Historical Series Data Research Articles

Análise de cluster e regionalização hidrológica para Estados brasileiros

ABSTRACT Streamflow data from gauging stations are essential for effective water resources management. However, some regions in Brazil lack the necessary data. Hydrological regionalization is an alternative technique for obtaining data such regions. However, not all regions in Brazil have defined hydrological regionalization models, including the state of Goiás and the Brazilian Federal District. The objective of this study was to develop a hydrological regionalization methodology based on the separation of hydrologically homogeneous regions and multiple linear regression, using a Geographic Information System (GIS) program. Historical series data were used to calculate reference flows with 90 or 95% duration over time in the watercourse (Q90 and Q95) and the mean flow (Ǭ). Rain gauge station data were used to calculate the mean annual rainfall in each watershed through spatial interpolation by ordinary kriging. Subsequently, the physiographic characteristics of each watershed were calculated. The hydrologically homogeneous regions were delimited based on these data using cluster analysis, which identified seven hydrologically homogeneous regions in Goiás, two of them belonging to the Federal District. Multiple regression allowed the development of seven regionalization models. Models for regions 1, 3, 4, 5, and 7 showed better performance.

Continuity of long-term climate data series after the transition from manual to automatic weather station

In climatology, transitioning from conventional manual weather stations (MWS) to automatic weather stations (AWS) presents challenges in maintaining the homogeneity of long-term data series. At the Skalnaté Pleso Observatory (1778 m a.s.l., High Tatras), manual meteorological measurements have been conducted since 1943 using the same methods and devices at the same location. In 2014, the AWS Physicus was installed to record meteorological data parallelly with manual measurements. This study processed six years of parallel measurements (2017–2022) to derive corrections for AWS monthly atmospheric precipitation totals and maximum (Tmax), minimum (Tmin) and mean air temperatures (Tmean) measured in 2023. Two correction approaches were proposed: monthly regressions (MR) and cumulative distribution functions (CDF), including Generalized Extreme Value (GEV) distribution, normal distribution (GAUSS), and Gamma distribution (GAMMA). Analyses of monthly data revealed an underestimation of precipitation totals in AWS data, with a mean bias error (MBE) of −6.8 mm and a root mean squared error (RMSE) of 20.0 mm. The most suitable correction method was the MR, which decreased MBE to 1.5 mm and RMSE to 11.1 mm. AWS monthly air temperature data were overestimated by 0.1 °C, 0.3 °C, and 0.1 °C for Tmax, Tmin, and Tmean, respectively. For Tmax, correction using the MR and GEV methods reduced the MBE to 0.0 °C and achieved RMSE of 0.1 °C for both. After applying GAUSS, MBE and RMSE decreased to 0.1 °C. The most appropriate correction method for Tmin was the MR resulting in MBE of 0.0 °C and RMSE of 0.3 °C. For Tmean, the MR and GEV methods reduced the MBE to 0.0 °C and the RMSE to 0.1 °C for both methods. These results demonstrate that AWS monthly data, when corrected using the presented methods, have the potential to maintain the continuity of historical climate data series at Skalnaté Pleso.

Declining research productivity and income inequality: A centenary perspective

Research productivity has been sharply declining since 1920. This decline has significant implications for inequality. Using an open-economy Schumpeterian growth model, we show that domestic research productivity influences income inequality through two opposing forces: a positive effect from increased asset returns and a negative effect from the erosion of innovation rents through creative destruction. In contrast, foreign research productivity affects inequality solely through the asset returns channel. By constructing a long historical data series for 21 OECD countries, we find that the asset returns channel has been the dominant driver of inequality over the past century. The reduction in both domestic and foreign R&D productivity accounts for 25% to 35% of the observed downward trend in income inequality over this period.

Estimated dissolved oxygen in a river located in the south of Rio Grande do Sul

The present study aimed to estimate the dissolved oxygen concentrations in the Piratini River Watershed (in Brazilian Portuguese: BHRP), located in the northwest region of the State of Rio Grande do Sul. DO is a crucial parameter in assessing water quality, as it is essential for the maintenance of aquatic ecosystems. To estimate this, global linear regression models were applied, statistical tools capable of evaluating the effect that one or more variables exert on another. The analyzed data come from collections carried out by Environmental Protection Foundation (FEPAM) between the years 2005 and 2013, period that, due to the rigorous protocol followed and sampling periodicity, revealed a robust and consistent data set. For the selection of parameters, we sought those that had the smallest sampling failures in the historical data series, namely: Thermotolerant Coliforms, Total Phosphorus, Ammoniacal Nitrogen and Total Nitrogen, Turbidity, Total Solids and Hydrogenionic Potential. After applying the linear regressions, the best global models generated were able to explain 91.90% of the variations in dissolved oxygen in the BHRP, once all the analyzed parameters were incorporated, consolidating the applicability of the method. It is also conclude that, with the exception of Total Phosphorus, all training periods are within the limits tolerated by legislation.

Six‐decade trend in minute‐scale short‐duration intense precipitation in Beijing and its association with urbanization

AbstractUnderstanding the characteristics of minute‐scale short‐duration intense rainfall and its possible link with urbanization is valuable for addressing urban climate change and evaluating biases in high‐resolution historical precipitation data series. This study analyses the long‐term trends of rainy‐season minute‐scale intense rainfall events at different continuous time periods based on precipitation data from 20 national meteorological stations in the Beijing area covering the period from 1961 to 2021. This study shows that during the whole period analysed, the number of events of intense rainfall lasting 15 min or longer decreased, while the number of shorter‐duration events and the accumulated precipitation increased. The frequency and amount of minute‐scale intense rainfall recorded by urban stations are generally higher than those recorded by rural stations in terms of climatological means. The signal of the urbanization effect in minute‐scale precipitation changes is detectable, with the contribution of urbanization to the trends in average precipitation of 30‐min intense rainfall being 68.8%, whereas the contributions with respect to the frequency of 20‐ and 30‐min intense rainfall and the maximum precipitation of intense rainfall with a duration of 45 min reach 100%. The results of the analysis can provide reference for understanding long‐term changes in short‐duration rainfall and the possibility that they are driven by local anthropogenic activity, as well as bias in the minute‐scale precipitation data series of urban observation sites when they are used in studies of large‐scale intense precipitation changes.

Cropping calendar, agroclimatic and agroecological zoning for rainfed maize (Zea mays L.) under different rainfall scenarios in a semi-arid region of Brazil

Context The productivity of crops is influenced by the available net water balance, which relies on a complex interplay of edaphic and climatic factors. In this sense, rainfall is the leading water input for crop growth and yield, especially in water-scarce regions under semi-arid climates worldwide. Aims This study aimed to define the optimum cropping calendar and the agroclimatic and agroecological zones for rainfed maize in the state of Ceará, which is predominantly located within the Brazilian semi-arid region, assuming different rainfall scenarios. Methods We considered the subdivision of the study area into eight homogeneous precipitation regions according to the regional rainfall pattern. The climatic water balance for each region over two decades was calculated from historical weather data series, assuming three rainy scenarios: dry, regular, and rainy. The agroclimatic zoning was defined through the crop water balance results, soil classification, land use and occupation, slope and temperature, weighted through a multi-criteria decision analysis based on the Analytic Hierarchy Process. Key results The recommended planting period was broader in rainy conditions. However, there may be some dry spells between two 10-day intervals of the calendar, which impair plant growth depending on their intensity. For the dry scenario, the agroclimatic zoning showed that the marginal water deficiency covered 96.55% of the study area. Conclusions Rainfed maize cultivated under the semi-arid climate of Ceará state is viable in rainy conditions, partially restricted in regular years, and not recommended in dry years. Implications This study emphasises the importance of climate monitoring in effective planning for rainfed maize cultivation in semi-arid regions.

Chondrichthyes in the north-western Ionian Sea (central Mediterranean): Species diversity, abundance and spatio-temporal changes

This study provides an updated list of fish in the Class Chondrichthyes sampled during all research projects conducted in the north-western Ionian Sea from 1985 to 2020. A total of 36 Chondrichthyes species belonging to 8 orders and 19 families were detected, most of them distributed over a wide depth range, between shelf and slope. For species assessed by the International Union for Conservation of Nature (IUCN), 8 species are listed as Critically Endangered, 3 as Endangered, 6 Vulnerable and 5 as Near Threatened species. The distribution of cartilaginous fish was examined exploring the historical series of MEDITS data (1995–2020. Changes in the species diversity and in their abundance with time and space were detected and some environmental and anthropogenic drivers that could have caused these changes were analysed. An increase in the presence, number of species and biomass of Chondrichthyes was detected from 1995 to 2020. The GAM analyses seem to indicate that the observed increase in species number and biomass is due to the decrease in fishing capacity and increase in salinity and temperature on the bottom. Furthermore, for the most abundant species, Galeus melastomus, an increase of biomass, size and a deeper distribution over the last 26 years were detected. Significant changes in biomass were also observed for other abundant species in the study area (Scyliorhinus canicula and Raja miraletus). The increase in biomass over time, different from several Mediterranean observations, is discussed in relation to the presence of refuge areas in the north-western Ionian Sea.

Climate zoning: identifying suitable regions for the occurrence of Alternaria Brown spot in tangerine trees in Brazil.

Brazilian citrus farming has been migrating to nontraditional citrus-growing regions, which can be considered a challenge for citrus growers, as not all these areas are suitable for tangerine cultivation. Thus, the mapping of regions exhibiting favorable climatic conditions for Alternaria brown spot has become crucial in the selection of appropriate locations for the establishment of new orchards. This mapping enables the implementation of an avoidance strategy, which entails steering clear of areas where the disease is prevalent, aligning with fundamental principles of disease control. Thus, this study seeks to zone areas with high and low climatic favorability for the occurrence of Alternaria brown spot in tangerine trees in Brazil. Historical climate data series from the NASA-POWER database were used for all municipalities in Brazil. Agrometeorological variables used to determine the development of Alternaria brown spot were average monthly air temperature (Tmean) and duration of leaf wetness period (LWD). Areas were considered unsuitable climatically when Tmean was <17 °C or >33 °C, relatively suitable when Tmean was between 13 °C and 33 °C and LWD <10%, and climatically suitable when Tmean was between 13 °C and 33 °C and LWD >10%. The states of Paraná, Santa Catarina and Rio Grande do Sul showed greater thermal amplitude within months and throughout the year. The southern region of the country has harsher winters, with minimum temperatures below 15 °C, which is unfavorable for the disease incidence. The states with the greatest favorability for Alternaria brown spot were Paraná, Santa Catarina, and Rio Grande do Sul, mainly from May to September. Rio Grande do Sul was the state in this region that showed the greatest favorability, as a consequence of leaf wetness exceeding 10 h. The main tangerine-producing regions in Brazil, including the southern part of Minas Gerais, the state of São Paulo and the metropolitan region of Porto Alegre, were mostly classified as relatively favorable areas for the occurrence of Alternaria brown spot. It is recommended that when establishing new tangerine orchards, regions with lower favorability for the occurrence of Alternaria brown spot, such as the North and Central-West regions of Brazil, particularly the states of Amazonas, Pará and Mato Grosso, should be selected. © 2023 Society of Chemical Industry.

Stock Market Analysis using Long Short-Term Model

In today's world of value and improved investments, financial analysis has become a difficult task. The implementation of recurrent neural networks (RNN) and long short-term memory (LSTM) cells for stock market forecasting using time series of historical portfolio stock data is demonstrated in this study. In this study, we applied LSTM to predict stock market values using Yahoo Finance data along with Python modules Pandas and Matplotlib to evaluate the performance of the model. Our results show that the LSTM model is able to make accurate predictions of stock market prices and trends using historical data. The results of the correlation study showed a significant relationship between the daily return and the closing price of four randomly chosen companies. Overall, using LSTM, Yahoo Finance, Python Pandas, and Matplotlib modules to predict stock prices and provide useful information to investors was a successful strategy.

Ecological gradients in a marine cave revisited 26 years after

Marine caves are unique and vulnerable habitats threatened by multiple global and local pressures. Changes in the sessile epibenthic community of the Grotta Azzurra of Capo Palinuro (Salerno, Italy) during the last 26 years have been evaluated using growth forms and trophic guilds as non-taxonomic descriptors, informative of the structure and the functioning of the cave ecosystem, respectively. Photographic samples were collected in 1992 and 2018 to investigate the stability of the ecosystem response to the three main environmental gradients occurring inside the cave: i) confinement, ii) depth, and iii) light. Benthic communities were well spatially distributed along the three gradients inside the cave according to both the morphology and the feeding strategy of organisms, although the zonation was less marked considering trophic guilds. Gradients persisted and zones remained distinct through time, notwithstanding significant changes in community composition. The trophic organization was preserved throughout the cave, but significant changes in the spatial organization of benthic assemblages were widely observed, mainly due to 3-dimensional growth forms replaced by 2-dimensional (i.e., encrusting) growth forms and by increased sedimentation, thus causing the structural homogenization of the cave community. The outermost and the deepest sectors of the cave experienced higher changes compared to the most confined, blind-ended, sector and to the shallowest depths, which resulted comparatively more stable. From a functional point of view, filter feeders always dominated the communities (even considering their shift from erect to encrusting morphologies) and the few observed changes were due to the reduction of photophilic autotrophs and active ciliate without lophophore and the increase of mixotrophic sponges. As local human pressures remained low through the period considered, the changes observed are likely to be mainly due to global warming, which caused increased sea surface water temperatures and repeated summer heat waves (reported in 1994, 2003, 2008, 2012, and 2015). Regular monitoring programs on marine caves should be promoted to create historical series of data precious to better understand magnitude and pattern of change in the long-term evolution, and to estimate the recovery time of such priority habitats.

Climate risk zoning for wheat crops in the southeastern region of Brazil.

Wheat (Triticum aestivum L.) is the second most consumed food in the world. One way to meet this demand is the expansion of wheat cultivation to the Brazilian Cerrado in the southeastern region. However, one of the major limitations is that there are few studies related to wheat climate risk zoning. Thus, this study aimed to determine the agroclimatic zoning of wheat by estimating the water needs satisfaction index (ISNA) in the southeastern region of Brazil. For this purpose, a 60-year historical series of meteorological data was used to calculate the potential evapotranspiration, crop evapotranspiration, and climatological water balance values. To define the agroclimatic zones of wheat and sowing date, the ISNA method was used. The data were analyzed using descriptive statistics to determine the variations. To obtain the agroclimatic zoning of wheat, the geostatistical method of kriging interpolation was used. The regions with the highest rainfall are the south of Minas Gerais and the coast of São Paulo. The sowing period directly impacts the development of the crop, the available water capacity and the ISNA values indicated the spring and summer had better cultivation conditions, and the best window for wheat cultivation is concentrated in the fall due to the limitation of biotic factors. In terms of altitude (>700 m), Minas Gerais has 39.4% of the area suitable for wheat cultivation. Thus, climatic variations within and between the states of the southeastern region should be considered for the positioning of wheat cultivars in these regions to obtain the maximum yield. © 2023 Society of Chemical Industry.

Impact of electric vehicle emissions in the Brazilian scenario of energy transition and use of bioethanol

The transport sector stands out on the world stage due to its significant share of greenhouse gas emissions, which currently equate to 25% of global emissions. In the global context of energy transition, the electrification of mobility is shown as one of the alternatives to mitigate the impacts of global warming. In Brazil, the conjuncture is of extensive use of renewable energy sources, in which bioethanol accounts for about 35% of the primary source in flex-fuel vehicles and the electricity generation matrix comprises about 83% of renewable sources. This study aims to evaluate the global warming potential (GWP) resulting from the insertion of electric vehicles in the long term (2023–2050), in scenarios of the Brazilian light vehicle fleet, considering the transition of the electric generation matrix and the use of bioethanol. In this sense, statistical methodologies applied to historical data series and regression algorithms were used to estimate the fleet. This is in addition to the life cycle assessment approach applied to the manufacture and use of vehicles, and also fuel production in order to estimate the global warming potential. The results of the comparative evaluation showed that the ICE and HEV vehicles fueled by bioethanol had the lowest emissions per kilometer driven, compared to purely electric vehicles. In the long term, the scenario with a greater share of bioethanol proved to be more favorable for reducing annual emissions, presenting annual emissions of 69 Mt CO2eq/year in 2050 compared to 81 Mt CO2eq/year in 2020. The results of this study will provide fundamental information for planning to mitigate the environmental impacts of the transportation sector.

Historical control data of ecotoxicological test with the copepod Tigriopus fulvus

ABSTRACT The aim of this study was to evaluate the variability of naupliar mortality and sensitivity of T. fulvus to copper sulphate, (positive control), reared in laboratory by using a series of historical control data (HCD). Historical data were obtained by a dataset of 68 mortality acute tests (negative control) and by about 30 values of LC50 determined at 24–48 and 96 h (positive control with copper sulphate), in over 12 years. Data, arranged in Shewhart-like control charts, displayed a mean mortality of 3.53% in negative controls, well below the reference limit of ≤10%, with a low level of variability over time. Lethal concentration (LC50) for copper sulphate showed a stable mean of 0.67 ± 0.12 at 24 h, 0.30 ± 0.08 at 48 h and 0.12 ± 0.05 mg Cu++/L, over time. These data confirm Tigriopus fulvus as a good bioindicator, even after years of culturing in laboratory conditions, since even over time, its responses to negative and positive controls (copper) have not undergone changes.

The role of migration networks in the development of Botswana's generalized HIV epidemic.

The majority of people with HIV live in sub-Saharan Africa, where epidemics are generalized. For these epidemics to develop, populations need to be mobile. However, the role of population-level mobility in the development of generalized HIV epidemics has not been studied. Here we do so by studying historical migration data from Botswana, which has one of the most severe generalized HIV epidemics worldwide; HIV prevalence was 21% in 2021. The country reported its first AIDS case in 1985 when it began to rapidly urbanize. We hypothesize that, during the development of Botswana's epidemic, the population was extremely mobile and the country was highly connected by substantial migratory flows. We test this mobility hypothesis by conducting a network analysis using a historical time series (1981-2011) of micro-census data from Botswana. Our results support our hypothesis. We found complex migration networks with very high rates of rural-to-urban, and urban-to-rural, migration: 10% of the population moved annually. Mining towns (where AIDS cases were first reported, and risk behavior was high) were important in-flow and out-flow migration hubs, suggesting that they functioned as 'core groups' for HIV transmission and dissemination. Migration networks could have dispersed HIV throughout Botswana and generated the current hyperendemic epidemic.

Landsat-7 ETM+, Landsat-8 OLI, and Sentinel-2 MSI Surface Reflectance Cross-Comparison and Harmonization over the Mediterranean Basin Area

In the Mediterranean area, vegetation dynamics and phenology analysed over a long time can have an important role in highlighting changes in land use and cover as well as the effect of climate change. Over the last 30 years, remote sensing has played an essential role in bringing about these changes thanks to many types of observations and techniques. Satellite images are to be considered an important tool to grasp these dynamics and evaluate them in an inexpensive and multidisciplinary way thanks to Landsat and Sentinel satellite constellations. The integration of these tools holds a dual potential: on the one hand, allowing us to obtain a longer historical series of reflectance data, while on the other hand making data available with a higher frequency even within a specific timeframe. The study aims to conduct a comprehensive cross-comparison analysis of long-time-series pixel values in the Mediterranean regions. For this scope comparisons between Landsat-7 (ETM+), Landsat-8 (OLI), and Sentinel-2 (MSI) satellite sensors were conducted based on surface reflectance products. We evaluated these differences using Ordinary Least Squares (OLS) and Major Axis linear regression (RMA) analysis on points extracted from over 15,000 images across the Mediterranean basin area from 2017 to 2020. Minor but consistent differences were noted, necessitating the formulation of suitable adjustment equations to better align Sentinel-2 reflectance values with those of Landsat-7 or Landsat-8. The results of the analysis are compared with the most-used harmonization coefficients proposed in the literature, revealing significant differences. The root-mean-square deviation, the mean difference and the orthogonal distance regression (ODR) slope show an improvement of the parameters for both models used (OLS and RMA) in this study. The discrepancies in reflectance values leads to corresponding variations in the estimation of biophysical parameters, such as NDVI, showing an increase in the ODR slope of 0.3. Despite differences in spatial, spectral, and temporal characteristics, we demonstrate that integration of these datasets is feasible through the application of band-wise regression corrections for a sensitive and heterogeneous area like those of the Mediterranean basin area.

Rainfall analysis over 31 years of Chintapalle, Visakhapatnam, High Altitude and Tribal zone, Andhra Pradesh, India

Climate change and variability, particularly which of the annual rainfall, has received a great deal of interest to researchers worldwide. The extent of the variability of rainfall varies according to locations. Consequently, investigating the dynamics of rainfall variable in the perspective of changing climate is important to evaluate the impact of climate change and adapt potential mitigation strategies. To gain insight, trend analysis has been employed to inspect and quantify the rainfall distribution in the Chintapalli, Visakhapatnam district of Andhra Pradesh, India. Thirty-one years for a period of 1990–2020 long historical rainfall data series for different temporal scales (Monthly, Seasonal and Annual) of the study region was used for the analysis. Statistical trend analysis techniques namely Mann–Kendall (MK) test was used to detect the trend. To compute trend magnitude, Theil–Sen approach (TSA) was used for calculation of Sen’s slope. The detailed analysis of the data for 31 years indicates positive increasing trend with 2.13mm per year derived from the linear regression. MK test detected that there were rising and falling trends for various time scales in the study area. Departure analysis of rainfall indicated that a possible chance of normal rainfall, more frequently in the area. Rainfall Anomaly Index (RAI) analysis revealed that normal for most of the years, however, 2002 is the very dry year. While last ten years, the frequency of drought occurrence is thrice, but the magnitude is low. The study results will help in persuading the rainfall risks with effective use of water resources which can increase crop productivity and likely to manage natural resources for sustainability at HAT zone of Andhra Pradesh.

Impacts of a tailings dam failure on water quality in the Doce river: The largest environmental disaster in Brazil

BackgroundOn November 5, 2015, the Fundão Dam, owned by Samarco Mineração SA, experienced a catastrophic failure, releasing approximately 55 to 62 million m3 of mining ore waste into the Doce River. This event had severe consequences as the Doce River serves as a vital source of water and food for numerous communities. ObjectiveEvaluate the spatiotemporal distribution of water parameters before, during, and after the failure of the Samarco tailings dam, with the goal of support the management and restoration of water quality. The study utilizes a long historical series of data from 1997 to 2018. MethodologyIn this study, data from 13,916 measurements of variables including HCO3−, Mg2+, Ca2+, SO42−, pH, EC, and metal(oids) such as Fe, Mn, As, Cu, Ba, Pb, Zn, Cd, Cr, and Ni were compiled. These measurements were collected from 1997 to 2018 to investigate the impact of the dam rupture. ResultsThe dam failure resulted in an increase in high turbidity events and total suspended solids (TSS), leading to a decline in water quality. The Kruskal-Wallis test revealed a homogenization effect along the river watershed caused by the accident, particularly for Pb, Ni, Cd, Cr, and coliforms. Additionally, there was a change in the seasonal distribution of PO43−, NH4+, Na+, SO42−, and Zn. The factor analysis identified four main factors: i) HCO3−, Mg2+, Ca2+, SO42−, pH, and electrical conductivity (37.55%) were indicative of weathering processes from lithotypes; ii) Fe+Mn and the metal(oids) (As, Cu, Ba, Pb, Zn, Cd, Cr, and Ni) (15.89%) were associated with ore waste and regional geology; iii) total coliforms, NO3−, NH4+, and PO43− (9.39%) were linked to anthropogenic influences in the river; and iv) K+, Na+, and Cl− (8.22%) represented the impact of domestic and industrial sewage discharge as well as the marine environment in the lower sector of the Doce River. ConclusionThe dam rupture significantly altered water quality in the Doce River, particularly in relation to metal(oids), three years after the disaster. The presence of ore waste in the bottom sediments creates an environmental liability as it can be remobilized by human activities or extreme weather events, further impacting water quality.

Advanced well stimulation selection with gradient boosting

To ensure the required level of production of brown fields, it is necessary to plan and implement an efficient program of well stimulation in a timely manner. This program has a significant impact on the further development of the oil field in terms of its productivity and economics. Nowadays, the decisions about well stimulations are made by experts mostly based on their experience and the company’s heuristics. In the work, a novel approach to fast, accurate, and computationally efficient selection of candidates for well stimulations was presented. We explored the ability of Machine Learning algorithms to solve the problem. The predictive pipeline was built on the basis of production and pressure time series of historical data. We developed and applied a comprehensive prepossessing workflow to a real field data to prepare a training data set. Finally, two Gradient-Boosting models were developed, tuned, trained and validated. The first model was used for prediction of the necessity for the well stimulation and the second model—for identification of the required treatment type. Blind test of the models was provided. It resulted in a 0.80 recall score and 0.79 balanced accuracy score.

A New Hybrid Forecasting Model Based on Dual Series Decomposition with Long-Term Short-Term Memory

In recent years, ozone (O3) has gradually become the primary pollutant plaguing urban air quality. Accurate and efficient ozone prediction is of great significance to the prevention and control of ozone pollution. The air quality monitoring network provides multisource pollutant concentration monitoring data for ozone prediction, but ozone prediction based on multisource monitoring data still faces the challenges of each station’s series of data. Aiming at the problems of low prediction accuracy and low computational efficiency in traditional atmospheric ozone concentration prediction, ozone concentration prediction using dual series decomposition was proposed by variational mode decomposition (VMD), ensemble empirical mode decomposition (EEMD), and long short-term memory (LSTM). First, the historical data series of Nanjing air quality monitoring stations is decomposed by VMD, and then the EEMD algorithm is applied to the residual of VMD to obtain several characteristic intrinsic mode function (IMF) components; each characteristic IMF component is trained by LSTM to obtain the prediction result of each component, and then the final result can be obtained by linear superposition. The proposed method achieved the best results with R2 = 99%, MSE = 5.38, MAE = 4.54, and MAPE = 3.12. Because LSTM has strong adaptive learning ability and good memory function, it has the learning advantage of long-term memory for long-term data, and the prediction results are more accurate. According to the data, the proposed method is superior to the baseline models in terms of statistical metrics. As a result, the proposed hybrid method can serve as a reliable model for ozone forecasting.

Energy impact assessment of electric vehicle insertion in the Brazilian scenario, 2020 – 2050: a machine learning approach to fleet projection

The transportation sector stands out on the world stage as one of the largest consumers of energy and the trend towards electrification of the light vehicle fleet is proving to be challenging, given the impacts generated by growing demand of energy and mainly in the generation of electricity. In Brazil, the electricity generation mix comprises around 83% of renewable sources, basically composed of hydraulic, wind and biomass sources, making it reasonable to consider the electrification of the transportation sector as one of the measures to be adopted in the promotion of sector sustainability. This article contributes to energy planning through the long-term projection of the Brazilian fleet of light vehicles, and the simulation of their impacts on energy demand, in three scenarios of insertion of plug-in electric vehicles. In this sense, statistical methodologies applied to historical data series and supervised Machine Learning algorithms for curve fitting were used, as well as the well-to-wheels approach in the fuel life cycle to estimate the energy demand. The projections included three scenarios for the insertion of electric vehicles with annual increases in demand for electricity generation which would correspond to 0.6%, 0.9% and 4.2% of additional electricity generation demand in the year 2050, with reference to the year 2020. The electrification of the fleet also proved to be advantageous in terms of reducing the use of fossil fuels and promising in terms of sustainability in the transport sector. The results of these studies will provide fundamental information for energy planning, as well as for public policy decision-making.