Search

This paper presents a methodological tool for dynamic reconstruction of the state of the ocean, based, as an example, on observations from the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) experiment. The data used in this study were collected in the Amundsen Basin between October 2019 and January 2020. Analysing observational data to assess tracer field and upper-ocean dynamics is highly challenging when measurement platforms drift with the ice pack due to continuous drift speed and direction changes. We have equipped the new version of the coastal branch of the global Finite-volumE sea ice–Ocean Model (FESOM-C) with a nudging method. Model nudging was carried out assuming a quasi-steady state. Overall, the model can reproduce the lateral and vertical structure of the temperature, salinity, and density fields, which allows for projecting dynamically consistent features of these fields onto a regular grid. We identify two separate depth ranges of enhanced eddy kinetic energy located around two maxima in buoyancy frequency: the depth of the upper halocline and the depth of the warm (modified) Atlantic Water. Simulations reveal a notable decrease in surface layer salinity and density in the Amundsen Basin towards the north but no significant gradient from east to west. However, we find a mixed-layer deepening from east to west, with a 0.084 m km−1 gradient at 0.6 m km−1 standard deviation, compared to a weak deepening from south to north. The model resolves several stationary eddies in the warm Atlantic Water and provides insights into the associated dynamics. The model output can be used to further analyse the thermohaline structure and related dynamics associated with mesoscale and submesoscale processes in the central Arctic, such as estimates of heat fluxes or mass transport. The developed nudging method can be utilized to incorporate observational data from a diverse set of instruments and for further analysis of data from the MOSAiC expedition.

This investigation addresses the tsunami inundation in Lima and Callao caused by the massive 1746 earthquake (Mw 9.0) along the Peruvian coast. Numerical modeling of the tsunami inundation processes in the nearshore includes strong nonlinear numerical terms. In a comparative analysis of the calculation of the tsunami wave effect, two numerical codes are used, Tsunami-HySEA and TsunAWI, which both solve the shallow water (SW) equations but with different spatial approximations. The comparison primarily evaluates the flow velocity fields in inundated areas. The relative importance of the various parts of the SW equations is determined, focusing on the nonlinear terms. Particular attention is paid to the contribution of momentum advection, bottom friction, and volume conservation. The influence of the nonlinearity on the degree and volume of inundation, flow velocity, and small-scale fluctuations is determined. The sensitivity of the solution concerning the bottom friction parameter is also investigated, showing the effects of nonlinearity processes in the inundated areas, wave heights, current velocity, and the spatial structure variations shown in tsunami inundation maps.

In this paper, we anticipate geospatial population distributions to quantify the future number of people living in earthquake-prone and tsunami-prone areas of Lima and Callao, Peru. We capitalize upon existing gridded population time series data sets, which are provided on an open-source basis globally, and implement machine learning models tailored for time series analysis, i.e., based on long short-term memory (LSTM) networks, for prediction of future time steps. Specifically, we harvest WorldPop population data and teach LSTM and convolutional LSTM models equipped with both unidirectional and bidirectional learning mechanisms, which are derived from different feature sets, i.e., driving factors. To gain insights regarding the competitive performance of LSTM-based models in this application context, we also implement multilinear regression and random forest models for comparison. The results clearly underline the value of the LSTM-based models for forecasting gridded population data; the most accurate prediction obtained with an LSTM equipped with a bidirectional learning scheme features a root-mean-squared error of 3.63 people per 100 × 100 m grid cell while maintaining an excellent model fit (R2= 0.995). We deploy this model for anticipation of population along a 3-year interval until the year 2035. Especially in areas of high peak ground acceleration of 207–210 cm s−2, the population is anticipated to experience growth of almost 30 % over the forecasted time span, which simultaneously corresponds to 70 % of the predicted additional inhabitants of Lima. The population in the tsunami inundation area is anticipated to grow by 61 % until 2035, which is substantially more than the average growth of 35 % for the city. Uncovering those relations can help urban planners and policymakers to develop effective risk mitigation strategies.