Abstract.A physically based hydroclimatological model (AMUNDSEN) is used to assess future climate change impacts on the cryosphere and hydrology of the Ötztal Alps (Austria) until 2100. The model is run in 100 m spatial and 3 h temporal resolution using in total 31 downscaled, bias-corrected, and temporally disaggregated EURO-CORDEX climate projections for the representative concentration pathways (RCPs) 2.6, 4.5, and 8.5 scenarios as forcing data, making this -to date -the most detailed study for this region in terms of process representation and range of considered climate projections. Changes in snow coverage, glacierization, and hydrological regimes are discussed both for a larger area encompassing the Ötztal Alps (1850 km 2 , 862-3770 m a.s.l.) as well as for seven catchments in the area with varying size (11-165 km 2 ) and glacierization (24-77 %).Results show generally declining snow amounts with moderate decreases (0-20 % depending on the emission scenario) of mean annual snow water equivalent in high elevations (> 2500 m a.s.l.) until the end of the century. The largest decreases, amounting to up to 25-80 %, are projected to occur in elevations below 1500 m a.s.l. Glaciers in the region will continue to retreat strongly, leaving only 4-20 % of the initial (as of 2006) ice volume left by 2100. Total and summer (JJA) runoff will change little during the early 21st century with simulated decreases (compared to 1997-2006) of up to 11 % (total) and 13 % (summer) depending on catchment and scenario, whereas runoff volumes decrease by up to 39 % (total) and 47 % (summer) towards the end of the century (2071-2100), accompanied by a shift in peak flows from July towards June.

Abstract. The ongoing glacier shrinkage in the Alps requires frequent updates of glacier outlines to provide an accurate database for monitoring, modelling purposes (e.g. determination of run-off, mass balance, or future glacier extent), and other applications. With the launch of the first Sentinel-2 (S2) satellite in 2015, it became possible to create a consistent, Alpine-wide glacier inventory with an unprecedented spatial resolution of 10 m. The first S2 images from August 2015 already provided excellent mapping conditions for most glacierized regions in the Alps and were used as a base for the compilation of a new Alpine-wide glacier inventory in a collaborative team effort. In all countries, glacier outlines from the latest national inventories have been used as a guide to compile an update consistent with the respective previous interpretation. The automated mapping of clean glacier ice was straightforward using the band ratio method, but the numerous debris-covered glaciers required intense manual editing. Cloud cover over many glaciers in Italy required also including S2 scenes from 2016. The outline uncertainty was determined with digitizing of 14 glaciers several times by all participants. Topographic information for all glaciers was obtained from the ALOS AW3D30 digital elevation model (DEM). Overall, we derived a total glacier area of 1806±60 km2 when considering 4395 glaciers >0.01 km2. This is 14 % (−1.2 % a−1) less than the 2100 km2 derived from Landsat in 2003 and indicates an unabated continuation of glacier shrinkage in the Alps since the mid-1980s. It is a lower-bound estimate, as due to the higher spatial resolution of S2 many small glaciers were additionally mapped or increased in size compared to 2003. Median elevations peak around 3000 m a.s.l., with a high variability that depends on location and aspect. The uncertainty assessment revealed locally strong differences in interpretation of debris-covered glaciers, resulting in limitations for change assessment when using glacier extents digitized by different analysts. The inventory is available at https://doi.org/10.1594/PANGAEA.909133 (Paul et al., 2019).

Instrumental time series are often affected by inhomogeneities which can mask or amplify climate change signals. Various procedures for the detection and adjustment of breaks exist for monthly and annual time series. Homogenization methods on a daily basis are scarce and often disregard uncertainties accompanying the break adjustment. We present a complete homogenization procedure for daily extreme temperature series based on the break detection method PRODIGE and SPLIDHOM for break correction. Both parts of the homogenization rely on the existence of highly correlated reference stations. After the statistical comparison with neighbouring stations, detected breaks are verified and further localized by metadata. Uncertainties of the break adjustments are estimated by altering reference stations and by applying a bootstrapping technique providing an objective indication about the reliability of the homogenization.The method was tested and applied to 71 time series of daily minimum (TN) and maximum temperatures (TX) in Austria covering the period 1948-2009. For some series homogenization was not possible due to large uncertainties in the adjustments or a lack of suitable reference series. In the remaining 57 TN and 54 TX series a total number of 139 breaks were detected. Seventy-five percent of those breaks are documented in the metadata archive, with most of them being caused by station relocations and instrumentation changes. In general, the mean over the temperature dependent adjustments of all stations show a temperature reduction. However, the majority of breaks have mean amplitudes of less than 0.5°C. A comprehensive analysis was performed on the new homogenized daily dataset, showing a widespread warming trend in both TN and TX series. The warming trend is in general amplified due to the homogenization. However, significant changes in the trend are only observed at very few stations. In autumn, however, the trend is reversed in many temperature based 'climate change detection indices'.

Abstract. The on-going glacier shrinkage in the Alps requires frequent updates of glacier outlines to provide an accurate database for monitoring or modeling purposes (e.g. determination of run-off, mass balance, or future glacier extent) and other applications. With the launch of the first Sentinel-2 (S2) satellite in 2015, it became possible to create a consistent, Alpine-wide glacier inventory with an unprecedented spatial resolution of 10 m. Fortunately, already the first S2 images acquired in August 2015 provided excellent mapping conditions for most of the glacierised regions in the Alps. We have used this opportunity to compile a new Alpine-wide glacier inventory in a collaborative team effort. In all countries, glacier outlines from the latest national inventories have been used as a guide to compile a consistent update. However, cloud cover over many glaciers in Italy required including also S2 scenes from 2016. Whereas the automated mapping of clean glacier ice was straightforward using the band ratio method, the numerous debris-covered glaciers required in-tense manual editing. The uncertainty in the outlines was determined with a multiple digitising experiment of 14 glaciers by all participants. Topographic information for all glaciers was derived from the ALOS AW3D30 DEM. Overall, we derived a total glacier area of 1806 ± 60 km2 when considering 4394 glaciers > 0.01 km2. This is 14 % (−1.2 %/a) less than the 2100 km2 derived from Landsat scenes acquired in 2003 and indicating an unabated continuation of glacier shrinkage in the Alps since the mid-1980s. Due to the higher spatial resolution of S2 many small glaciers were additionally mapped in the new inventory or increased in size compared to 2003. An artificial reduction to the former extents would thus result in an even higher overall area loss. Still, the uncertainty assessment revealed locally considerable differences in interpretation of debris-covered glaciers, resulting in limitations for change assessment when using glacier extents digitised by different analysts. The inventory is available at: https://doi.pangaea.de/10.1594/PANGAEA.909133 (Paul et al., 2019).

No abstract

Three homogenization methods (ACMANT, MASH and HOMOP) have been evaluated for their efficiency in homogenizing daily relative humidity data. A homogeneous surrogate data set based on Austrian stations was created and perturbed to simulate inhomogeneous, realistic time series ("validation data sets"). Two validation data sets ("simple" and "complex") were created. In both data sets the magnitude of the breaks depends on the time of year and the measured values. They differ in the number of missing values and especially on whether the break signal was perturbed by white noise. In the latter case, the noise also changed to take into account changes in random measurements errors and other physical factors. The evaluation showed high agreement in statistical characteristics between the real data and the surrogate data set. The homogenization methods were compared in their ability both to detect breaks and to reproduce the homogeneous surrogate data set. For the evaluation of the final data set the distribution, trends and root-mean-square error (RMSE) were analysed. The percentage of improved time series depends on the evaluation parameter considered. Less stations were improved when using the "complex" validation data set. Because of the large number of breaks and the small signal-to-noise ratio, an improvement of the data by homogenization was non-ideal for all methods used, with each having its advantages and disadvantages. The quality of the ACMANT and HOMOP methods is comparable, with ACMANT solving less stations but declaring less stations falsely as homogeneous. To get an impression of the influence on real data, ACMANT was applied to homogenize daily Austrian time series of relative humidity. While the quality of data from some stations can be improved through the homogenization, this is not the case for all time series. A final evaluation of homogenized time series should be performed to ensure their quality before further use.

Abstract.A physically based hydroclimatological model (AMUNDSEN) is used to assess future climate change impacts on the cryosphere and hydrology of the Ötztal Alps (Austria) until 2100. The model is run in 100 m spatial and 3 h temporal resolution using in total 31 downscaled, bias-corrected, and temporally disaggregated EURO-CORDEX climate projections for the RCP2.6, RCP4.5, and RCP8.5 scenarios as forcing data. Changes in snow coverage, glacierization, and hydrological regimes are discussed both for a larger area encompassing the Ötztal Alps (1850 km (compared to 1997-2006) of up to 11 % (total) and 13 % (summer) depending on catchment and scenario, whereas runoff volumes decrease by up to 39 % (total) and 47 % (summer) towards the end of the century (2071-2100), accompanied by a shift in peak flows from July towards June.

This study explores the potential of different predictor strategies for improving the performance of regression-based downscaling approaches. The investigated local-scale target variables are precipitation, air temperature, wind speed, relative humidity, and global radiation, all at a daily time scale. Observations of these target variables are assessed from three sites in close proximity to mountain glaciers: 1) the Vernagtbach station in the European Alps, 2) the Artesonraju measuring site in the tropical South American Andes, and 3) the Mount Brewster measuring site in the Southern Alps of New Zealand. The large-scale dataset being evaluated is the ERA-Interim dataset. In the downscaling procedure, particular emphasis is put on developing efficient yet not overfit models from the limited information in the temporally short (typically a few years) observational records of the high mountain sites. For direct (univariate) predictors, optimum scale analysis turns out to be a powerful means to improve the forecast skill without the need to increase the downscaling model complexity. Yet the traditional (multivariate) predictor sets show generally higher skill than the direct predictors for all variables, sites, and days of the year. Only in the case of large sampling uncertainty (identified here to particularly affect observed precipitation) is the use of univariate predictor options justified. Overall, the authors find a range in forecast skill among the different predictor options applied in the literature up to 0.5 (where 0 indicates no skill, and 1 represents perfect skill). This highlights that a sophisticated predictor selection (as presented in this study) is essential in the development of realistic, local-scale scenarios by means of downscaling.