Environmental data ===== Description ^^^^^^^^^^^ From a health point of view, meteorological factors influencing human health have been given greater prominence by the need to better understand the effects of urban environmental changes on population health and possible vulnerabilities to climate variability scenarios. Conducting research in urban biometeorology has been pointed out as an important area of study to understand the relationships between climate and health in cities, to understand both the processes that can trigger diseases and those that create healthy environments. Therefore, the study of the relations between climate and human health, especially from the perspective of global climate change, to predict its likely effects on the health of the population and vulnerabilities to these changes, becomes relevant. Biometeorological studies are not limited by physical or temporal scales, nor by the intensity of the atmospheric signal, which can vary from small local random variability to extreme events such as tropical cyclones, extreme ultraviolet radiation and thermal hazards. Clearly, within these broad limits, biometeorology lends itself naturally to interdisciplinary research and collaboration, involving areas such as medicine, biology, sociology, social policies, etc. Biometeorology is related to an infinite number of geographical scales. For example, a large-scale study would be the impact of climate change on human well-being. Or on a micro-scale: an investigation of the effects of temperature change on the spread of vectors of tropical diseases. All of these aspects are associated with the understanding of the spatial and temporal distribution of climatic, meteorological, environmental and geographical variables for the region of interest. Such information provides support for the understanding of potential relationships between the variables mentioned and the spatial and temporal evolution of epidemics. Such understanding will be determined from the acquisition of observational data and the execution of regional atmospheric computational models. The observational data to be used come from local sensors (eg, weather stations) and remote data from sensors fixed to satellites. It is important to emphasize the use of regional atmospheric modelling to cover potential blanks on the available climate database and the possibility of guaranteeing consistent space and time variables distribution for the study region. The meteorological and weather parameters could influence the infection rate and disease spread via various channels. For instance, the temperature level, pressure, winds, humidity, or air quality might increase or decrease the transmission rate, risk, and survival period of the virus in the air or on surfaces. Additionally, extreme weather or long-term meteorological events, such as heat waves or severe storms, that cause persistent conditions for several days, can increase the risk of respiratory disease transmission through various mechanisms, including displacement of populations, disruption of healthcare systems, and increased exposure to pollutants. Overall, while climate is not the sole determinant of respiratory disease transmission, it is certainly one of many factors that can play a role. The seasonal cycle of respiratory viral diseases has long been acknowledged, with annual outbreaks of the common cold and influenza occurring predictably during the winter season in temperate regions. This regular occurrence can be attributed to two primary factors: changes in environmental conditions and human behaviour. Extensive research has demonstrated the impact of temperature and humidity on respiratory viruses' stability and transmission rates. Numerous studies have also indicated potential seasonal determinants in the epidemics of respiratory viruses and the influence of these contributing factors on host characteristics. These determinants encompass seasonal fluctuations in temperature, absolute humidity (AH), sunlight exposure, vitamin levels, and host behaviour. These proposed factors can be categorised as alterations in environmental conditions, patterns of human behaviour, and viral attributes Data access information ^^^^^^^^^^^^^^^^^^^^^^^ The analysis of the information will be based on the creation of a data acquisition system integrated with a database of historical and spatial series. The implementation of the database will come from: - local weather stations owned by the proponent Institutions, as well as partnerships established with municipal, state and federal government agencies; - orbital/satellite sensors (optical and SAR): use of existing infrastructure to receive data in real-time covering the national territory and involving meteorological, oceanographic and ground cover information; - Weather Research Forecast Model: daily, 9km spatial resolution data. Data-specific information ^^^^^^^^^^^^^^^^^^^^^^^^^ The proposed methodology is based on climatological analysis of a 22-year data series (2000-2022) for daily weather conditions throughout the state of Rio de Janeiro. For the proposed study, initially, three main variables are being evaluated (temperature, relative humidity, and precipitation) and were composed of the integration of in-situ information collected by weather stations, satellite observation, as well as regional atmospheric models. As shown in Figure 1, the analysis begins with the integration of data with different characteristics to generate a continuous field of spatialized data for the area of interest with a 10km spatial resolution. With the creation of the database, the climatological analysis made it possible to evaluate the temporal and spatial anomalies for each of the variables to create a reference level for the study and thus make it possible to compare the time periods for each point. The next step involves climate regionalization based on clustering techniques. Regionalization allows for a spatial understanding of the distribution and classification of the area into regions with similar weather patterns associated with the primary variables (temperature, humidity, precipitation, and bioclimatic indices). Temporal analysis is then carried out based on wavelet techniques to assess seasonal and temporal patterns related to the different atmospheric phenomena that may influence or act as forcing factors in certain regions. With the spatial and temporal characterization computed, it was possible to obtain and establish combined and characteristic bioclimatic indices for each region/municipality. They represent annual trends (e.g., mean annual temperature, annual precipitation) seasonality (e.g., annual range in temperature and precipitation) and extreme or limiting environmental factors (e.g., temperature of the coldest and warmest month, and precipitation of the wet and dry quarters). The bioclimatic variables include isothermally (ratio of diurnal variation to annual variation in temperatures), temperature seasonality (standard deviation of the monthly mean temperatures), and precipitation seasonality (the Coefficient of Variation of the monthly precipitation estimates expressed as a percentage of the mean of those estimates). A complete list of the bioclimatic indices that served as the basis for the analysis for this study, as well as their description, can be seen in Table 1. Based on the spatial and temporal bioclimatic analysis and characterization, it is possible to assess and scale the persistence of a given atmospheric condition in a region. This is necessary due to the need for a certain atmospheric phenomenon to act in each area over a period of time (days/weeks) for there to be a consequent effect on the population and thus make it possible to correlate clinical data, such as the number of hospital admissions. As previously described, to compare and analyze seasonality, bioclimatic indices associated with the climatology of 22 years of data for Rio de Janeiro were calculated. Maps shown in the figures below represent two of the associated variables: precipitation seasonality and isothermally. These bioclimatic variables describe changes in temperature and rainfall variability, as well as potential changing interactions between the two. In addition, it was possible to use statistical time series decomposition into significant components. This is a statistical method used to deconstruct a time series into several components, each representing underlying patterns in the data: trend, seasonal, and residual. Trend (Step 1): trend is the long-term movement of the series. Typically, we use a smoother or moving average to calculate the trend. The key is that it removes the seasonal variation from the time series; Detrended Time Series (Step 2): we remove the trend component from the time series. This has the effect of making the time series "stationary". Stationary just means the detrended series no longer goes up or down but is centered; Seasonal (Step 3): the seasonal component captures regular patterns of variability within specific, fixed periods, such as daily, weekly, monthly, or quarterly fluctuations. The seasonal component is commonly calculated by using an average or median value at a seasonal frequency (e.g. daily, monthly, etc); Residuals (Remainder) (Step 4): the irregular component, also known as the residual or noise, represents the random variation in the data that cannot be attributed to the trend, seasonal, or cyclical components. These are unforeseen variations that do not follow a predictable pattern. Limitations of Environmental dataset ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ The above data are related to past analysis and real-time observation of the present moment and involve primary variables such as temperature, humidity, precipitation, wind, solar radiation, etc. Associated with this information will be integrated into the database information on short, medium and long term forecasts using computational modelling methodologies / environmental mathematics. The generation of physical fields from computational modelling will also allow reconstruction of historical series and absence of spatial and temporal data, reconstruction of local and specific physical data fields, and past scenarios in regions without available data. Due to the different sampling characteristics of each data source, the information to be acquired and integrated will undergo pre-processing to standardize the representativeness and distribution. The spatial and temporal sampling will be standardized with the objective of correlating and comparing with variables from other sources within the context of the project, such as those related to the clinic. .. rubric:: References - Technical presentations on the methodology by Fabio Hochleitner, Meteorologist, and member of the AESOP team: a. Monitoring Bioclimatological Parameters for Environmental Analysis Related to Respiratory Diseases: https://youtu.be/twGVPPHdcvQ b. Bioclimate data: https://youtu.be/TDNtPxcjnO4 c. Data Dictionary: https://fiocruzbr.sharepoint.com/:x:/r/sites/HPC-AESOP-FiocruzBA/Documentos%20Compartilhados/Governan%C3%A7a%20e%20curadoria%20de%20dados/Mapeamento%20dos%20dados%20-%20Documenta%C3%A7%C3%A3o/Metadata.xlsx?d=w39ad84c743fe4b3db22e3d267108155a&csf=1&web=1&e=XS7uHN - Slides, graphs and maps: a. Figure 1: https://drive.google.com/open?id=1vBN9EHgCXw0HcFQ7Bcy-IWprY7r2I90O&usp=drive_fs b. https://docs.google.com/presentation/d/19JXuJhDug7iit7qSERysBZHHddsUMW3g/edit?usp=sharing&ouid=102771901574537620671&rtpof=true&sd=true