Glacial melting in the Tibetan Plateau affects the water resources of millions of people. This study finds that—partly owing to changes in atmospheric circulations and precipitation patterns—the most intensive glacier shrinkage is in the Himalayan region, whereas glacial retreat in the Pamir Plateau region is less apparent.

/pdf/different-glacier-status-with-atmospheric-circulations-in-9xtro65rsi.pdf

Different glacier status with atmospheric circulations in Tibetan Plateau and surroundings

Himalayan glaciers are a focus of public and scientific debate. Prevailing uncertainties are of major concern because some projections of their future have serious implications for water resources. Most Himalayan glaciers are losing mass at rates similar to glaciers elsewhere, except for emerging indications of stability or mass gain in the Karakoram. A poor understanding of the processes affecting them, combined with the diversity of climatic conditions and the extremes of topographical relief within the region, makes projections speculative. Nevertheless, it is unlikely that dramatic changes in total runoff will occur soon, although continuing shrinkage outside the Karakoram will increase the seasonality of runoff, affect irrigation and hydropower, and alter hazards.

https://www.zora.uzh.ch/id/eprint/72075/1/2012_BolchT_BolchAl12_science_GlacHimal_supplm_subm_Kopie_.pdf

The state and fate of Himalayan glaciers

Glaciers are among the best terrestrial climate indicators, an important water resource in mountains1,2 and a major contributor to global sea level rise3,4. In the Hindu Kush - Karakoram - Himalaya region (HKKH), a paucity of appropriate glacier data has prevented a comprehensive assessment of current regional mass balance5. However, there are indirect evidences of a complex pattern of glacial responses5-8 in reaction to heterogeneous climate change signals9. Here, we provide the first coherent data set of detailed glacier thickness changes over the HKKH during 2003-2009 by combining satellite laser altimetry and a global elevation model. In the eastern, central and south-western parts of the HKKH, glacier wastage is widespread with regional thinning rates up to 0.66 ± 0.09 m a-1 in the Jammu-Kashmir region. Conversely, in the Karakoram, glaciers are close to balance with only a slight thinning of 0.07 ± 0.04 m a-1. Regionally averaged thinning rates under debris-mantled ice are similar to those of clean ice despite insulation by debris covers. The 2003-2008 specific mass balance for our HKKH study region is -0.21 ± 0.05 m a-1 water equivalent (WE), significantly less negative than the global average of ~ -0.7 m a-1 WE for glaciers and ice caps4,10. This difference is mainly an effect of the balanced glacier mass budget in the Karakoram. The corresponding HKKH sea level contribution is +0.035 ± 0.009 mm a-1 amounting to 1% of the present-day sea level rise11. Our 2003-2008 mass budget of -12.8 ± 3.5 Gt a-1 is more negative than recent satellite gravimetry based estimates of -5 ± 6 Gt a-1 over 2003-2010 (ref. 12). For the mountain catchments of the Indus and Ganges basins13, the glacier imbalance contributes ~3.5% and ~2.0%, respectively, to the annual average river discharge13, and up to ~10% for the Upper Indus basin14.

Contrasting patterns of early twenty-first-century glacier mass change in the Himalayas

Riv ers originating in the high mountains of Asia are among the most meltwater-dependent river systems on Earth, yet large human populations depend on their resources downstream 1 . Across High Asia’s river basins, there is large variation in the contribution of glacier and snow melt to total runo 2 , which is poorlyquantified.Thelackofunderstandingofthehydrological regimes of High Asia’s rivers is one of the main sources of uncertainty in assessing the regional hydrological impacts of climate change 3 . Here we use a large-scale, high-resolution cryospheric‐hydrological model to quantify the upstream hydrological regimes of the Indus, Ganges, Brahmaputra, Salween and Mekong rivers. Subsequently, we analyse the impacts of climate change on future water availability in these basins using the latest climate model ensemble. Despite large dierences in runo composition and regimes between basins and between tributaries within basins, we project an increase in runo at least until 2050 caused primarily by an increase in precipitation in the upper Ganges, Brahmaputra, Salween and Mekong basins and from accelerated melt in the upper Indus Basin. These findings have immediate consequences for climatechangepolicieswhereatransitiontowardscopingwith intra-annual shifts in water availability is desirable. In general, the climate in the eastern part of the Himalayas is characterized by the East-Asian and Indian monsoon systems, causing the bulk of precipitation to occur during JuneSeptember (Supplementary Fig. 4). The precipitation intensity shows a strong northsouth gradient caused by orographic eects 4 . Precipitation patterns in the Hindu Kush and Karakoram ranges in the west are characterized by westerly and southwesterly flows, causing the precipitation to fall more equally distributed over the year 5 (Supplementary Fig. 4). In the Karakoram, up to two-thirds of the annual high-altitude precipitation occurs during the winter months 6,7 . In addition, basin hypsometry determines the ratio of solid and liquid precipitation within a basin. Solid precipitation can be stored long-term as perennial snow, and ice or short-term as seasonal snow before turning into runo by melting, whereas liquid

/pdf/consistent-increase-in-high-asia-s-runoff-due-to-increasing-1ld0ks8341.pdf

Consistent increase in High Asia's runoff due to increasing glacier melt and precipitation

Although reliable figures are often missing, considerable detrimental changes due to shrinking glaciers are universally expected for water availability in river systems under the influence of ongoing global climate change. We estimate the contribution potential of seasonally delayed glacier melt water to total water availability in large river systems. We find that the seasonally delayed glacier contribution is largest where rivers enter seasonally arid regions and negligible in the lowlands of river basins governed by monsoon climates. By comparing monthly glacier melt contributions with population densities in different altitude bands within each river basin, we demonstrate that strong human dependence on glacier melt is not collocated with highest population densities in most basins.

https://www.pnas.org/content/pnas/107/47/20223.full.pdf

Contribution potential of glaciers to water availability in different climate regimes

. The El Nino/Southern Oscillation (ENSO) is a major driver of climate variability in the tropical Andes, where recent Nino and Nina events left an observable footprint on glacier mass balance. The nature and strength of the relationship between ENSO and glacier mass balance, however, varies between regions and time periods, leaving several unanswered questions about its exact mechanisms. The starting point of this study is a 4-year long time series of distributed surface energy and mass balance (SEB/SMB) calculated using a process-based model driven by observations at Shallap Glacier (Cordillera Blanca, Peru). These data are used to calibrate a regression-based downscaling model that links the local SEB/SMB fluxes to atmospheric reanalysis variables on a monthly basis, allowing an unprecedented quantification of the ENSO influence on the SEB/SMB at climatological time scales (1980–2013, ERA-Interim period). We find a stronger and steadier anti-correlation between Pacific sea-surface temperature (SST) and glacier mass balance than previously reported. This relationship is most pronounced during the wet season (December–May) and at low altitudes where Nino (Nina) events are accompanied with a snowfall deficit (excess) and a higher (lower) radiation energy input. We detect a weaker but significant ENSO anti-correlation with total precipitation (Nino dry signal) and positive correlation with the sensible heat flux, but find no ENSO influence on sublimation. Sensitivity analyses comparing several downscaling methods and reanalysis data sets resulted in stable mass balance correlations with Pacific SST but also revealed large uncertainties in computing the mass balance trend of the last decades. The newly introduced open-source downscaling tool can be applied easily to other glaciers in the tropics, opening new research possibilities on even longer time scales.

/pdf/enso-influence-on-surface-energy-and-mass-balance-at-shallap-4baofxuqm5.pdf

ENSO influence on surface energy and mass balance at Shallap Glacier, Cordillera Blanca, Peru

A study shows that regional atmospheric change driven by land-cover change contributes little to glacier mass loss on Tanzania’s Mount Kilimanjaro. More generally, this finding suggests that local land-cover change may have limited impact on mountain glaciers in the tropics and elsewhere, compared with that of global climate change. Global climate change is primarily linked to changes in greenhouse gases, but land-cover change (LCC) has increasingly been recognized as another forcing on the regional scale1,2. The related effects on alpine glaciers are, however, not yet known. Here we present the first quantification of the contribution of LCC-driven atmospheric change to glacier mass loss, illustrated by the well-studied case of Kilimanjaro in tropical Africa3,4,5. We employ a novel multi-scale modelling approach6, which links atmospheric dynamics and local glacier mass balance in a fully physical way and is validated by in situ measurements. Using different model settings, this shows that local LCC since the 1970s has contributed 7±6% (17±12%) to mass loss of a southern slope glacier in the dry (wet) season, but this effect could reverse in the other mountain sectors and also decrease glacier mass loss. Thus, for the moment, the hypothesis that local LCC is another forcing of glacier loss on Kilimanjaro7,8 cannot be corroborated. More generally, our results indicate that the impact of local LCC on mountain glaciers is constrained by regional circulation (moisture trajectories), altitude (distance to forest), and outside the tropics by precipitation mechanisms (frontal systems). We therefore argue that attribution of glacier change and variability to large-scale climate dynamics3,9,10 is unlikely to be distorted by local LCC.

Limited forcing of glacier loss through land-cover change on Kilimanjaro

Deep neural networks (DNNs) have been recently proposed for quartet tree phylogeny estimation. Here, we present a study evaluating recently trained DNNs in comparison to a collection of standard phylogeny estimation methods on a heterogeneous collection of datasets simulated under the same models that were used to train the DNNs, and also under similar conditions but with higher rates of evolution. Our study shows that using DNNs with quartet amalgamation is less accurate than several standard phylogeny estimation methods we explore (e.g., maximum likelihood and maximum parsimony). We further find that simple standard phylogeny estimation methods match or improve on DNNs for quartet accuracy, especially, but not exclusively, when used in a global manner (i.e., the tree on the full dataset is computed and then the induced quartet trees are extracted from the full tree). Thus, our study provides evidence that a major challenge impacting the utility of current DNNs for phylogeny estimation is their restriction to estimating quartet trees that must subsequently be combined into a tree on the full dataset. In contrast, global methods (i.e., those that estimate trees from the full set of sequences) are able to benefit from taxon sampling, and hence have higher accuracy on large datasets.

Re-evaluating Deep Neural Networks for Phylogeny Estimation: The Issue of Taxon Sampling

Motivation Predicting gene expression from DNA is an open field of research. As in many areas, labeled data is dwarfed by unlabelled data, i.e. species with a sequenced genome but no gene expression assay data. Pretraining on unlabelled data using masked language modeling has proven highly successful in overcoming data constraints in natural language and proteomics. However, in genomics, this approach has so far been applied only to single genomes, neither leveraging conservation of regulatory sequences across species nor the vast amount of available genomes. Results Here we train a masked language model on more than 800 species spanning over 500 million years of evolution. We show that explicitly modeling species is instrumental in capturing conserved yet evolving regulatory elements and in controlling for oligomer biases. We extract embeddings for 3’ untranslated regions of Saccharomyces cerevisiae and Schizosaccharomyces pombe and use them to achieve prediction of mRNA half-life that is better or on-par with the state-of-the-art, demonstrating the utility of the approach for regulatory genomics. Moreover, we show that the per-base reconstruction probability of our model significantly predicts RNA-binding protein bound sites directly. Altogether, our work establishes a self-supervised framework to leverage large genome collections of evolutionary distant species for regulatory genomics and contributes to alignment-free comparative genomics. Availability and implementation The source code and trained models are available at: https://github.com/DennisGankin/species-aware-DNA-LM

Martin Großhauser

Papers

Contribution potential of glaciers to water availability in different climate regimes

ENSO influence on surface energy and mass balance at Shallap Glacier, Cordillera Blanca, Peru

Limited forcing of glacier loss through land-cover change on Kilimanjaro

Re-evaluating Deep Neural Networks for Phylogeny Estimation: The Issue of Taxon Sampling

Species-aware DNA language modeling