Hourly Time Series Research Articles

Application of Chaos Theory to Time-Series Urban Measurements of Meteorological Variables and Radon Concentration: Analysis and Interpretation

Through chaos theory, experimental data of hourly time series are analyzed. These time series consist of Radon concentration levels and meteorological variables of temperature, pressure, and relative humidity within the boundary layer and very close to the ground. Results were obtained in two urban dwellings for family use and for two different periods of time, of the order of one month and one month plus one week, respectively. Each time series was subjected to a chaotic analysis showing the existence of the characteristic chaotic parameters in the appropriate ranges: Lyapunov coefficient (λ), correlation dimension (Dc), Kolmogorov entropy (SK), Lempel-Ziv complexity (LZ), Hurst coefficient (H), maximum predictability time (τ), lost information (<ΔI>) and fractal dimension (D). The studied processes show to be irreversible. From the chaotic parameters, it is shown that the ratio between the entropy of each meteorological variable and the radon concentration is very sensitive to relative humidity. Likewise, the meteorological variables that most affect the concentration of Radon are relative humidity and temperature. The concordance between the results obtained and those delivered by analyzes carried out through other methodologies in longer periods is verified.

A hybrid intelligent framework for forecasting short-term hourly wind speed based on machine learning

With the development of wind power which is the great substitute for traditional energy, it is worth conducting an in-depth exploration of the hourly wind speed time series which is chaotic due to the complex weather. In this paper, the hybrid intelligent framework is proposed as an integrated prediction tool with high accuracy for signal pre-processing, data prediction, and result optimization of short-term hourly wind speed. Specifically, its excellent performance is guaranteed through adequate information extraction on three stages. The first stage is completed in the signal pre-processing module that the interference information is cleaned up from the hourly wind speed signal via de-noising. The second stage is conducted in the data prediction module that the hidden regular information is fully extracted via the decomposition method. The third stage is performed in the resulting optimization module that the residual information is recovered via error modification. For illustration, the performance of the proposed framework is evaluated through historical hourly wind speed, taken from publicly available Sotavento wind farms. The obtained experimental results indicate that de-noising is beneficial to capturing the real trend, but it may negatively impact short-term prediction accuracy. For the hybrid prediction model based on the empirical mode decomposition-based method, the de-noising mode integrating into the decomposition process is more effective than independent de-noising. The second stage is the key to improving the forecasting performance that, adopting the decomposition method, the average fitting performance is improved by 52.84% than the single models. Before the third stage, chaos test is necessary to determine whether there is a requirement of error modification. In summary, the proposed prediction framework can capture the complex characteristics for different short-term hourly wind speed time series, achieving greater performance than the other comparative models.

Probabilistic forecasting method for mid-term hourly load time series based on an improved temporal fusion transformer model

The growth of distributed renewable energy and demand-side responsiveness has increased the difficulty of mid-term hourly load time-series forecasting. This study presents a probabilistic forecasting method for hourly load time series based on an improved temporal fusion transformer (ITFT) model to achieve more accurate and thorough forecasting results. The raw univariate time series of the hourly load was reconstructed into multiple day-to-day load time series at different hour-points to reconcile the contradiction between learning the temporal dependence on a long prediction horizon and reducing model complexity. The corresponding hour point was used as a static covariable input to distinguish the differences. Based on the original temporal fusion transformer (TFT) model, the ITFT model replaces the long short-term memory (LSTM) with a gated recurrent unit (GRU) to learn long-term dependence more efficiently. Furthermore, quantile constraints and prediction interval (PI) penalty terms were incorporated into the original quantile loss function to prevent quantile crossover and construct more compact prediction intervals (PIs). The results of two actual examples show that the proposed method is explanatory and can significantly improve the reliability and compactness of probabilistic load forecasting (PLF) results compared with other popular methods.

High frequency UV–Vis sensors estimate error in riverine dissolved organic carbon load estimates from grab sampling

High frequency ultraviolet – visible (UV–VIS) sensors offer a way of improving dissolved organic carbon (DOC) load estimates in rivers as they can be calibrated to DOC concentration. This is an improvement on periodic grab sampling, or the use of pumped sampling systems which store samples in-field before collection. We hypothesised that the move to high frequency measurements would increase the load estimate based on grab sampling due to systemic under-sampling of high flows. To test our hypotheses, we calibrated two sensors in contrasting catchments (Exe and Bow Brook, UK) against weekly grab sampled DOC measurements and then created an hourly time series of DOC for the two sites. Taking this measurement as a ‘true’ value of DOC load, we simulated 1,000 grab sampling campaigns at weekly, fortnightly and monthly frequency to understand the likely distribution of load and error estimates. We also performed an analysis of daily grab samples collected using a pumped storage sampling system with weekly collection. Our results show that: a) grab sampling systemically underestimates DOC loads and gives positively skewed distributions of results, b) this under-estimation and positive skew decreases with increasing sampling frequency, c) commonly used estimates of error in the load value are also systemically lowered by the oversampling of low, stable flows due to their dependence on the variance in the flow-weighted mean concentration, and d) that pumped storage systems may lead to under-estimation of DOC and over estimation of specific ultra-violet absorbance (SUVA), a proxy for aromaticity, due to biodegradation during storage.

Short-term solar irradiance forecasting based on a novel Bayesian optimized deep Long Short-Term Memory neural network

Accurate forecasting is indispensable for improving solar renewables integration and minimizing the effects of solar energy's intermittency. Existing research on time series solar forecasting confronts challenges such as determining the accurate hyperparameters and flexibility in considering meteorological parameters. This study proposes a novel deep learning model, namely an optimized stacked Bi-directional Long Short-Term Memory(BiLSTM)/ Long Short-Term Memory(LSTM) model to forecast univariate and multivariate hourly time series data by integrating stacked LSTM layers, drop out architecture, and LSTM based model. The performance of the model is enhanced by Bayesian optimization with the tuning of six relevant hyperparameters. To evaluate the model, standard Global Horizontal Irradiance (GHI) and observed Plane of Array (POA) irradiance with meteorological real-world solar data from Sweihan Photovoltaic Independent Power project in Abu Dhabi, UAE, and NREL solar data for year-round data are forecasted. Furthermore, the performance of the proposed algorithm is also evaluated under weather uncertainty for different climate types. The forecasting accuracy is evaluated based on various performance metrics and it is observed that the proposed model offered the best R2 values, 0.99 for univariate as well as multivariate models using GHI data and 0.97 using POA data. The findings suggest that the proposed model is a reliable technique for solar prediction due to its comparable performance with both GHI and POA in terms of accuracy.

The impact of variable renewables on the distribution of hourly electricity prices and their variability: A panel approach

This paper investigates the impact of intermittent renewable generation on the distribution of electricity prices and their variability in Denmark and Germany. We exploit hourly data from 2015 to 2020 and employ a novel panel quantile approach - the Quantiles via moments (MMQR) method. Previous research has mainly used aggregated-daily data and have applied a time-series setting. We argue that since the electricity price formation and renewable energy generation can show great variations during a day, a panel setting with 24 individuals-hours could offer higher accuracy. Therefore, we apply a panel approach that accounts for both the time and cross-sectional dimension of electricity prices. The panel allows us to control for time-invariant (hourly-specific) characteristics and can reveal hidden market dynamics that exist during a day. The combination of hourly-specific effects and the quantile approach enable us to estimate the renewable sources effect on various price quantiles while controlling for market dynamics. In this way, we investigate extreme market cases accounting for the range and distribution of the electricity prices data. The results suggest that the merit-order effect occurs in both countries, with wind and solar generation having diverse effects on the electricity price distribution. Thus, policy makers should consider this diversifying effect to develop efficient renewable support schemes. We also explore non-linearities by including different demand levels in our model and investigate price variability. The outcomes indicate that wind generation increases (decreases) the occurrence of price fluctuations for low demand (high demand) in both countries. Meanwhile, in Germany, solar power stabilizes price fluctuations for high demand levels, stronger than wind. Market risk information could be useful for organizations in recognizing beneficial investment opportunities or hedging strategies. We finally aggregate the hourly observations into daily and compare the estimation outcomes. The results prompt us to believe that aggregated time-series tend to underestimate the RES impact on prices. In addition, we estimate the same models using hourly data in a time-series approach in order to verify that the diverse effect between aggregated time-series and hourly panel data is driven by the time-invariant characteristics, and not the data resolution. We find that the hourly time-series underestimate the merit-order effect, like in the aggregated time-series case, which supports our claims that the results are steered by the cross-sectional dimension. Thus, a panel approach could provide higher accuracy estimates of the RES influence on electricity prices. In conclusion, hourly-related features seem to affect the merit-order effect and its robustness, and a panel approach should be considered when investigating electricity markets.

Study of the Effect of Urban Densification and Micrometeorology on the Sustainability of a Coronavirus-Type Pandemic

This research examines the persistence of a pandemic in urban environments subjected to intensive densification processes, applying chaotic analysis tools to hourly time series constructed by relating accumulated patients with meteorological and pollutant variables (measured at ground level). To investigate this objective, seven communes of the metropolitan region of Santiago de Chile that present intensive urbanization processes that affect urban micrometeorology, favoring the concentration of pollutants, were considered. Quotients were constructed between the number of hourly patients with SARS-CoV-2 that accumulated in each commune over a period of two years and the hourly variables of urban micrometeorology (temperature, magnitude of wind speed, relative humidity) and pollutant concentration (tropospheric ozone, particulate material of 2.5 and 10 μm) constituting a new family of time series. Chaos theory was applied to these new time series, obtaining the chaotic parameters Lyapunov coefficient, correlation entropy, Lempel–Ziv complexity, Hurst coefficient and the fractal dimension in each measurement commune. The results showed that the accumulated patients (2020–2022), of the order of 400,000, belonged to the five communes (with a built area of approximately 300,000 m2 in recent years) that had the highest urban densification, which affected urban meteorology, favored the concentration of pollutants and made the SARS-CoV-2 pandemic more persistent. The “ideal” density of built housing should balance a pandemic and nullify its expansion.

Wind speed forecasting using a hybrid model considering the turbulence of the airflow

This work presents the conformation of a hybrid model aimed at forecasting the wind speed one step ahead. The model comprises the Multiplicative Cascade and Persistence. To couple both techniques, a transition point was used, in this case, 6.6 m/s, defined from the spectrum of wind speed near the ground. The Multiplicative Cascade was used when speeds were higher than the transition point, while Persistence was used when lower. Two time series measured in Tamaulipas and Veracruz, Mexico, were used to build the hybrid model. The performance of the individual models concerning the hybrid model was compared, finding that the adjustment of the hybrid model exceeds that of the Multiplicative Cascade and Persistence models for both series. Different error metrics were used in the quantitative analysis, where the hybrid model was the most competitive. For example, in Mean Absolute Percentage Error, the difference in performance between the hybrid model and Persistence is 2.72% points for the 10-min time series. In contrast, this difference is 27.89% points for the hourly time series. Finally, this work shows that for the cases analyzed, differentiating the effects of turbulent from non-turbulent flows and modeling with the proper techniques improves forecasting accuracy.

TEMPLATE-BASED MODEL FOR SHORT-TERM FORECASTING OF THE NUMBER OF TRANSACTIONS IN RETAIL CLOTHING STORES

Obtaining predictive values of indicators based on historical data represented by time series plays a crucial role in making business decisions in various industries. One of these areas of application is the task of predicting the number of transactions in retail stores in order to optimally plan the working hours of employees and achieve maximum customer satisfaction with the quality of service. The choice of an appropriate time series forecasting model depends on the forecast horizon, as well as the characteristics of the time series, namely trend, seasonality, cyclicality, and irregularity. Traditional time series analysis and forecasting methods are designed to handle a single seasonality in a time series, but in the presence of multiple seasonality, these methods do not work satisfactorily. The use of time series decomposition methods is characterized by computational complexity. The use of machine learning methods is also not always advisable for a number of different reasons. Thus, it is necessary to use simple adaptive models, based on selected patterns, for recurring seasonal data of complex structure. The main goal of this article is to develop a successful adaptive model and propose methods for using it for short-term forecasting of the number of transactions based on time series data. For estimation purposes, a set of hourly time series of the number of customers (transactions) of some retail chain stores, characterized by multiple seasonality, is used. The results of computational experiments show that the proposed template-based model is quite effective for obtaining short-term predictive values. This model, characterized by simplicity, intuitiveness and a minimum number of tuning parameters, can actually be applied to any area of data represented by time series.

Hourly historical and near-future weather and climate variables for energy system modelling

Abstract. Energy systems are becoming increasingly exposed to the impacts of weather and climate due to the uptake of renewable generation and the electrification of the heat and transport sectors. The need for high-quality meteorological data to manage present and near-future risks is urgent. This paper provides a comprehensive set of multi-decadal, time series of hourly meteorological variables and weather-dependent power system components for use in the energy systems modelling community. Despite the growing interest in the impacts of climate variability and climate change on energy systems over the last decade, it remains rare for multi-decadal simulations of meteorological data to be used within detailed simulations. This is partly due to computational constraints, but also due to technical barriers limiting the use of meteorological data by non-specialists. This paper presents a new European-level dataset which can be used to investigate the impacts of climate variability and climate change on multiple aspects of near-future energy systems. The datasets correspond to a suite of well-documented, easy-to-use, self-consistent, hourly- and nationally aggregated, and sub-national time series for 2 m temperature, 10 m wind speed, 100 m wind speed, surface solar irradiance, wind power capacity factor, solar power factor, and degree days spanning over 30 European countries. This dataset is available for the historical period 1950–2020 and is accessible from https://doi.org/10.17864/1947.000321 (Bloomfield and Brayshaw, 2021a). As well as this a companion dataset is created where the ERA5 reanalysis is adjusted to represent the impacts of near-term climate change (centred on the year 2035) based on five high-resolution climate model simulations. These data are available for a 70-year period for central and northern Europe. The data are accessible from https://doi.org/10.17864/1947.000331 (Bloomfield and Brayshaw, 2021b). To the authors’ knowledge, this is the first time a comprehensive set of high-quality hourly time series relating to future climate projections has been published, which is specifically designed to support the energy sector. The purpose of this paper is to detail the methods required for processing the climate model data and illustrate the importance of accounting for climate variability and climate change within energy system modelling from the sub-national to European scale. While this study is therefore not intended to be an exhaustive analysis of climate impacts, it is hoped that publishing these data will promote greater use of climate data within energy system modelling.

A DEEP LEARNING APPROACH FOR CROP TYPE MAPPING BASED ON COMBINED TIME SERIES OF SATELLITE AND WEATHER DATA

Abstract. Global Earth Monitor (GEM, Horizon 2020) takes advantage of the large volumes of available Earth Observation (EO), weather, climate and other non-EO data to establish economically viable continuous monitoring of the Earth. Within the GEM framework, the development of scalable and cost-effective solutions is being tested on several use-cases, with crop identification being one of them.Crop identification uses a combination of EO and weather data to enable automatic identification of crops. The use case supports operational decisions when managing crops and the monitoring of actual vs. planned or reported agricultural land use (e.g., Common Agricultural Policy monitoring). Satellite data and weather data come at very different temporal and spatial resolutions: Sentinel-2 constellation nominally provides an observation of a field every 5 days at 10 m spatial resolution, while weather data has continuous hourly time series at multi-km spatial resolution. We have designed ad-hoc routines to spatially aggregate satellite data at field level and to systematically compose layers of different time discretization series, so that each EO is associated with a complete time series (of opportune length) of weather variables at daily resolution. For each field, we extract the time series of the median over field pixels of Sentinel-2 L1C bands, cloud mask and cloud probability. For doing this we take advantage of Sentinel Hub's Statistical API (Sinergise, 2020), that enables the retrieval of statistics of band values and derived indices over a specified geographic area and time range. Using meteoblue dataset API (meteoblue, 2017), complete time series of daily weather data (NEMS4 model, meteoblue, 2008) are then associated to each field observation, following the systematic layer composition approach mentioned above. An opportune time series length is defined for each of the 17 weather variables we considered. To handle this kind of multi-dimensional layered data, we use a flexible encoding-decoding framework (FlexMod, designed by TUM as part of GEM project): multiple encoders are designed for features of different time length (namely EO data and weather variables) and are then passed to the decoder via a mediator. Thanks to the flexible design of FlexMod framework, different models and architectures can be easily tested by simply defining new encoders and/or decoders. We present results obtained on a dataset in Slovenia, where crop fields are labelled according to a Hierarchical Crop and Agriculture Taxonomy (HCAT). This taxonomy, based on the EAGLE-Matrix and EU regulations, is the one adopted in the EuroCrops project (Schneider et al. 2021). The classification of field crops takes advantage of Sentinel-2 satellite data and Numerical Weather Prediction model output data. We exploit the potential of FlexMod to test different feature extractors, temporal encoding frameworks and decoders and we present a comparison between results obtained training a long-short term memory (LSTM) implementation (Breizhcrops, Rußwurm et al. 2020) and a Self-attention transformer model (Vaswani et al. 2017), the latter showing the best performances with accuracy 0.904 and Cohen’s kappa 0.824. We moreover investigate the role of weather data by benchmarking results against those obtained with just satellite imagery. To better appraise the influence of the weather data we analyse how perturbing weather data in the testing dataset affects the final results. So far, we obtain in both cases very similar accuracies and Cohen’s kappa. A deeper analysis of crop-specific scores (precision, recall, F1) suggests that the training and testing datasets are too limited in terms of size and crop variability to draw any general conclusion over the role of weather. As future developments, once the EuroCrops datasets are ready, we plan to expand the training and testing dataset to cover a higher variability of climatological areas and increase the numerosity of the so far under-represented crops, in the attempt to draw more general conclusions around the influence of weather and the predictability of specific crop classes. Moreover, given the encouraging scores, we aim to perform crop type mapping at least at European scale, thanks to the availability of the EuroCrops data and the cost-effective big data solutions developed during GEM project.

Sporadic Low Salinity Signals in the Oceanic Mixed Layer Observed by the Kuroshio Extension Observatory Buoy

The hourly measurements of air-sea interaction parameters made at the Kuroshio Extension Observatory (KEO) buoy since 2004 have produced an extensive dataset that can be used to understand the dynamics of the oceanic mixed-layer under extreme weather conditions. Our analysis of the KEO-buoy measurements reveals the presence of sporadic low-salinity signals (SLSSs) with duration of 1–2 days in the mixed-layer when, for example, tropical cyclones (TCs) approach. Here SLSS events are identified within the buoy data by using a flag which is set to unity when the value of normalized salinity falls below −1 (minus one). Application of this definition to the hourly time series of salinity measured at 10 m depth by the KEO buoy has allowed us to identify 244 SLSS events over the 11-year analysis period. Both winter (January-March) and spring (April-June) seasons show a peak in the composite time series of wind velocity subdiurnal variation (WVDV) near the start time of SLSS events. SLSS events can be terminated by the entrainment of saltier subsurface water parcels to the mixed-layer in response to storms. The discovery of the WVDV peaks following SLSS commencement is the reason the present study has adopted a time scale of 1-day wind velocity analysis. This temporal resolution has enabled us to identify a TC-related delay in the development of the WVDV peaks, which may represent a fundamental characteristic of the influence of TCs on these events.

The prediction of Malaysian Borneo tide water level using a Chaotic approach

Predicting sea level is crucial since the rising of sea levels can cause flood, inundation and coastal erosion. This study investigates the potential of the chaotic model to forecast the future sea level in Malaysian Borneo (Sabah). The studied data is an hourly time series of water level collected at the particular benchmark location (station 74003) in the district of Sandakan, Sabah. This study employed 5136 time series data from June 2017 to November 2017. The research goals are to discover the existence of chaotic dynamics by reconstructing the multi-dimensional phase space and Cao method, and to predict future water level data using the local linear approximation approach. The outcome reveals that the correlation coefficient betwixt actual and forecasted data is 0.9228 which is close to 1. It thus reveals the reliability of the prediction method for forecasting the sea level time series data, as well as encouraging indication that the chaotic approach is appropriate to a time series of Malaysian sea level. In implication, these findings are expected to help agencies, particularly the Malaysian Department of Survey and Mapping (JUPEM), organise better sea-level management.

Storage requirements in a 100% renewable electricity system: extreme events and inter-annual variability

In the context of 100% renewable electricity systems, prolonged periods with persistently scarce supply from wind and solar resources have received increasing academic and political attention. This article explores how such scarcity periods relate to energy storage requirements. To this end, we contrast results from a time series analysis with those from a system cost optimization model, based on a German 100% renewable case study using 35 years of hourly time series data. While our time series analysis supports previous findings that periods with persistently scarce supply last no longer than two weeks, we find that the maximum energy deficit occurs over a much longer period of nine weeks. This is because multiple scarce periods can closely follow each other. When considering storage losses and charging limitations, the period defining storage requirements extends over as much as 12 weeks. For this longer period, the cost-optimal storage needs to be large enough to supply 36 TWh of electricity, which is about three times larger than the energy deficit of the scarcest two weeks. Most of this storage is provided via hydrogen storage in salt caverns, of which the capacity is even larger due to electricity reconversion losses (55 TWh). Adding other sources of flexibility, for example with bioenergy, the duration of the period that defines storage requirements lengthens to more than one year. When optimizing system costs based on a single year rather than a multi-year time series, we find substantial inter-annual variation in the overall storage requirements, with the average year needing less than half as much storage as calculated for all 35 years together. We conclude that focusing on short-duration extreme events or single years can lead to an underestimation of storage requirements and costs of a 100% renewable system.

Classification of synoptic and local-scale wind patterns using k-means clustering in a Tyrrhenian coastal area (Italy)

In coastal regions, the complex interaction of synoptic-scale dynamics and breeze regimes influence the local atmospheric circulation, permitting to distinguish typical yet alternative patterns. In this paper, the k-means clustering algorithm is applied to the hourly time series of wind intensity and direction collected by in-situ weather stations at seven locations within 30 km from the western coastline of central Italy, in the proximity of Rome, over the period 2014–2020. The selection of both wind-integral quantities and ad hoc objective parameters allows for the identification of three characteristic clusters, two of which are closely related to the synoptic circulation and governed by persistent winds, blowing from either the northeast or the southeast direction throughout the day. In the latter case, synoptic and mesoscale contributions add up, giving rise to a complex circulation at the ground level. On the contrary, the third cluster is closely related to the sea breeze regime. The results allow the identification of some general information about the low-level circulation, showing that the synoptic circulation dominates in winter and, partly, in spring and autumn, when high ventilation and low recirculation conditions occur. Conversely, during summer the sea breeze regime is more frequent and stronger, generating intense air recirculation. Our analysis permits to discern rigorously and objectively the typical coastal meteorological patterns, only requiring anemological in-situ data.

Dampak Siklon Tropis Savannah Terhadap Karakteristik Hidrogeokimia Aliran pada Mata Air Guntur, Kawasan Karst Gunungsewu

Tropical Cyclones often occur in Indonesia and have disastrous impacts. Until now, no research has focused on time series data related to groundwater discharge and chemistry due to tropical cyclones. Therefore, this study aims to analyze and compare the characteristics of the hourly time series, both discharge and chemistry of the Guntur karst springs. Guntur Spring was affected by the tropical cyclone of Savannah on March 17, 2019. The method used in this study consisted of two parts, namely the analysis of the discharge and chemistry of Guntur Spring. Discharge analysis was performed by correlation test with rain intensity. Chemical analysis was carried out using the Minister of Health Regulation No. 90 of 2002 and expert standards. Other chemical analyzes were performed using a triangular piper, rectangular piper, and stiff diagram of the parameters Na+, K+, Ca2+, Mg2+, SO42-, Cl-, and HCO3-. The results of the relationship between discharge and rain intensity are directly proportional and increase significantly during a cyclone. The results of analysis of the Guntur Springs during Tropical Cyclone Savannah, all parameters are in accordance with the quality standard. In addition, there was a reduction in ion content due to the dilution process of Ca2+, Mg2+, Cl-, HCO3-., DHL, and pH ions compared to non-cyclone, Beton Resurgence, and Gremeng Resurgence ions. The chemical results when the cyclone is different compared to non-cyclone, Gremeng Resurgence and Beton Resurgence with a characteristic decrease in ionic content.

Reprint of: Modelling long-term shoreline evolution in highly anthropized coastal areas. Part 2: Assessing the response to climate change

Here, a methodology to obtain ensemble shoreline change projections at regional scale by combining multi-model projections of wave climate and water levels and the reduced-complexity shoreline evolution model in Alvarez-Cuesta et al. (2021) is presented. In order to account for climate change uncertainty, dynamically downscaled and bias corrected projected waves and storm surge series from five different combinations of global and regional climate models and three potential mean sea-level rise (SLR) trajectories for two representative concentration pathways, are used to force the erosion impact model IH-LANS . The methodology is applied to a 40 km highly anthropized coastal stretch in the Mediterranean coast of Spain. Thirty hourly time series of shoreline evolution between 2020 and 2100 are obtained, each of them linked to one future realization of waves and water levels. From the shoreline time-series analysis, long and short -term processes are unraveled, yielding permanent retreats and beach area losses, contribution of individual physical processes (longshore, short-term cross-shore, and SLR) to shoreline change and non-stationary extreme retreats. The methodology presented herein is intended to be a useful tool for evaluating potential climate change risks while enabling the evaluation and prioritization of adaptation measures.

Modelling long-term shoreline evolution in highly anthropized coastal areas. Part 2: Assessing the response to climate change

Here, a methodology to obtain ensemble shoreline change projections at regional scale by combining multi-model projections of wave climate and water levels and the reduced-complexity shoreline evolution model in Alvarez-Cuesta et al. (2021) is presented. In order to account for climate change uncertainty, dynamically downscaled and bias corrected projected waves and storm surge series from five different combinations of global and regional climate models and three potential mean sea-level rise (SLR) trajectories for two representative concentration pathways, are used to force the erosion impact model IH-LANS . The methodology is applied to a 40 km highly anthropized coastal stretch in the Mediterranean coast of Spain. Thirty hourly time series of shoreline evolution between 2020 and 2100 are obtained, each of them linked to one future realization of waves and water levels. From the shoreline time-series analysis, long and short -term processes are unraveled, yielding permanent retreats and beach area losses, contribution of individual physical processes (longshore, short-term cross-shore, and SLR) to shoreline change and non-stationary extreme retreats. The methodology presented herein is intended to be a useful tool for evaluating potential climate change risks while enabling the evaluation and prioritization of adaptation measures.

INFLUENCE OF SPATIAL AND TEMPORAL RESOLUTION ON TIME SERIES-BASED COASTAL SURFACE CHANGE ANALYSIS USING HOURLY TERRESTRIAL LASER SCANS

Abstract. Near-continuously acquired terrestrial laser scanning (TLS) data contains valuable information on natural surface dynamics. An important step in geographic analyses is to detect different types of changes that can be observed in a scene. For this, spatiotemporal segmentation is a time series-based method of surface change analysis that removes the need to select analysis periods, providing so-called 4D objects-by-change (4D-OBCs). This involves higher computational effort than pairwise change detection, and efforts scale with (i) the temporal density of input data and (ii) the (variable) spatial extent of delineated changes. These two factors determine the cost and number of Dynamic Time Warping distance calculations to be performed for deriving the metric of time series similarity. We investigate how a reduction of the spatial and temporal resolution of input data influences the delineation of twelve erosion and accumulation forms, using an hourly five-month TLS time series of a sandy beach. We compare the spatial extent of 4D-OBCs obtained at reduced spatial (1.0 m to 15.0 m with 0.5 m steps) and temporal (2 h to 96 h with 2 h steps) resolution to the result from highest-resolution data. Many change delineations achieve acceptable performance with ranges of ±10 % to ±100 % in delineated object area, depending on the spatial extent of the respective change form. We suggest a locally adaptive approach to identify poor performance at certain resolution levels for the integration in a hierarchical approach. Consequently, the spatial delineation could be performed at high accuracy for specific target changes in a second iteration. This will allow more efficient 3D change analysis towards near-realtime, online TLS-based observation of natural surface changes.

Do ARMA Models Provide Better Gap Filling in Time Series of Soil Temperature and Soil Moisture? The Case of Arable Land in the Kulunda Steppe, Russia

The adoption of climate-smart agriculture requires the comprehensive development of environmental monitoring tools, including online observation of climate and soil settings. They are often designed to measure soil properties automatically at different depths at hour or minute intervals. It is essential to have a complete dataset to use statistical models for the prediction of soil properties and to make short-term decisions regarding soil tillage operations and irrigation during a vegetation period. This is also important in applied hydrological studies. Nevertheless, the time series of soil hydrological measurements often have data gaps for different reasons. The study focused on solving a problem of gap-filling in hourly time series of soil temperature and moisture, measured at the 30 cm depth using a weighted gravitation lysimeter station while meteorological data were recorded simultaneously by a weather station. The equipment was installed in the Kulunda Steppe in the Altai Krai, Russia. Considering that climate conditions affect soil temperature and moisture content directly, we did a comparative analysis of the gap-filling performance using the three imputation methods—linear interpolation, multiple linear regression, and extended ARMA (p,q) models with exogenous climatic variables. The results showed that, according to the minimum of the mean absolute error, ARMA (p,q) models with optimally selected order parameters, and an adaptive window, had some advantages compared to other single-imputation methods. The ARMA (p,q) model produced a good quality of gap-filling in time series with the mean absolute error of 0.19 °C and 0.08 Vol. % for soil temperature and moisture content, respectively. The findings supplemented the methodology of hydrological data processing and the development of digital tools for the online monitoring of climate and soil properties in agriculture.