Jaakko Putkonen

A large fraction of the Alaska terrain is underlain by regolith that contains excess ice (ice that in volume exceeds the natural pore space). Possible loss of excess ice would lead to ground subsidence, which in turn would disturb and destabilize the ground surface. Depending on the volume fraction and local variations in excess ice content and soil properties, the subsidence would lead to forest disturbances, increased soil erosion, hydrological changes, and potential increase in sediment supply to rivers and creeks. In addition to InSAR annual evaluations, currently, the best available long‐term data of the ground subsidence come from historical observations (~20 years) on a few field sites of limited spatial extent in Alaska and Western Canada. Due to the projected changes of environmental conditions in Alaska in the coming decades, there is a reasonable expectation of widespread ground subsidence. To better understand the spatial variations and varying magnitudes of expected subsidence, we set out to model and map the future ground subsidence across north‐central Alaska, along the Dalton Highway corridor. We analyzed the current thermal regime on the dominant landcovers on various amounts of mapped excess soil ice. The in situ observations and the purpose‐built subsidence model, coupled with the existing landcover and excess ice maps and analytical projections under varying assumptions, allowed us to model the future ground subsidence by 2050. We run the model for two separate excess ice scenarios for north‐central Alaska: 50%, and 25% near surface excess ice guided by published map. The results show the projected average subsidence rates of 1.79–2.33 cm/year for 2025–2050 in the southern part (Yukon River area) of the study area for 25% and 50% excess ice, respectively, whereas the smallest projected subsidence rates of < 0.17 cm/year in the northern part (North Slope) of the study area. This broad variation is explained by the strong and persistent North–South climatic gradient. However, the small‐scale spatial variation in the subsidence, shown in the produced maps, reflects both the spatially varying amounts of soil excess ice and mapped landcovers.

Local weather forecasts in the Arctic outside of settlements are challenging due to the dearth of ground-level observation stations and high computational costs. During winter, these forecasts are critical to help prepare for potentially hazardous weather conditions, while in spring, these forecasts may be used to determine flood risk during annual snow melt. To this end, a hybrid VMD-WT-InceptionTime model is proposed for multi-horizon multivariate forecasting of remote-region temperatures in Alaska over short-term horizons (the next seven days). First, the Spearman correlation coefficient is employed to analyze the relationship between each input variable and the forecast target temperature. The most output-correlated input sequences are decomposed using variational mode decomposition (VMD) and, ultimately, wavelet transform (WT) to extract time-frequency patterns intrinsic in the raw inputs. The resulting sequences are fed into a deep InceptionTime model for short-term forecasting. This hybrid technique has been developed and evaluated using 35+ years of data from three locations in Alaska. Different experiments and performance benchmarks are conducted using deep learning models (e.g., Time Series Transformers, LSTM, MiniRocket), and statistical and conventional machine learning baselines (e.g., GBDT, SVR, ARIMA). All forecasting performances are assessed using four metrics: the root mean squared error, the mean absolute percentage error, the coefficient of determination, and the mean directional accuracy. Superior forecasting performance is achieved consistently using the proposed hybrid technique.

The well-documented decrease in the annual minimum Arctic sea ice extent over the past few decades is an alarming indicator of current climate change. However, much less is known about the thickness of the Arctic sea ice. Developing accurate forecasting models is critical to better predict its changes and monitor the impacts of global warming on the total Arctic sea ice volume (SIV). Significant improvements in forecasting performance are possible with the advances in signal processing and deep learning. Accordingly, here, we set out to utilize the recent advances in machine learning to develop non-physics-based techniques for forecasting the sea ice volume with low computational costs. In particular, this paper aims to provide a step-wise decision process required to develop a more accurate forecasting model over short- and mid-term horizons. This work integrates variational mode decomposition (VMD) and bidirectional long short-term memory (BiLSTM) for multi-input multi-output pan-Arctic SIV forecasting. Different experiments are conducted to identify the impact of several aspects, including multivariate inputs, signal decomposition, and deep learning, on forecasting performance. The empirical results indicate that (i) the proposed hybrid model is consistently effective in time-series processing and forecasting, with average improvements of up to 60% compared with the case of no decomposition and over 40% compared with other deep learning models in both forecasting horizons and seasons; (ii) the optimization of the VMD level is essential for optimal performance; and (iii) the use of the proposed technique with a divide-and-conquer strategy demonstrates superior forecasting performance.

The lack of ground level observation stations outside of settlements makes monitoring and forecasting local weather and permafrost challenging in the Arctic. Such predictive pieces of information are essential to help prepare for potentially hazardous weather conditions, especially during winter. In this study, we aim at enhancing predictive analytics in Alaska of permafrost and temperature by using a hybrid forecasting technique. In particular, we propose VMD-WT-InceptionTime model for short-term air temperature forecasting.This proposed technique incorporates data preprocessing techniques and deep learning to enhance the accuracy of the next seven days air temperature forecasts. Initially, the Spearman correlation coefficient is utilized to examine the relationship between different inputs and the forecast target temperature. Following this, Variational Mode Decomposition (VMD) is used to decompose the most output-correlated input variables (i.e., temperature and relative humidity) to extract intrinsic and non-stationary time-frequency features from the original sequences. The Wavelet Transform (WT) is then employed to further extract intrinsic multi-resolution patterns from these decomposed input variables. Finally, a deep InceptionTime model is used for multi-step air temperature forecasting using these processed sequences. This forecasting technique was developed using an open dataset holding 20+ years of data from three locations in Alaska: North Slope, Alaska, Arctic National Wildlife Refuge, Alaska, and Diomede Island region, Bering Strait. Model performance has been rigorously evaluated of metrics including RMSE, MAPE and error.Results highlight the effectiveness of the proposed hybrid model in providing more accurate short-term forecasts than several baselines (GBDT, SVR, ExtraTrees, RF, ARIMA, LSTM, GRU, and Transformer). More specifically, this technique reported RMSE and MAPE average increase rates amounting to 11.21% and 16.13% in North Slope, 30.01% and 34.97% in Arctic National Wildlife Refuge, and 16.39%, 23.46% in Diomede Island region. In addition, the proposed technique produces forecasts over all seven horizons with a maximum error of -1.2K, and an average error lower than 0.18K for North Slope. For Arctic National Wildlife Refuge, a maximum error of -0.9K, and an average of < 0.1K. While a maximum error of -0.8K, and an average of <0.13K, for Diomede Island region. However, the worst performances achieved were errors of around 6K in the third horizon (i.e., 3rd day) for North Slope and the Arctic National Wildlife Refuge and the last horizon (i.e., 7th day) for the Diomede Islands region. Most of the worst performances of the proposed technique in all three locations can be attributed to having to produce forecasts of higher variations and wider temperature ranges than their averages.Overall, this research highlights the potential of the decomposition techniques and deep learning to: 1) reveal and effectively learn the underlying cyclicity of air temperatures at varying resolutions that allows for accurate predictions without any knowledge of the governing physics, 2) produce accurate multi-step temperature forecasts in Arctic climates.

Glacial ice contains information on paleoclimate, but typically is less than 1 Myrs old because glaciers continuously flow and melt. A recently discovered buried ice mass in Antarctica is dated to 3-5 Myrs and highlights the potential for long preservation of ancient ice under a layer of debris. Only two such ice masses are so far known in Antarctica. Given the significant scientific potential and lack of systematic mapping, we set out to locate all buried ice masses in Transantarctic Mountains (TAM). We visually analyzed >8,000 high-resolution satellite images covering much of the TAM. We searched for the polygonal patterned ground that signifies ice in the ground and once detected the corresponding digital elevation model was inspected for presence of a convex landform. When both features coincide, they indicate the presence of a buried ice body. We identified 22 new sites that are likely to conceal massive buried ice masses.

The current knowledge of the spatial variability of precipitation in High Mountain Asia is based on the remotely-sensed estimates (coarse spatial and temporal resolution) or data from sparsely-distributed rain gauges. However, as precipitation is strongly affected by topography in mountainous terrain, the spatially varying precipitation and the resulting water balances are currently poorly understood. To fill this gap in knowledge, we studied the spatial variation of the precipitation and its impact on water balance in a small headwater basin located in the foothills of the Himalaya, Nepal. We deployed ten rain gauges and climate stations, spanning the whole elevation range 700–4500 m above sea level (masl) for a period of four years. Our results show a quadratic polynomial relationship between annual precipitation and station elevation, which are used to produce annual precipitation maps. The performance of the elevation-based precipitation estimates is adequate in closing the water balance while the performances of average precipitation and Thiessen polygon method are poor and inconsistent in closing the water balance. We also demonstrate that precipitation estimates from one or two gauges at the lowest basin elevation substantially underestimate the water balance. However, the precipitation from one or two rain gauges at 2000–3000 masl provide a significantly better estimate of the water balance of a small headwater basin.