροι της με την βοήθεια της μεθόδου μέγιστης πιθανοφάνειας. Στα πλαίσια της φασματικής ανάλυσης, για να εξακριβωθεί η ύπαρξη ή οχι περιοδικότητας στην τιμή της διασποράς των εποχικών και ετήσιων τιμών βροχόπτωσης χρησιμοποιούνται οι 7 σταθμοί για τους οποίους υπάρχει επάρκεια μετρήσεων με εξεταζόμενη περίοδο την πεντηκονταετία 1955-2004 και κάνοντας χρήση της μεθόδου Blackman-Tukey. Προκύπτει με την εν λόγω μέθοδο ότι κατά την διάρκεια του φθινοπώρου και της άνοιξης δεν διαφαίνονται κάποια σαφή στοιχεία περιοδικότητας στην διασπορά των υψών υετού των 7 σταθμών. Αυτό δεν ισχύει όμως για τον χειμώνα αλλά και σε ετήσια βάση, όπου στα φάσματα των τιμών υετού των σταθμών αποκαλύπτονται τρεις περιοχές συχνοτήτων περιοδικότητας που μοιάζουν αρκετά μεταξύ τους. Αυτό αντανακλά το γεγονός ότι σε γενικές γραμμές οι σταθμοί της Δυτικής Ελλάδας επηρεάζονται από τα ίδια περιπου βαρομετρικά συστήματα και άρα είναι φυσιολογικό να εμφανίζουν παρόμοιες συνιστώσες περιοδικότητας στις διασπορές των τιμών υετού τους. / This work deals with the processing of annual and seasonal precipitation series from 12 stations of West Greece for a 30-year study period (1975-2004). Moreover for 8 out for τηε 12 stations where possible, the processing uses a 50-year study period (1956-2005). By using the annual precipitation height as an climatic index it follows that the total of the twelve stations is characterized generally by a combination of semi-wet and wet climatic type. Making use of nonparametric Mann-Kendall test for ascertaining the existence of trend, it doesn't follow any significant trend for the 30-year period (1975-2004), with the exception of the annual precipitation heights of Pirgos that show a significant negative trend. During the 50-year period (1956-2005) significant negative trends occur in seasonal (mainly during spring) and annual basis as well, for half of the eight stations which have been examined. Gamma distribution is tha type of statistical distribution that describes more effectively the physical quantity precipitation height, and in our case its parameters per station are being computed for the period 1975-2004, by using the maximum likelihood method. Under the framework of a Spectral analysis of the precipitation series (for the verification of periodicity in the variances of precipitation rates) , 7 stations are used for a 50-year study period (1955-2004) by using the Blackman-Tukey method. It follows after this method has been used, that precipitation series don't appear any periodicity during autumn and spring seasons. This is in contrast with the winter season and the annual rainfall values as well, where three parts of periodicity in the spectra of the stations appear that bear a common resemblance. This depicts the fact that genarally the total of West Greece stations are influenced by almost the same barometric pressure systems which leads to the variances of precipitation rates to appear common periodicity components.

Ο Δείκτης της Νότιας Κύμανσης (Southern Oscillation Index, SOI) περιγράφει την ατμοσφαιρική κυκλοφορία στον τροπικό Α. Ειρηνικό Ωκεανό σε σχέση με το παγκόσμιας κλιματικής σημασίας φαινόμενο ΕΝSO (El Niño-Southern Oscillation). Μετά τα τέλη της δεκαετίας του 1970, βρέθηκε ότι η ισχυρότερη αρνητική φάση του Δείκτη της Νότιας Κύμανσης των τελευταίων 150 ετών συμπίπτει με την ισχυρότερη θετική φάση του Δείκτη της Βορειοατλαντικής Κύμανσης (North Atlantic Oscillation Index, ΝΑΟΙ) των τελευταίων 180 ετών. Μάλιστα, η ανάλυση της συσχέτισης μεταξύ των Δεικτών της Νότιας και της Βορειοατλαντικής Κύμανσης την περίοδο 1950-2007 εμφάνισε, αλλά μόνο μετά το 1978, μια στατιστικά σημαντική σχέση ανάμεσα στις 20 ισχυρότερες αρνητικές φάσεις των τριμήνων Σεπτεμβρίου-Νοεμβρίου του Δείκτη της Νότιας Κύμανσης και των χειμερινών (Δεκεμβρίου-Φεβρουαρίου) φάσεων του Δείκτη της Βορειοατλαντικής Κύμανσης που χρονικά ακολούθησαν. Παρόμοια, οι 20 ισχυρότερες θετικές φάσεις των τριμήνων Οκτωβρίου-Δεκεμβρίου και Νοεμβρίου-Ιανουαρίου του Δείκτη της Νότιας Κύμανσης την περίοδο 1950-2007 παρουσίασαν, μετά τα τέλη της δεκαετίας του 1970, μια στατιστικά σημαντική συσχέτιση με τις χειμερινές (Δεκεμβρίου-Φεβρουαρίου) φάσεις του Δείκτη της Βορειοατλαντικής Κύμανσης που ακολούθησαν. Οι σχέσεις αυτές συνέπεσαν με μειωμένη χειμερινή βροχόπτωση στην Α. Μεσόγειο. Εάν οι σχέσεις μεταξύ των Δεικτών της Νότιας και της Βορειοατλαντικής Κύμανσης διατηρηθούν και στο μέλλον, επιτρέπουν, με βάση τη φθινοπωρινή φάση του Δείκτη της Νότιας Κύμανσης, τη δυνατότητα πρόγνωσης της χειμερινής βροχόπτωσης στις περιοχές που επηρεάζονται από τη Βορειοατλαντική Κύμανση (ΝΑΟ) (για περιοχές της Ελλάδας ακόμη και με πιθανότητα 90%). Αναλύθηκε η ετήσια κατά πάχος προσαύξηση του βλαστού σε δένδρα χαλεπίου (Pinus halepensis subsp. halepensis) και τραχείας (Pinus halepensis subsp. brutia) πεύκης από τη θερμο-μεσογειακή ζώνη βλάστησης των νήσων Ζακύνθου, Σκύρου, Σάμου και Κρήτης (Ιεράπετρα). Το πλάτος των αυξητικών δακτυλίων τους βρέθηκε πολύ ευαίσθητο στις μεταβολές της βροχόπτωσης και εμφάνισε στατιστικά σημαντική συσχέτιση με το μέσο όρο της μέσης ετήσιας βροχόπτωσης 37 μετεωρολογικών σταθμών της Α. Μεσογείου. Οι αυξητικοί δακτύλιοι κατέδειξαν, επίσης, ότι η ετήσια βροχόπτωση στην Α. Μεσόγειο μετά το 1970 εμφάνισε τη σημαντικότερη μείωση τουλάχιστον των τελευταίων 200 ετών. Η μείωση της αύξησης των δένδρων συμπίπτει χρονικά με την πρόσφατη αύξηση της θερμοκρασίας του πλανήτη. Συμφωνεί, δε, με την πρόβλεψη, μέσω των κλιματικών μοντέλων, της Διακυβερνητικής Επιτροπής για την Kλιματική Aλλαγή (Intergovernmental Panel on Climate Change, IPCC, 2007) ότι στη Μεσόγειο αναμένεται σημαντική μείωση των βροχοπτώσεων λόγω περαιτέρω αύξησης της θερμοκρασίας του πλανήτη, πράγμα που παρατηρήθηκε στο πρόσφατο παρελθόν. Τίθεται, επίσης, το ερώτημα εάν οι πρόσφατες διασυνδέσεις μεταξύ των Δεικτών της Νότιας και της Βορειοατλαντικής Κύμανσης (που σχετίζονται με την πρόσφατη μείωση της βροχόπτωσης στην Α. Μεσόγειο) συνδέονται και με την αύξηση της θερμοκρασίας του πλανήτη. Κατά τις υγρές περιόδους του 20ου αιώνα (620-760 mm μέση ετήσια βροχόπτωση) η ετήσια αύξηση των δένδρων στις περιοχές μελέτης καθορίστηκε σε μεγάλο βαθμό από βροχοπτώσεις που σημειώθηκαν εντός λίγων εβδομάδων ή μηνών πριν ή και κατά την έναρξη της αυξητικής περιόδου. Όμως, κατά την ξηρότερη περίοδο που καταγράφηκε (1990-96, 480 mm μέση ετήσια βροχόπτωση) η αύξηση εξαρτήθηκε από βροχόπτωση 3-4 ετών πριν, συμπεριλαμβανομένου του έτους σχηματισμού του δακτυλίου. Αυτό δείχνει ότι το νερό από βαθύτερα στρώματα του εδάφους, συσσωρευμένο από βροχοπτώσεις προηγούμενων ετών, κατέστη ιδιαιτέρως σημαντικό καθώς η ξηρασία εντάθηκε. Αυτή η διαδικασία πρέπει να σχετίζεται με το βαθύ ριζικό σύστημα των πεύκων. Όμως, μια σειρά ξηρών ετών μπορεί να εξαντλήσει τα υπόγεια «αποθέματα υγρασίας». Στην περίπτωση αυτή τα πεύκα μπορεί να φτάσουν πολύ κοντά στα όρια της επιβίωσης τους, ακόμη και να ξεραθούν. Τέτοια περιστατικά καταγράφηκαν στη νήσο Σάμο και στην Αχαΐα (Πελοπόννησο), όπου ακόμη και 80-χρονα πεύκα ξεράθηκαν στο τέλος των καλοκαιριών του 2000 και 2007. Με βάση το σενάριο Α1B-SRES (IPCC 2007), η μέση ετήσια βροχόπτωση στην περιοχή μελέτης προβλέπεται να μειωθεί την περίοδο 2090-2099 ακόμη και κατά 30% σε σχέση με τα επίπεδα του 1980-1999, φτάνοντας τα 390 mm. Τα επίπεδα αυτά βροχόπτωσης είναι πολύ χαμηλότερα από την οριακή τιμή των 480 mm που προσδιορίστηκε ως κρίσιμη για τους μελετηθέντες πληθυσμούς πεύκων (Pinus) στο να επιβιώσουν υπό ξηρασία στηριζόμενοι σε βαθύτερα αποθέματα υγρασίας. Συνεπώς, εάν τέτοιες συνθήκες επικρατήσουν κατά τον 21ο αιώνα, θα αυξηθεί σημαντικά ο κίνδυνος καταστροφής των δασών της θερμο-μεσογειακής ζώνης βλάστησης. Επίσης, με τη συνδυασμένη επίδραση αυξημένων θερινών θερμοκρασιών και πυρκαγιών θα μεγαλώσει και ο κίνδυνος ερημοποίησης. / The Southern Oscillation Index (SOI) describes atmospheric circulation in the eastern tropical pacific related to ENSO (El Niño-Southern Oscillation), a phenomenon of global climatic significance. After the late 1970s, SOI’s strongest negative phase in 150 year was found to coincide with the strongest positive phase of the North Atlantic Oscillation Index (NAOI) in 180 years. Correlation analysis between SOI and NAOI during 1950-2007 revealed a statistically significant connection between the 20 strongest negative phases of the Sept.-Nov. SOI and the following winter’s (Dec.-Feb.) NAOI phases only after 1978. Similarly, the 20 strongest positive phases of the Oct.-Dec. and Nov.-Jan. SOI of 1950-2007 produced a statistically significant correlation with the following winter’s (Dec.-Feb.) NAOI phases after the late 1970s. Such relationships coincided with reduced winter precipitation in the eastern Mediterranean. If these SOI-NAOI connections hold, the possibility exists to forecast winter precipitation conditions in regions effected by NAO from the previous autumn’s SOI state (with even a 90% accuracy for regions of Greece). Annual radial stem increment was analysed in Pinus halepensis subsp. halepensis and Pinus halepensis subsp. brutia trees for Thermo-Mediterranean vegetation zones of the Greek islands of Zakinthos, Skiros, Samos and Crete (Ierapetra). Tree-ring width was found to be very sensitive to precipitation and produced a statistically significant correlation with annual rainfall from mean of 37 meteorological stations of the eastern Mediterranean. Tree-rings also indicated that annual rainfall reached its lowest values in nearly 200 years after the 1970s. This reduction in growth coincides with recent global warming. Thus, it is in line with IPCC (2007) climate model projections’ that the Mediterranean will experience a significant decline in precipitation as global warming progresses, as was the case in the recent past. It also raises the question whether recent SOI-NAOI links (involved in the recent decline in precipitation in the eastern Mediterranean) are also connected to global warming. During moist periods of the 20th century (ca. 620-720 mm average annual precipitation) annual tree growth in the regions under investigation was largely controlled by rainfall during a few weeks or months before or during the beginning of the growing season. In contrast, during the driest period on record (1990-1996; 480 mm average annual precipitation) growth depended on rainfall of 3-4 years before, including the year of tree ring formation. This suggests that water from deeper ground, accumulated during rainfall of previous years and became increasingly important as drought intensified. Deep rooting must be involved in such a process. However, a series of dry years may exhaust deeper ground “moisture reserves”. In this case pines may be pushed very close to their survival limits and can even be desiccated. Such incidents were recorded in Samos and Achaia (Peloponnesus) of Greece where pines died in late summer 2000 and 2007, including some 80-year-old trees. Mean annual precipitation for the studied area in 2090-2099 is projected to decrease by even 30% compared to 1980-99 levels, based on Α1B-SRES (IPCC 2007), reaching 390 mm. These levels of rainfall are far bellow the threshold of 480 mm determined as critical for the investigated populations of Pinus to survive drought by relying on deeper moisture reserves. Thus, if such conditions persist during the 21st century they will contribute to the risk of devastation for Thermo-Mediterranean zone forests. Combined with higher summer temperatures and fire outbreaks the risk of desertification will also increase.