Assessments of Data-Driven Deep Learning Models on One-Month Predictions of Pan-Arctic Sea Ice Thickness

In recent years,deep learning methods have gradually been applied to prediction tasks related to Arctic sea ice concentration,but relatively little research has been conducted for larger spatial and temporal scales,mainly due to the limited time coverage of observations and reanalysis data.Meanwhile,deep learning predictions of sea ice thickness(SIT)have yet to receive ample attention.In this study,two data-driven deep learning(DL)models are built based on the ConvLSTM and fully convolutional U-net(FC-Unet)algorithms and trained using CMIP6 historical simulations for transfer learning and fine-tuned using reanalysis/observations.These models enable monthly predictions of Arctic SIT without considering the complex physical processes involved.Through comprehensive assessments of prediction skills by season and region,the results suggest that using a broader set of CMIP6 data for transfer learning,as well as incorporating multiple climate variables as predictors,contribute to better prediction results,although both DL models can effectively predict the spatiotemporal features of SIT anomalies.Regarding the predicted SIT anomalies of the FC-Unet model,the spatial correlations with reanalysis reach an average level of 89%over all months,while the temporal anomaly correlation coefficients are close to unity in most cases.The models also demonstrate robust performances in predicting SIT and SIE during extreme events.The effectiveness and reliability of the proposed deep transfer learning models in predicting Arctic SIT can facilitate more accurate pan-Arctic predictions,aiding climate change research and real-time business applications.

Deep Learning Shows Promise for Seasonal Prediction of Antarctic Sea Ice in a Rapid Decline Scenario

The rapidly changing Antarctic sea ice has garnered significant interest. To enhance the prediction skill for sea ice and respond to the Sea Ice Prediction Network-South's latest call, this study presents the reforecast results of Antarctic sea-ice area and extent from December to June of the coming year with a Convolutional Long Short-Term Memory(Conv LSTM)Network. The reforecast experiments demonstrate that Conv LSTM captures the interannual and interseasonal variability of Antarctic sea ice successfully, and performs better than the European Centre for Medium-Range Weather Forecasts. Based on this, we present the prediction from December 2023 to June 2024, indicating that the Antarctic sea ice will remain at lows, but may not create a new record low. This research highlights the promising application of deep learning in Antarctic sea-ice prediction.

Engineering behaviour of in situ cored deep cement mixed marine deposits subjected to undrained and drained shearing

The deep cement mixing(DCM)is used to improve the capacity and reduce the settlement of the soft ground by forming cemented clay columns.The investigation on the mechanical behaviour of the DCM samples is limited to either laboratory-prepared samples or in-situ samples under unconfined compression.In this study,a series of drained and undrained triaxial shearing tests was performed on the in-situ cored DCM samples with high cement content to assess their mechanical behaviours.It is found that the drainage condition affects significantly the stiffness,peak and residual strengths of the DCM samples,which is mainly due to the state of excess pore water pressure at different strain levels,i.e.being positive before the peak deviatoric stress and negative after the peak deviatoric stress,in the undrained tests.The slope of the failure envelope changes obviously with the confining pressures,being steeper at lower stress levels and flatter at higher stress levels.The strength parameters,effective cohesion and friction angle obtained from lower stress levels(c′0 andφ′0)are 400 kPa and 58°,respectively,which are deemed to be true for design in most DCM applications where the in-situ stress levels are normally at lower values of 50-200 kPa.Additionally,the computed tomography(CT)scanning system was adopted to visualize the internal structures of DCM samples.It is found that the clay pockets existing inside the DCM samples due to uneven mixing affect markedly their stress-strain behaviour,which is one of the main reasons for the high variability of the DCM samples.

Optimizing the LSTM Deep Learning Model for Arctic Sea Ice Melting Prediction

The National Oceanic and Atmospheric Administration reports a 95% decline in the oldest Arctic ice over the last 33 years [1], while the National Aeronautics and Space Administration states that summer Arctic Sea Ice Extent (SIE) is shrinking by 12.2% per decade since 1979 due to warmer temperatures [2]. Given the rapidly changing Arctic conditions, accurate prediction models are crucial. Deep learning models developed for Arctic forecasts primarily focus on exploring convolutional neural networks (CNN) and convolutional Long Short-Term Memory (LSTM) networks, while the exploration of the power of LSTM networks is limited. In this research, we focus on enhancing the performance of an LSTM network for predicting monthly Arctic SIE. We leverage five climate and atmospheric variables, validated for their correlation with SIE in prior studies [3]. We utilize the Spearman’s rank correlation and ExtraTrees regressor to enhance our understanding of the importance of the five variables in predicting SIE. We further enhance our predictor variables with seasonal information, lagged time steps, and a linear regression simulated SIE that accounts for the influence of past SIE on current SIE. Statistical methods guide our selection of data scalers and best evaluation metrics for our model. By experimenting with hyperparameter optimization and advanced deep learning training techniques, such as batch sizes, number of neurons, early stopping, and model checkpoint, our model achieved a Mean Absolute Error (MAE) of 0.191 and R2 of 0.996, underscoring its ability to account for nearly all the variance in our data and holds great promise for the prediction of SIE.

Dome Argus: Ideal site for deep ice drilling

Located on the centre of ice drainage range, the highest Dome Argus （Dome A） of East Antarctic Ice Sheet （EAIS）, could be represented as an ideal site for deep ice cores drilling containing oldest paleo-climate records. To select a suitable drilling site for deep ice core, it needs gather all information pertaining to the local meteorology, ice sheet landforms, ice thickness, subgla- cial topography of bed rocks, ice velocity, internal structures of ice sheet, etc. Based on the International Partnerships in Ice Core Sciences （IPICS）, we present recent achievement of glaciological research and its perspective at Dome A in this paper. We system- atically discussed the merits and possible ventures of potential drilling sites around Dome A. Among all the candidates, we find that the Chinese Antarctic Kunlun Station is the best site for and assess further the possibility to obtain a replicate core for carrying out the first deep ice core drilling campaign. We emphasize studying dynamics and evolution of climate change.

Tibetan Plateau Geladaindong black carbon ice core record(1843-1982):Recent increases due to higher emissions and lower snow accumulation

Black carbon(BC) deposited on snow and glacier surfaces can reduce albedo and lead to accelerated melt. An ice core recovered from Guoqu glacier on Mt. Geladaindong and analyzed using a Single Particle Soot Photometer(SP2) provides the ?rst long-term(1843-1982) record of BC from the central Tibetan Plateau. Post 1940 the record is characterized by an increased occurrence of years with above average BC, and the highest BC values of the record. The BC increase in recent decades is likely caused by a combination of increased emissions from regional BC sources, and a reduction in snow accumulation. Guoqu glacier has received no net ice accumulation since the 1980 s, and is a potential example of a glacier where an increase in the equilibrium line altitude is exposing buried high impurity layers. That BC concentrations in the uppermost layers of the Geladaindong ice core are not substantially higher relative to deeper in the ice core suggests that some of the BC that must have been deposited on Guoqu glacier via wet or dry deposition between 1983 and 2005 has been removed from the surface of the glacier, potentially via supraglacial or englacial meltwater.

Variation of culturable bacteria along depth in the East Rongbuk ice core,Mt.Everest

Ice melt water from a 22.27 m ice core which was drilled from the East Rongbuk Glacier, Mt. Everest was incubation in two incubation ways： plate melt water directly and enrichment melt water prior plate, respectively. The abundance of cultivable bacteria ranged from 0-295 CFU mL-I to 0--1720 CFU mL-1 in two incubations with a total of 1385 isolates obtained. Comparing to direct cultivation, enrichment cultivation recovered more bacteria. Pigment-producing bacteria accounted for an average of 84.9% of total isolates. Such high percentage suggested that pigment production may be an adaptive physiological feature for the bacteria in ice core to cope with strong ultraviolet radiation on the glacier. The abundances of cultivable bacteria and pigment-producing isolates varied synchronously along depth： higher abundance in the middle and lower at the top and bottom. It indicated that the middle part of the ice core was hospitable for the microbial survival. Based on the physiological properties of the colonies, eighty-nine isolates were selected for phylogenetic analysis. Obtained 16S rRNA gene sequences fell into four groups： Firmicutes, Alpha-Proteobacteria, Gamma-Proteobacteria, and Actinobacteria, with the Firmicutes being dominant. Microbial compositions derived from direct and enrichment cultivations were not overlapped. We suggest that it is a better way to explorethe culturable microbial diversity in ice core by combining the approaches of both direct and enrichment cultivation.

Core discing characteristics and mitigation approach by a novel developed drill bit in deep rocks

Core discing often occurs in deep rocks under high-stress conditions and has been identified as an important characteristic for deep rock engineering.This paper presents the formation mechanism of core discing firstly.Then,the interaction between diamond drill bits and rock was analyzed based on numerical modeling.A novel drill bit with an inner conical crown for the mitigation of core discing was designed and verified by simulation experiments.The mitigation method was applied in the cavern B1 of CJPL-Ⅱand satisfactory results had been achieved.The percentage of core discing had been obviously decreased from 67.8%when drilling with a rectangular crown drill bit,to 26.5%when an inner conical crown drill bit had been adopted.This paper gives full insight into core discing characteristics and provides a new method for core discing mitigation;it will potentially contribute to stress measurement in deep rock engineering.

Comparison and Calibration of Elemental Measurements in Sediments Using X-Ray Fluorescence Core Scanning with ICP Methods: A Case Study of the South China Sea Deep Basin

The X-ray fluorescence(XRF)core scanning method is widely applied in studies of sedimentary paleoenvironments due to its convenient pretreatment,nondestructive characteristics,fast execution,continuous scanning,and high resolution.XRF core scanning for sediments is commonly used in the studies on the South China Sea.This study compares XRF-scanned intensities and measured inductively coupled plasma(ICP)elemental contents of core CS11 in the northeast South China Sea deep basin.The results show that the analyzed elements can be separated into three classes.Class I includes elements with high correlation coefficients,such as Ca,Sr,and Zr;Class II contains elements with average correlation coefficients,such as Fe,Mn,Ti,and Cu;and Class III comprises elements with low correlation coefficients,such as K,Ni,Zn,Rb,and Al.In the South China Sea deep basin,pore water,compaction,and grain size have weak effects on the elemental intensities and contents of short core sediments.Hence,for elements with high correlation coefficients,a linear relationship model can be established by the least-squares method,in which the converted XRF intensities are approximately equal to the measured ICP contents.Based on the established log-ratio calibration model,the resulting ln(K/Ca),ln(Ti/Ca),ln(Fe/Ca),and ln(Zr/Ca)values generally display the same variation trends as the measured curves.The elemental contents and ratios produced by the linear model via the least-squares method and the log-ratio calibration model are expected to provide high-resolution data support for future paleoenvironmental research on the South China Sea deep basin.

Sea Salt Sodium Record in a Shallow Ice Core from East Antarctica as a Potential Proxy of the Antarctic Sea Ice Extent in Southern Indian Ocean

Antarctic sea ice has experienced an increasing trend in recent decades,especially in the Ross Sea and Indian Ocean sectors.Sea ice variability affects greatly the maritime airmass transport from high latitude to Antarctic continent.Here we present a new ice core record of sea salt sodium(ssNa+)concentration at annual-resolution in the Princess Elizabeth Land spanning from 1990 to 2016,showing that this marker could be used as a potential proxy for reconstructing the sea ice extent(SIE)in the Southern Indian Ocean(SIO)given their significant correlation(R=-0.6,P<0.01)over the past 27 years.The correlation and composite analyses results show that the ssNa^+at the 202 km inland from Zhongshan Station and the SIE changes in SIO are closely related to the Indian Ocean Dipole(IOD)and Southern Annular Mode(SAM).The northward wind in central SIO occurs during positive IOD and the strengthened westerlies occurs during positive SAM,both of which favor increased sea ice in SIO and lead to the decreased ssNa^+concentration at the coastal site.

INTERACTION BETWEEN THE ENSO AND ASIAN MONSOON RECORDED IN DASUOPU ICE CORE FROM HIMALAYAS

Two of the most important climate interactions between the ocean and atmosphere are the Asian monsoon in the Indian Ocean and El Niň o\|Southern Oscillation (ENSO) in the Pacific Ocean. Previous analyses have variously concluded either that the Asian monsoon affects the ENSO or that the converse is true. But previous research about the relation between ENSO and Asian monsoon was focused on tropic or low\|latitude ocean. There was litter research on this relation in mid\|latitude continent.Dasuopu glacier is located in middle Himalayas (85 43°E, 28 23°N) at an altitude over 7000m . In 1996 and 1997, a collaboration between China and American scientists had a series of ice core drilling performed down to the bed of Dasuopu glacier. The core analysis now is still in progress. Here a record of oxygen isotopic composition and accumulation of the ice core provides a record of relation between ENSO phenomenon and the Asian monsoon.The growth rate of the core averages over 1m per year, which provides a high resolution to date layer. There is a strong seasonal cycle in δ 18 O, thus we used the seasonal δ 18 O cycle to count annual layers in the core. According to previous studying, we assigned the high δ 18 O extremes to winter. The complete δ 18 O time series indicates a increasing trend since 1980.The δ 18 O in the core correlate strongly with observed sea surface temperature (SST) over the last decade years. Thus, the δ 18 O in the ice core at the location of south Tibetan Plateau is a relatively simple proxy for SST, which is very different from that at the location of north Plateau where δ 18 O proxy for local temperature. Furthermore, the δ 18 O in the core is also corresponds with an index of monsoon strength, which demonstrates that the ice core successfully captures the large\|scale aspects of Asian monsoon behavior.Generally, negative δ 18 O anomalies in the core correspond with weak monsoons, warm SST and El Niň o, on the other hand, positive anomalies correlate with strong monsoons, cool SST and La Niň o. These imply interaction between the tropics and mid\|latitudes, channeled through the monsoon system.

Glaciochemical records from Naimona’Nyi ice core in the Himalayas

A 6-m ice core was recovered in 2004 from the Naimona＇Nyi Glacier, the middle Himalayas. Empirical orthogonal function （EOF） analysis on the major ion reveals that EOF1 represents the variations of majority of ions which may be originated from crustal aerosols. Comparing the calcium concentrations from the Naimona＇Nyi with these from Dasuopu, East Rongbuk and Guliya ice cores, it is observed that calcium, a good indicator of the input of crustal aerosol in snow, concentrates mostly in the Guliya ice core located on the northern Tibetan Plateau, and gradually decreases from west to east in the Himalayas.

A concise overview on historical black carbon in ice cores and remote lake sediments in the northern hemisphere

Black Carbon(BC),as a driver of environmental change,could significantly impact the snow by accelerating melting and decreasing albedo.Systematic documentation of BC studies is crucial for a better understanding of its spatial and temporal trends.This study reviewed the BC studies in the ice core and remote lake sediments and their sources in the northern hemisphere.The literature surveyed points to around 2.9 to 3.7 times increase of BC in the European Alps and up to a three-fold increase of BC in the Himalayan-Tibetan Plateau(HTP)after the onset of industrialization in Europe and Asia,respectively.BC concentration from Greenland ice core showed seven times increase with an interrupted trend after 1950's.South Asian emissions were dominant in the HTP along with a contribution from the Middle East,whereas Western European and local emissions were responsible for the change in BC concentration in the European Alps.In the Arctic,contributions from North America,Europe and Asia persisted.Similarly,a historical reconstruction of lake sediments records demonstrates the effects of emissions from long-range transport,sediment focusing,local anthropogenic activities,precipitation and total input of flux on the BC concentration.

Anthropogenic trace metals in an ice core at Vestfonna,Svalbard, Norway

A 211 m depth ice core observation was carried out at the top of the Vestfonna Ice Cap in Nordaustlandet, Svalbard, Norway in 1995. Chronology of the ice core was determined by tritium analysis and comparison to the volcanic eruption of Laki; the resulting accumulation rate is 0.34 0.35 m water eq.·yr -1 for the last 400 year. Concentrations of Al, V, Cr, Fe, Cu, Zn, As, Ag, Cd, Pb and U in an ice core have been determined by an inductively coupled plasma mass spectrometry (ICP MS) with a desolvated micro concentric nebulizer, which is a recent development and can achieve high sensitivity with low uptake rate of 60 mL/min. The concentrations of Pb, Cu, and Zn had increased from 1940s declined from 1970s to present. However, the profiles of Pb, Cu, and Zn were different and they seem to be influenced by the difference of sources. Since the ratios between Cu, Pb, and Zn in Svalbard is similar to that in French Alps, the source area of these elements is estimated to be Europe.

Source of major ions from an ice core of the No.12 Glacier in Laohugou Valley,Qilian Mountain

In June 2006,a 20.12 m shallow ice core was recovered from an elevation of 5,040 m in the northern branch firn basin of No.12 Glacier,Great Snow Mountain,in the western part of Qilian Mountain,China.Isotopes(δ 18 O),major soluble ions,and radionuclide(β-activity) measurements from the ice core revealed a 46-year record(1960-2006).In this paper,the method of sea-salt ion tracer,correlation analysis and trend analysis were used in this research to confirm the source of the chemical composition.The correlation analysis and HYSPLIT backward trajectory analysis suggests that atmospheric soluble dust species dominate the chemical signature.

Mineral and Sea-Salt Aerosol Fluxes over the Last 340 kyr Reconstructed from the Total Concentration of Al and Na in the Dome Fuji Ice Core

A quantitative analysis of the total concentrations of Al and Na in the Antarctic ice sheet during the past 340 kyr was performed by applying the acid digestion method to the Dome Fuji ice core. Atmospheric fluxes of mineral and sea-salt aerosol to Dome Fuji were calculated from the total concentration. The average fluxes of mineral aerosol to Dome Fuji in the periods of glacial maximum, 18.6 ± 10.1 mg·m–2·yr–1, were larger than the value in the interglacial periods, 3.77 ± 2.20 mg·m–2·yr–1. Conversely, the fluxes of sea-salt have no significant difference between the average value of glacial maximum, 130 ± 55 mg·m–2·yr–1, and that of interglacial, 111 ± 54 mg·m–2·yr–1. The results obtained in this study suggest that the variation of mineral aerosol flux in Dome Fuji, together with climate change, was much larger than that of sea-salt aerosol flux. This result may have occurred because the variety in the intensity of the source and transport during the glacial-interglacial cycle is more significant for mineral aerosol than that for sea-salt aerosol.

Turbidite Events Recorded in Deep-sea Core IR-GC1 off Western Sumatra： Evidence from Grain-size Distribution

Sediment core IR-GC1, from the abyssal basin of the Indian Ocean off Sumatra, may provide important information on depositional events related to earthquakes and tsunamis. In this study, based on a combination of grain-size analysis with lithological studies and oxygen isotope stratigraphy, seven deep-sea turbidite layers were identified, corresponding to seven turbidity events that occurred at 128- 130, 105-107, 98-100, 86-87, 50-53, 37-41 and 20-29 ka. The sediments of the turbidite deposits are characterized by coarse grain sizes, poor sorting, wide kurtosis, bimodal frequency distributions and clear depositional variations. Particle size grading is also an important signature of deep-sea turbidite deposits and can be used as an indicator to identify turbidite layers. Possible triggering mechanisms for the turbidite events include tsunamis, earthquakes, volcanic eruptions and sea-level changes.

Geochemical characteristics and their significances of rare-earth elements in deep-water well core at the Lingnan Low Uplift Area of the Qiongdongnan Basin

A geochemical analysis of rare-earth elements （REEs） in 97 samples collected from the core of deep-water Well LS-A located at the Lingnan Low Uplift Area of the Qiongdongnan Basin is conducted, with the pur-pose of revealing the changes of sedimentary source and environment in the study region since Oligocene and evaluating the response of geochemical characteristics of REEs to the tectonic evolution. In the core samples, both∑REE and∑LREE （LREE is short for light-group REEs） fluctuate in a relatively wide range, while∑HREE （HREE is short for heavy-group REEs） maintains a relatively stable level. With the stratigraphic chronology becoming newer, both∑REE and∑LREE show a gradually rising trend overall. The∑REE of the core is relatively high from the bottom of Yacheng Formation （at a well depth of 4 207 m） to the top of Ledong Formation, and the REEs show partitioning characteristics of the enrichment of LREE, the stable content of HREE, and the negative anomaly of Eu to varying degrees. Overall the geochemical characteristics of REEs are relatively approximate to those of China＇s neritic sediments and loess, with significant ＂continental ori-entation＂. The∑REE of the core is relatively low in the lower part of Yacheng Formation （at a well depth of 4 207-4 330 m）, as shown by the REEs partitioning characteristics of the depletion of LREE, the relative enrich-ment of HREE, and the positive anomaly of Eu; the geochemical characteristics of REEs are approximate to those of oceanic crust and basalt overall, indicating that the provenance is primarily composed of volcanic eruption matters. As shown by the analyses based on sequence stratigraphy and mineralogy, the provenance in study region in the early Oligocene mainly resulted from the volcanic materials of the peripheral uplift ar-eas; the continental margin materials from the north contributed only insignificantly; the provenance devel-oped to a certain extent in the late Oligocene. Since the Miocene, the provenance has ceaselessly expanded from proximal to distal realm, embodying a characteristic of multi-source sedimentation. In the core strata with 31.5, 28.4, 25.5, 23, and 16 Ma from today, the geochemical parameters of REEs and Th/Sc ratio have significant saltation, embodying the tectonic movement events in the evolution of the Qiongdongnan Basin. In the tectonic evolution history of the South China Sea, the South China Sea Movement （34-25 Ma BP, early expansion of the South China Sea）, Baiyun Movement （23 Ma BP）, late expansion movement （23.5-16.5 Ma BP）, expansion-settlement transition, and other important events are all clearly recorded by the geochemi-cal characteristics of REEs in the core.

RECENT 200 YEARS CLIMATIC AND ENVIRONMENTAL RECORDS FROM THE FAR EAST RONGBUK ICE CORE, MT. QOMOLANGMA (EVEREST)

During the Sino\|American Expedition to Mt. Qomolangma in May 1997, a 41m ice core was recovered from an elevation of 6500m from the northern branch firn basin of the Far East Rongbuk Glacier in Mt. Everest. The ice core was dated down to 1814 by counting δ 18 O peaks and referring to the variations of β activity and major ion concentrations. The average annual accumulation is 224mm (ice equivalent). Five cold periods and five warm periods have been reconstructed from the ice core for the last 200 years and the general tendency of climatic change is warming, which is agree with the temperature change in the Northern Hemisphere. Also the climatic records in Far East Rongbuk ice core has good agreement with that in the Guliya ice core. This indicates that the climatic changes are consistent in the northwestern and southern Qinghai—Tibetan Plateau, and the ranges of climatic changes are larger in southern Plateau than that in northwestern Plateau. Though the δ 18 O variations has some negatively correlation with precipitation amount for short time scale, these do not effect δ 18 O changes reflecting temperature for long time scale.

Review of pre-processing technologies for ice cores

High-resolution ice core records covering long time spans enable reconstruction of the past climatic and environmental conditions allowing the investigation of the earth system's evolution. Preprocessing of ice cores has direct impacts on the data quality control for further analysis since the conventional ice core processing is time-consuming, produces qualitative data, leads to ice mass loss, and leads to risks of potential secondary pollution. However, over the past several decades, preprocessing of ice cores has received less attention than the improvement of ice drilling, the analytical methodology of various indices, and the researches on the climatic and environmental significance of ice core records. Therefore, this papers reviews the development of the processing for ice cores including framework, design as well as materials, analyzes the technical advantages and disadvantages of the different systems. In the past, continuous flowanalysis(CFA) has been successfully applied to process the polar ice cores. However, it is not suitable for ice cores outside polar region because of high level of particles, the memory effect between samples, and the filtration before injection. Ice core processing is a subtle and professional operation due to the fragility of the nonmetallic materials and the random distribution of particles and air bubbles in ice cores, which aggravates uncertainty in the measurements. The future developments of CFA are discussed in preprocessing, memory effect, challenge for brittle ice, coupling with real-time analysis and optimization of CFA in the field. Furthermore, non-polluting cutters with many different configurations could be designed to cut and scrape in multiple directions and to separate inner and outer portions of the core. This system also needs to be coupled with streamlined operation of packaging, coding, and stacking that can be implemented at high resolution and rate, avoiding manual intervention. At the same time, information of the longitudinal sections could be scanned andidentified, and then classified to obtain quantitative data. In addition, irregular ice volume and weight can also be obtained accurately. These improvements are recorded automatically via user-friendly interfaces. These innovations may be applied to other paleomedias with similar features and needs.