Resumo A intensificação da demanda hídrica decorrente do crescimento populacional e econômico exige a adoção de ferramentas eficazes para gestão dos recursos hídricos. A previsão de vazões, por meio da modelagem hidrológica, é essencial para subsidiar a alocação racional da água e mitigar impactos em múltiplos setores. Modelos determinísticos exigem grande volume de dados e conhecimento detalhado da bacia. Em contrapartida, modelos estocásticos como SARIMA e técnicas de aprendizado de máquina, como Redes Neurais Artificiais (RNA), oferecem alternativas viáveis, diante da escassez de dados. Assim, este estudo avaliou o desempenho preditivo dos modelos SARIMA e Multi-Layer Perceptron (MLP), aplicados às séries temporais de vazão média mensal, na Bacia Hidrográfica do Rio Itiquira, que pertence à Bacia do Rio Paraguai, conhecida por abrigar o Pantanal. A MLP apresentou desempenho superior no período de teste, com RMSE de 7,95 m3/s, NSE de 0,841, KGE de 0,906 e PBIAS de 4,05% para um horizonte de 9 meses, enquanto o SARIMA obteve RMSE de 11,291 m3/s, NSE de 0,679, KGE de 0,770 e PBIAS de 19,60%. Os resultados evidenciam a elevada capacidade da MLP em modelar padrões sazonais, reforçando seu potencial como ferramenta de suporte à gestão hidrológica em regiões com dados limitados.
Abstract The urban heat island (UHI) is a well-known consequence of urbanization on local climate. This study investigates the UHI in the Metropolitan Area of São Paulo (MASP), focusing on urban morphology and meteorological conditions. We utilized the Weather Research and Forecasting (WRF) model, incorporating the Local Climate Zones (LCZ) classification, to analyze simulations for September from 2014 to 2023. Results reveal a distinct southeast-to-northwest temperature gradient, where locations in the same LCZ class showed different temperatures based on their position. Proximity to the coast on the southeastern side provided cooling effects from the sea and valley-mountain breezes. UHI intensity was greater at night under clear skies and calm winds, peaking around 18:00 LT. The choice of rural reference point had a significant impact on the measured UHI magnitude. Furthermore, densely built-up areas (LCZ 1-3) retained more heat overnight compared to more open zones (LCZ 6, 9). These findings underscore the importance of comprehensive UHI assessments, which must consider urban morphology, meteorological conditions, regional circulation patterns, and the careful selection of rural reference points.
Resumo A temperatura do ar e o fotoperíodo são duas variáveis meteorológicas que desempenham um papel fundamental no desenvolvimento vegetativo e reprodutivo de espécies florestais por influenciar o aparecimento de estruturas vegetativas e reprodutivas. O objetivo deste trabalho foi avaliar a resposta de seis métodos de cálculo dos graus-dia e a influência do fotoperíodo na variável filocrono em duas espécies florestais nativas na fase de muda. Foi instalado um experimento a campo, sob delineamento inteiramente casualizado, em esquema fatorial 2 × 12, sendo duas espécies florestais nativas (pau ferro - Caesalpinia ferrea Mart. ex. Tul. Var. leiostachya Benth e angico vermelho - Anadenanthera macrocarpa (Benth) Brenan) e doze épocas de semeadura, com cinco repetições por tratamento. A variável filocrono é influenciada pelos métodos de cálculo dos graus-dia, sendo o melhor método, aquele que considera as três temperaturas cardinais. Foi verificada também a influência do fotoperíodo nos valores de filocrono, em que os menores valores para as duas espécies concentraram-se em épocas de temperaturas mais elevadas e maior fotoperíodo, instaladas em 20/10, 20/11 e 20/12.
Abstract Taylor diffusion model and forms for the autocorrelation functions are employed to derive turbulent dissipation rates. These formulations contain a numerical coefficient nc and are written in terms of turbulent scales associated with the energy-containing eddies. The derivation generates negative terms in the dispersion parameters that decrease the turbulent dispersion process. The used methodology allows one to connect the negative terms with the inertial range frequencies. The introduction of the inertial range autocorrelation functions in the Taylor diffusion model provides a dispersion parameter with a negative term. Therefore, the negative terms of the dispersion parameters are compared with the negative term of the inertial range dispersion parameter. The comparison provides equations for the turbulent dissipation rates containing the Kolmogorov constant and numerical coefficients. The analysis comparing observed and derived dissipation rate magnitudes selects a magnitude of nc = 2 for the numerical coefficient. Therefore, this value of nc = 2, originated by the choice of the exponential autocorrelation function, seems to be the magnitude that should be used in turbulence models. The present investigation exhibits the non-universal character of the dissipation coefficient and establishes magnitudes for this coefficient that are in accordance with values discovered in the literature.
Abstract Reference evapotranspiration (ETo) has a wide application in agriculture and the Penman-Monteith FAO-56 (PM) method is considered standard to calculate it, but this model requires several meteorological data, which are not always readily available. As an alternative, other models that require less data and according to the climatic characteristics of the region are chosen. Thus, the objective of this work was to carry out a comparative study of 12 methods of daily ETo with the PM standard method for 584 automatic weather stations grouped by climate types in Brazil and to evaluate the performance of the different methods according to the climate. To evaluate the performance of the models, the coefficient of determination (R2), the root mean square error (RMSE, mm dia−1), the Kling-Gupta efficiency index (KGE) and percent bias (PBIAS) were used. Overall, the Turc method had the best performance in relation to the other methods, followed by the Penman FAO-24, Makkink, Priestley-Taylor, Hargreaves-Samani and Tanner-Pelton methods. In addition, it is concluded that for the most appropriate choice of a model it is necessary to consider the climatic characteristics of the studied region and the availability of meteorological data.
Abstract Maize plays a fundamental role in the economy of Rio Grande do Sul. However, the occurrence of droughts has impacted the maize yield. To mitigate these effects, investment in irrigation and the adoption of efficient strategies for irrigation management are essential. Thus, the aim of the study was to evaluate the water balance and maize yield data, analyze their relationship with the ENSO phenomenon, and propose region-specific irrigation strategies for Rio Grande do Sul. This study was developed based on historical series of maize yield and meteorological data from weather stations in ten municipalities. Three irrigation management strategies were studied. The interannual variability in maize yield across the studied cities can be attributed to fluctuations in water availability according water balance during the crop cycle. In El Niño years, maize yield was above average in 85% of the years, while in La Niña years yield below the mean yield in 78% of the cases. The adoption of deficit irrigation can be an alternative in scenarios where water efficiency is prioritized while maintaining yield levels similar to full irrigation. Adopting irrigation practices in the state of Rio Grande do Sul can result in an average yield increase of 21.8 kg mm-1.
Abstract Solar energy emerges as a promising sustainable source of energy, capable of meeting energy demand and mitigating issues related to climate change. To evaluate its availability, we can use empirical models such as the Angstrom-Prescott (AP) equation. In this context, our objective was to calibrate and assess the AP model to estimate global solar radiation (Rs) in Minas Gerais. We utilized solar radiation and insolation data from seven municipalities in the state, extracted from the National Institute of Meteorology (INMET), calibrating the model's coefficients through linear regression. After calculating Rs with the calibrated linear and angular coefficient values, we evaluated the accuracy of the estimative by comparing it with observed data. For this, we used the following statistical indicators: linear, angular, determination, and correlation coefficients; mean absolute error; root mean square error; concordance index; and performance index. The results indicated that the model can be an effective tool for estimating Rs in the region (performance index above 0.80, except in one municipality). The use of the calibrated coefficients presented in this work or the standard method are both effective and show no difference between them. The Angstrom-Prescott model proved to be effective in estimating solar radiation in Minas Gerais, with results highlighting its accuracy and applicability.
Abstract Recent technological advancements and data assimilation techniques have elevated the precision of weather reanalysis, creating detailed historical atmospheric gridded datasets by combining diverse observational data and numerical models. This study assessed the ERA5-Land and NASA Power climate reanalysis products in tropical lowlands, using data from 44 ground weather stations in Colombian oil palm production zones from 2008 to 2022. Validation results varied based on variables, locations, and temporal resolutions, with each product showing distinct strengths. ERA5-Land demonstrated superior performance in temperature estimation, while NASA Power excelled in solar radiation estimates. Despite the tropical setting, proximity to mountainous terrain created challenging microclimates, impacting reanalysis accuracy particularly for rainfall and relative humidity. Temporal data aggregation from daily to monthly and annual scales revealed varying patterns of improvement across variables, with some metrics showing enhanced performance at coarser temporal resolutions while others remained relatively unchanged. The study emphasizes the importance of careful product selection based on the specific variable and temporal scale of interest, especially in regions with complex microclimates or unique environmental conditions. While reanalysis data offer valuable insights, their limitations must be acknowledged to ensure accurate interpretation and application in research endeavors, particularly in tropical agricultural contexts.
Abstract This study evaluates the performance of the Soil and Water Assessment Tool (SWAT) hydrological model in simulating daily river flow and quantifying flow components within the Isser catchment, upstream of the Koudiat Acerdoune dam in Algeria. Model parameters were optimized using the SWAT-CUP optimization method and the SUFI-2 algorithm. The model was calibrated (2002-2012) and validated (2013-2018) against observed daily streamflow data, following a three-year warm-up period (1999-2001). Sensitivity analysis identified key parameters influencing streamflow, including groundwater delay, baseflow processes, channel hydraulic conductivity, and surface roughness. The SWAT model demonstrated strong performance, achieving satisfactory to very good performance metrics (e.g., R2 Cal/Val = 0.86/0.82; NSE Cal/Val = 0.81/0.74), indicating its ability to accurately simulate daily streamflow in the Isser watershed. Water balance analysis revealed that actual evapotranspiration accounted for 51% of precipitation. Shallow aquifer recharge constituted a major loss at 28%, while total water yield represented the remaining 21%. Furthermore, the model estimated that surface runoff accounted for approximately 47.4% of the total water yield, lateral flow for 11.6%, while baseflow contributed the remaining 41%. These findings demonstrate the SWAT model's capability to effectively represent the hydrological processes governing the Isser watershed. The study underscores the importance of considering both surface and subsurface water dynamics to guide sustainable water resource planning in semi-arid regions.
Resumo O mapeamento e a análise espacial das mudanças no uso e cobertura da terra (UCT) são essenciais para monitorar dinâmicas ambientais e gerir recursos hídricos. Estudos que avaliam a evapotranspiração (ET) considerando alterações no UCT ajudam a quantificar os impactos antrópicos no ciclo hidrológico. Redes Neurais Artificiais (RNA) têm sido usadas para estimar a ET, mas dependem de séries meteorológicas. Portanto, o objetivo deste estudo é estimar a ET real utilizando uma RNA Multilayer Perceptron (RNA-MLP), para a área de Proteção Ambiental de Itupararanga, região de importância hídrica para a Região Metropolitana de Sorocaba, utilizando dados de UCT e bandas espectrais do satélite Landsat. A ET de 2021 foi calculada com o algoritmo SEBAL e utilizada como base para treinamento e classificação supervisionada no modelo RNA-MLP do software TerrSet. As estimativas da RNA foram avaliadas com critérios do TerrSet, destacando diferenças visuais que indicam a necessidade de ajustes nos parâmetros do MLP para maior aproximação das estimativas do SEBAL, previamente validadas. Esta abordagem inovadora complementa o uso de RNA na modelagem da ET, oferecendo novas perspectivas para a análise de dinâmicas ambientais.
Resumo Os eventos extremos de calor são uma realidade no Brasil, impactando os pavimentos asfálticos e os ambientes. Esta pesquisa analisou esses eventos em 10 diferentes locais do Brasil, no período de 1961-2023, por meio da temperatura máxima média anual por 7 dias consecutivos; e da ocorrência, duração e intensidade das ondas de calor. Os resultados demonstram tendências de aumento dos valores anuais médios dos 7 dias consecutivos mais quentes ao longo do período de 1961-2023, com maior destaque para os anos mais recentes. De maneira geral, identificou-se um aumento do número de ondas de calor por ano ao longo do tempo, com acréscimos expressivos nas cidades das regiões Norte, Nordeste e Centro-Oeste, especialmente após os anos 2000. Também, pode-se destacar que as ondas de calor mais intensas e duradouras ocorreram nos últimos anos. Além disso, observou-se que, muitas vezes, as temperaturas extremas do ar registradas durante as ondas de calor são acompanhadas por valores expressivos de radiação solar. A maior quantidade de eventos extremos de calor, aliados a maior intensidade e duração, demandam mais dos materiais asfálticos em termos de desempenho, afetando os custos da infraestrutura rodoviária brasileira.
Resumo A evapotranspiração de referência (ETo) é uma componente importante do ciclo hidrológico, influenciada por diversos elementos meteorológicos. Este estudo objetivou identificar variáveis meteorológicas que mais impactam ETo no Brasil, realizando uma análise de sensibilidade da equação de Penman-Monteith FAO-56 para as zonas climáticas Tropical, Subtropical e Semiárida. Utilizaram-se dados diários de 584 estações meteorológicas automáticas do Instituto Nacional de Meteorologia. Cada elemento meteorológico, temperaturas máximas (Tmax) e mínima (Tmin) do ar, radiação solar global (Qg), umidade relativa média (URmed) e velocidade do vento (U2), foi variado individualmente em dez intervalos iguais entre seus valores máximos e mínimos, mantendo os demais constantes nos valores médios. O índice de Sensibilidade (IS) quantificou a influência de cada variável sobre ETo. Os resultados mostraram que Qg e Tmax foram os principais fatores de influência positiva sobre ETo nos climas Tropical e Subtropical. No clima Semiárido, além de Tmax e Qg, U2 também apresentou influência significativa, enquanto a URmed apresentou predominantemente índices de sensibilidade negativos. Conclui-se que a influência dos elementos meteorológicos na ETo varia significativamente entre as regiões climáticas brasileiras, ressaltando a importância de considerar as características climáticas na modelagem da evapotranspiração, priorizando a medição de variáveis meteorológicas que exercem maior impacto na estimativa da ETo.
Abstract The study assesses the impact of diverse soil management techniques on water storage and soybean yield, considering changes in precipitation patterns in Rondônia and the potential water stress on the soybean crop due to alterations in land use and soil cover. The experiment, conducted in Vilhena-RO during the 2021/2022 harvest, included treatments with two management techniques (no tillage and minimal tillage) and soybean cultivation under three soil cover conditions (no cover, no-till, and intercropped with Brachiaria brizantha cv. marandu). Agrometeorological monitoring collected data such as precipitation, temperature, relative humidity, solar radiation, soil temperature, and humidity. Analysis of volumetric soil moisture indicated similar variability between treatments, but conditions without tillage showed greater water retention, associated with soil structuring. No statistical differences were observed between the soil cover conditions. The study underscores the importance of the soil's physical conditions in regulating surface water storage for soybean cultivation in Vilhena-RO.
Abstract The increased consumption of natural resources, such as water, has become a global concern. Consequently, determining information that can minimize water consumption, such as evapotranspiration, is increasingly necessary. This research evaluates the capacity of Genetic Algorithms (GAs) in training and fine-tuning the parameters of Artificial Neural Networks (ANNs) (MLP-GA) to obtain daily values of reference evapotranspiration (ETo) in accordance with the Penman-Monteith FAO-56 method. The method is employed to estimate ETo at 14 weather stations in Brazil. The findings are assessed based on the coefficient of correlation (r), mean absolute error (MAE), root mean square error (RMSE), and mean percentage error (MPE), and are contrasted with the Hargreaves-Samani, Jensen-Haise, Linacre, Benavides & Lopez, and Hamon methods, along with the Multilayer Perceptron (MLP) neural network, which is conventionally trained and employs hyperparameter tuning techniques such as Grid Search (MLP-GRID) and Random Search (MLP-RD). The results show that the MLP-GA is, on average, 12 times faster than MLP-RD and 60 times faster than MLP-GRID, while achieving the highest precision indices in most regions, with an r of 0.99, MAE ranging from 0.11 mm to 0.20 mm, RMSE between 0.14 mm and 0.27 mm, and MPE between 2.49% and 7.09%. These findings suggest the results generated achieve an precision between 92.91% and 97.51% in comparison to the Penman-Monteith method. This confirms that employing Genetic Algorithms (GA) to automate the training and optimization of the model is effective and enhances the neural network's capacity to predict ETo.
Abstract Anchieta-Imigrantes System (AIS) is one of Brazil's primary highways, connecting São Paulo to the Port of Santos. In AIS, Operação Comboio (OC) is activated when visibility drops below 100 m. This study investigated the meteorological factors that initiated OC in AIS during 2015 using a comprehensive array of data sources, for example, meteorological stations and synoptic charts. In 2015, OC was triggered on 106 days, predominantly in spring, mostly on the Anchieta Highway, and often in the afternoon. Meteorological systems triggering OC were classified as orographic, confined to the Serra do Mar, or non-orographic, involving broader cloudiness or precipitation detected by satellite or radar. Data from Itutinga Meteorological Station helped differentiate between visibility reductions caused by fog or precipitation. Among the 87 OCs studied in 2015, sea breeze (SB) contributed to 38% of cases. Generally, moisture was provided by advection by the SB or the Post-frontal anticyclone, occasionally from precipitating systems associated with frontal passage or thermodynamic instability. Moisture influx combined with temperature drops, induced by SB, post-frontal air, or nocturnal radiation loss, triggered OC.
Resumo A expansão da fronteira agropecuária impulsionada pelo desmatamento da Amazônia tem promovido uma crescente conversão de uso e cobertura da terra, onde florestas estão sendo convertidas em pastagens, e/ou, com o aumento populacional, dando espaço à criação de cidades. Essa transformação na cobertura vegetal é refletida nos elementos climáticos, e consequentemente, na resposta aos fenômenos atmosféricos de microescala. As pesquisas mais recentes realizadas com dados observacionais do Programa de Grande Escala da Biosfera-Atmosfera na Amazônia (LBA) indicam alterações nos ciclos biogeoquímicos da água como reduções de chuvas entre regiões florestadas e não florestadas. No entanto, poucos estudam avaliam a conversão de florestas em cidades. Diante disso, este estudo tem como objetivo avaliar como a modificação da cobertura do solo de floresta para cidade e de floresta para pastagem impacta as variáveis micrometeorológicas no ano de 2017, em Rondônia, sudoeste da Amazônia. Os resultados apontam que os elementos climáticos são modificados com a antropização, no entanto, entre os ambientes antropizados não foram evidenciadas diferenças significativas. A conversão de florestas aumentou a amplitude térmica em aproximadamente 2 °C.
Abstract In this study, the principles of complex climatology were applied to delineate topoclimates in the mountains of eastern Cuba. A regional numerical weather model, driven by reanalysis, was used to obtain temperature patterns at a resolution of 0.6 km. Unsupervised machine learning techniques were then utilized to identify weather types based on temperature during rainy and less rainy periods, as well as geographical location. This analysis was supplemented with historical precipitation and geographical data to identify 23 topoclimates in the study area. The methods allowed for the estimation of topoclimate identification errors and provided insights into the representativeness of surface meteorological stations for the study area. The results showed that the rainfall distribution of the identified topoclimates was consistent with that of local climate-forming factors and previous research. Furthermore, new insights into the climatological rainfall characteristics of the lower and middle heights of the mountains in eastern Cuba were identified.
Abstract A rapid verticalization to accommodate the citizens of the Metropolitan Region of São Paulo is altering the balance of radiation and atmospheric heat, highlighting the need to understand the impact that green and built infrastructure have on the canopy urban heat island phenomenon. This meteorological phenomenon occurs mainly due to the difference in landscape between urban and rural areas. Hypothetical scenarios with different green profiles were simulated using the WRF model coupled with SLUCM, and their results were compared to the current scenario using numerical data in order to observe the impact of green infrastructure. Comparison using output data showed that the total area of green infrastructure has great potential in reducing the intensity of the canopy urban heat island. The scenario with the largest total area and highest dispersion of green infrastructure recorded average urban temperatures 1.2 °C to 1.9 °C lower than the current scenario. Understanding the behavior of green infrastructure and its benefits is important for the development of municipal public policies that are in line with sustainable goals, and explicitly the relevance of urban parks and squares for local thermal regulation.
Abstract Downward longwave irradiance DLF is one of the main components of the surface radiation balance (SRB), but its direct measurement is currently limited. Clouds modulate its behavior, and clear-sky DLF0 is predominant in composition of the final DLF value. It is shown that in mid-latitude and tropical atmospheres, DLF0 can be represented as the sum of fluxes from three distinct spectral regions: R1 (λ < 7.5 µm), R2 (7.5 to 14 µm), and R3 (λ > 14 µm). R1 and R3 are closely described by blackbody radiation at screen temperature (Tscr), while R2 exhibits a mean emissivity that primarily depends on total precipitable water (w). It is presented a simple yet consistent physically-based model (hereafter denoted by OLD0), suitable for estimation of DLF0 at ground level. Validation of OLD0 with ground-based data of a worldwide set of 21 stations shows fair accuracy with bias MBE lower than 6 W.m-2 and spread (standard deviation STD) lower than 12 W.m-2 for typical values DLF0 ~ 300-400 W.m-2, compatible with surface pyrgeometer measures. The proposed algorithm outperforms existing methods, achieving a mean bias error (in module |MBE|) of approximately 2.8 W.m-2. In contrast, other widely used algorithms typically exhibit |MBEs| ranging from 8.1 to 15.9 W.m-2.
Resumo O presente trabalho objetivou obter um modelo para as séries temporais de vazão e precipitação, do grupo SARIMA, de dados observados em cinco estações pluviométricas e em uma estação fluviométrica, presentes na Bacia Hidrográfica do Rio Preto (BHRP), a fim de selecionar o melhor modelo que represente seu caráter preditivo. O modelo selecionado para os dados de precipitação total mensal (PTM), considerando os critérios de seleção de Akaike, Schwarz e Hannan e Quinn, foi o SARIMA (0,0,0)x(0,1,1), e o modelo selecionado para os dados de vazão mensal foi o SARIMA (1,0,0)x(0,1,1). Os modelos do tipo SARIMA demonstram boa capacidade em modelar e prever os dados de precipitação e vazão de uma bacia hidrográfica com relevo e sistemas atmosféricos heterogêneos. Esses modelos foram capazes de incorporar as características, como a sazonalidade e a correlação serial, e demonstraram bons ajustes aos dados, qualificando-os para a realização de previsões de longo termo, que foram feitas em um horizonte de 84 meses.
