ABSTRACT: This paper is a review, concerning the observed climatic changes in the precipitation regime of Greece, during the period 1956-2002, based on daily precipitation totals from 26 meteorological stations, assigned by the Hellenic National Meteorological Service. The annual trends of the precipitation time series were estimated using the Mann-Kendall rank statistic method. Besides, the probability of the extreme events was determined by the scale and shape parameters of the fitted Gamma distribution to the daily precipitation time series. The findings extracted by the analysis showed that, decreasing trends, mostly statistically significant at 95% confidence level, regarding the total annual precipitation and the annual number of rain days appear all over Greece. The fitted Gamma distributions to the precipitation datasets within 10-year sub periods indicate that the scale parameter increases for the western and southern-eastern sub regions, while the annual precipitation presents a downward trend, especially during the last decade 1991-2000. The increase in the variance and the scale parameter, as well as the shift of the mean towards higher values reveal the incidence of extreme daily precipitation values since 1980s. Key-words: Precipitation variability, Gamma distribution, extremes, climate change, Greece. ΠΕΡΙΛΗΨΗ: Αυτή η εργασία είναι µια επισκόπηση, η οποία αφορά τις παρατηρούµενες κλιµατικές µεταβολές στο βροχοµετρικό καθεστώς της Ελλάδος, κατά την διάρκεια της περιόδου 1956-2002, βασισµένη σε ηµερήσια δεδοµένα βροχής από 26 µετεωρολογικούς σταθµούς, που παραχωρήθησαν από την Εθνική Μετεωρολογική Υπηρεσία. Οι ετήσιες τάσεις των χρονοσειρών της βροχής εκτιµήθηκαν εφαρµόζοντας την στατιστική µέθοδο Mann-Kendall. Επίσης, η πιθανότητα των ακραίων γεγονότων καθορίστηκε από τις παραµέτρους (scale και shape) της κατανοµής Γάµµα που προσαρµόστηκε στις χρονοσειρές των ηµερήσιων δεδοµένων. Τα εξαγόµενα συµπεράσµατα από την ανάλυση έδειξαν ότι, σε όλη την Ελλάδα εµφανίζονται τάσεις µείωσης (στατιστικά σηµαντικές σε επίπεδο εµπιστοσύνης 95%) στις χρονοσειρές της ετήσιας βροχής καθώς επίσης και στον ετήσιο αριθµό ηµερών βροχής. Οι προσαρµοσµένες κατανοµές Γάµµα στα ηµερήσια βροχοµετρικά δεδοµένα 10-ετών υποπεριόδων έδειξαν ότι η παράµετρος “scale” αυξάνει στις δυτικές και νοτιο-ανατολικές περιοχές της Ελλάδος, ενώ η ετήσια βροχόπτωση παρουσιάζει τάση ελάττωσης, ιδιαίτερα την τελευταία δεκαετία 1991-2000. Η παρατηρούµενη αύξηση της µεταβλητότητας της βροχής και της παραµέτρου “scale” της κατανοµής Γάµµα καθώς επίσης και η µετατόπιση της µέσης τιµής της ηµερήσιας βροχής προς υψηλότερες τιµές αποκαλύπτουν την εµφάνιση ακραίων ηµερήσιων ποσών βροχής από τη δεκαετία του 1980. Λέξεις-κλειδιά: Μεταβλητότητα βροχής, κατανοµή Γάµµα, ακραίες τιµές, κλιµατική µεταβολή, Ελλάδα.

abstract Climatic conditions and hence climate change influence agriculture. Most studies that addressed the vulnerability of agriculture to climate change have focused on potential impacts without considering adaptation. When adaptation strategies are considered, socio-economic conditions and farm management are often ignored, but these strongly influence current farm performance and are likely to also influence adaptation to future changes. This study analysed the adaptation of farmers and regions in the European Union to prevailing climatic conditions, climate change and climate variability in the last decades (1990–2003) in the context of other conditions and changes. We compared (1) responses in crop yields with responses in farmers’ income, (2) responses to spatial climate variability with responses to temporal climate variability, (3) farm level responses with regional level responses and (4) potential climate impacts (based on crop models) with actual climate impacts (based on farm accountancy data). Results indicated that impacts on crop yields cannot directly be translated to impacts on farmers’ income, as farmers adapt by changing crop rotations and inputs. Secondly, the impacts of climatic conditions on spatial variability in crop yields and farmers’ income, with generally lower yields in warmer climates, is different from the impacts of temporal variability in climate, for which more heterogeneous patterns are observed across regions in Europe. Thirdly, actual impacts of climate change and variability are largely dependent on farm characteristics (e.g. intensity, size, land use), which influence management and adaptation. To accurately understand impacts and adaptation, assessments should consider responses at different levels of organization. As different farm types adapt differently, a larger diversity in farm types reduces impacts of climate variability at regional level, but certain farm types may still be vulnerable. Lastly, we observed that management and adaptation can largely reduce the potential impacts of climate change and climate variability on crop yields and farmers’ income. We conclude that for reliable projections of the impacts of climate change on agriculture, adaptation should not be seen anymore as a last step in a vulnerability assessment, but as integrated part of the models used to simulate crop yields, farmers’ income and other indicators related to agricultural performance.