1gi)26 and glaciers shrinking to higher elevations where precipitation rates are higher as a result of orographic precipitation enhancement27. Deep learning applied to glacier evolution modelling. regularized multilinear regression. Here, we perform the first-ever glacier evolution projections based on deep learning by modelling the 21st century glacier evolution in the French Alps. Park, and S. Beason. Solved Activity 13.3 Nisqually Glacier Response to Climate - Chegg From this behavior, inferences of past climate can be drawn. 6 (2018). On the one hand, MB nonlinearities for mountain glaciers appear to be only relevant for climate scenarios with a reduction in greenhouse gases emissions (Fig. is central to a glacier's response: Fig.2ashows 1L.t/for a warming trend of 1 C per century, for three glaciers with dierent (and fixed ). Ice thickness accuracy varied significantly, with an overall correct representation of the ice distribution but with local biases reaching up to 100%. The performance of this parametrization was validated in a previous study, indicating a correct agreement with observations31. Summer melt was also above average. Verfaillie, D., Dqu, M., Morin, S. & Lafaysse, M. The method ADAMONT v1.0 for statistical adjustment of climate projections applicable to energy balance land surface models. Res. In many aspects, it might be too optimistic, as many ice caps will have a negative impact on MB through thinning, bringing their mean surface elevation to lower altitudes, thus further warming their perceived climate. The anomaly in snowfall was evenly distributed for every month in the accumulation (October 1April 31) and ablation seasons, respectively. & Galiez, C. A deep learning reconstruction of mass balance series for all glaciers in the French Alps: 19672015. ADS Mer de Glace, 29km2 in 2015), which did show important differences under RCP 8.5 (up to 75%), due to their longer response time. Z. et al. (Springer, New York, 2009). A recent study he did found that 80 percent of the glaciers in Alberta and British Columbia could melt in the next 50 years. At this point, it is important to clarify the different ways of treating PDDs in the Lasso and the temperature-index MB models analysed in this study in order to justify analogies. This is well in agreement with the known uncertainties of glacier evolution models, with glacier ice thickness being the second largest uncertainty after the future GCM-RCM-RCP climate members used to force the model29. The Cryosphere 14, 565584 (2020). The two models with linear MB responses to PDDs and accumulation simulate more positive MB rates under RCP 2.6, highlighting their over-sensitivity to negative air temperature anomalies and positive snowfall anomalies (Fig. Grenoble Alpes, CNRS, IRD, G-INP, Institut des Gosciences de lEnvironnement, Grenoble, France, INRAE, UR RiverLy, Lyon-Villeurbanne, France, Institute for Marine and Atmospheric research Utrecht, Utrecht University, Utrecht, Netherlands, Univ. Nature Geosciences, https://doi.org/10.1038/s41561-021-00885-z (2022). Tour. We acknowledge the more than 50 years of glaciological monitoring performed by the GLACIOCLIM French National Observatory (https://glacioclim.osug.fr), which provided essential observations for our modelling study. Nature 577, 364369 (2020). & Funk, M. A comparison of empirical and physically based glacier surface melt models for long-term simulations of glacier response. Slider with three articles shown per slide. Interestingly, future warmer temperatures do not affect annual snowfall rates on glaciers as a result of both higher precipitation rates in the EURO-CORDEX ensemble (Fig. We argue that such models can be suitable for steep mountain glaciers. In order to do so, we applied a deterministic sampling process as a sensitivity analysis to both the deep learning and the Lasso MB models. 1d, g). a Projected mean glacier altitude evolution between 2015 and 2100. The model output data generated in this study have been deposited in netCDF and CSV format in a Zenodo repository under accession code Creative Commons Attribution 4.0 International. Since the neural network used here virtually behaves like a black box, an alternative way is needed to understand the models behaviour. Our projections highlight the almost complete disappearance of all glaciers outside the Mont-Blanc and Pelvoux (Ecrins region) massifs under RCP 4.5 (Fig. Our results also highlight the important role played by glacier geometry adjustment under changing climatic conditions, which is typical of mountain glaciers38. GLAMOS. This synthetic setup allowed us to reproduce the climatic conditions to be undergone by most ice caps, with their mean surface altitude hardly evolving through time. Reanalysis of 47 Years of Climate in the French Alps (19582005): Climatology and Trends for Snow Cover. This translates into more frequent extreme negative MB rates, and therefore greater differences due to nonlinearities for the vast majority of future climate scenarios (Fig. Interestingly, our analysis indicates that more complex models using separate DDFs for ice, firn and snow might introduce stronger biases than more simple models using a single DDF. With this setup, we reproduced the ice cap-like behaviour with a lack of topographical adjustment to higher elevations. Differences for individual glaciers can be much more pronounced, as large and flat glaciers will have topoclimatic configurations that produce more extreme MB rates than small and steep glaciers with a short response time. 3). Farinotti, D., Round, V., Huss, M., Compagno, L. & Zekollari, H. Large hydropower and water-storage potential in future glacier-free basins. 4a, b) and negative (Fig. Under warmer conditions (RCP 8.5), the differences between the linear and nonlinear MB model become smaller, as the topographical feedback from glacier retreat compensates for an important fraction of the losses induced by the late century warmer climate (Fig. These results revealed that the main uncertainties on glacier simulations arise from the initial ice thickness used to initialize the model. Fluctuations of the Nisqually Glacier, Mt. Rainier, Washington, since C.G. When working with spatiotemporal data, it is imperative to respect spatial and temporal data structures during cross-validation in order to correctly assess an accurate model performance48. and JavaScript. Summer climate is computed between April 1st and September 30th and winter climate between October 1st and March 31st. He, K., Zhang, X., Ren, S. & Sun, J. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification. The training was performed with an RMSprop optimizer, batch normalization46, and we used both dropout and Gaussian noise in order to regularize it. 5). GlacierMIP A model intercomparison of global-scale glacier mass-balance models and projections. Since the climate and glacier systems are known to be nonlinear13, we investigate the benefits of using a model treating, among others, PDDs in a nonlinear way in order to simulate annual glacier-wide MB at a regional scale. In the past, shortwave radiation represented a more important fraction in the glacier surface energy budget than the energy fluxes directly related to air temperature (e.g. With this cross-validation we determined a deep learning MB model spatiotemporal (LSYGO) RMSE of 0.59m.w.e. In our model, we specifically computed this parameterized function for each individual glacier larger than 0.5km2, representing 80% of the total glacierized area in 2015, using two DEMs covering the whole French Alps: a photogrammetric one in 1979 and a SPOT-5 one in 2011. PDF Centennial glacier retreat as categorical evidence of regional climate Models were trained using the SAFRAN reanalysis dataset47, including observations of mountain regions in France for the 19582015 period. This translates into a more linear response to air temperature changes compared to the ablation season (Fig. J. Glaciol. Change 120, 2437 (2014). Farinotti, D. et al. (Photograph by Klaus J. Bayr, Keene State College, 1990) One method of measuring glaciers is to send researchers onto the ice with . We performed a validation simulation for the 20032015 period by running our model through this period and comparing the simulated glacier surface area of each of the 32 glaciers with MB to observations from the 2015 glacier inventory16,52. Grenoble Alpes, CNRS, G-INP, Laboratoire Jean Kuntzmann, Grenoble, France, You can also search for this author in Share sensitive information only on official, secure websites.. Glacier response to climate perturbations: an accurate linear geometric Moreover these three aspects of glacier behavior are inextricably interwoven: a high sensitivity to climate change goes hand-in-hand with a large natural variability. However, both the climate and glacier systems are known to react non-linearly, even to pre-processed forcings like PDDs13, implying that these models can only offer a linearized approximation of climate-glacier relationships. Geophys. Multiple copies of this dataset were created, and for each individual copy a single predictor (i.e. MATH Since 2005, study finds that surface melt off glaciers in the North has risen by 900%. Sci. S10). Article Nat Commun 13, 409 (2022). 4e). The application of a non-linear back-propagation neural network to study the mass balance of Grosse Aletschgletscher, Switzerland. acknowledges the funding received from a EU Horizon 2020 Marie Skodowska-Curie Individual Fellowship (grant no. Future projections of glacier-wide MB evolution were performed using climate projections from ADAMONT25. Despite the differences in the two modelling approaches (TableS2), both regional glacier volume projections present relatively similar results by the end of the century, with volume differences ranging between 14% for RCP 2.6 to less than 2% for RCP 4.5 (Fig. Our previous work31 has shown that linear MB models can be correctly calibrated for data around the mean temperature and precipitation values used during training, giving similar results and performance to deep learning. Nonlinear sensitivity of glacier mass balance to future climate change unveiled by deep learning. Smiatek, G., Kunstmann, H. & Senatore, A. EURO-CORDEX regional climate model analysis for the Greater Alpine Region: performance and expected future change: climate change in the gar area. how climate change and glacier retreat are reshaping whole aquatic ecosystems, there is a need to develop an integrated understanding spanning multiple taxonomic groups and trophic levels in glacier-fed rivers (e.g., bacteria, protists, fungi, algae, diatoms, invertebrates, mammals, amphibians, and fish; Clitherow et al. GloGEMflow has been previously applied in a study over the whole European Alps, and its temperature-index model was mainly calibrated with MB data from the Swiss Alps. The Cryosphere 12, 13671386 (2018). Here, we compare our results with those from a recent study that focused on the European Alps10. Overall, this results in linear MB models overestimating both extreme positive (Fig. By 2100, under RCP 4.5, these two high-altitude massifs are predicted to retain on average 26% and 13% of their 2015 volume, respectively, with most of the ice concentrated in a few larger glaciers (>1km2, Fig. This parametrization reproduces in an empirical manner the changes in glacier geometry due to the combined effects of ice dynamics and MB.
Elijah Craig Barrel Proof Releases,
How To Change Address On Concealed Carry Permit Va,
What Does A Temporary License Look Like,
Empire At War Expanded: Fall Of The Republic Wiki,
South St Paul Apartments,
Articles N