ANALYSIS OF THE OCCURRENCE OF EXTREME EVENTS IN THE CITY OF RIO DE JANEIRO/

Objective: To investigate the trends of extreme rainfall and temperature indices in the city of Rio de Janeiro, as measures to assess climate change. Theoretical benchmark: Extreme climatic events have been observed more frequently than has become a concern all over the world, so the investigation of trends in extreme precipitation and temperature indices makes it necessary to enter how this dynamic in climatic conditions affects the city of Rio de Janeiro. Method: The methodology for assessing climate change will be used from the indicators defined by the Expert Team on Climate Change Detection, Monitoring and Indices (ETCCDMI) panel, but of the 27 indicators proposed by the panel, 12 indicators were used that were more adapted to the reality of the city of Rio de Janeiro, for the simulation of climate data, data from the stations of the National Institute of Meteorology (INMET) and for the simulation of the indicators in the RClimdex software. Daily data of 53 years of 19 (six) stations obtained from the historical series of the period between 1970 and 2023 were analyzed, for the precipitation indices: PRCPTOT, R10, R20, R95p and R99p, and for the temperature indices: TX90p, TN90p, TMAXmean, TMINmean and DTR. Results and conclusion: The results indicate a trend of decreasing rainfall of low rainfall and a significant increase of extreme rainfall, the data also show a decrease of thermal amplitude, with the increase of minimum temperatures and a moderate decrease of maximum temperatures. Implications of the research: The study in question presents internationally validated indicators that demonstrate how extreme events are intensifying in the city of Rio de Janeiro. Originality/value: The data demonstrated in this research are of extreme importance as a source of research for future works that address this theme, besides serving as a subsidy to decision makers to propose adaptive measures to the territory, in the face of the intensification of extreme events in the city of Rio de Janeiro.


INTRODUCTION
Extreme weather events have been observed around the world, in order to reach large populations.According to UN-Habtat in 2022, about 56% of the world's population lives in urban areas, and by 2050 this percentage can increase to 68%.These areas tend to exacerbate the effects of extreme climates due to soil sealing processes, irregular occupancy of slopes and areas susceptible to flooding, heavy presence of buildings with overload in urban infrastructure systems, thickening with heat islands, and concentration of socio-economic activities (JANKOVIĆ and HEBBERT, 2012;COELHO et al, 2022;ROCHA, 2021).
The report of the UN Intergovernmental Panel on Climate Change (IPCC), in the chapter on Managing the Risks of Extreme Events and Disasters to Advance Adaptation to Climate Change (UN, 2022) presents scientific monitoring indicators that demonstrate that extreme events are increasingly intense, in addition to having increased their frequency occurrence, so that natural disasters with loss of human life can intensify over the years (VISSIRINI et al, 2023).According to Vincent et al. (2005), the increase in economic losses and human lives caused by the occurrence of extreme events has stimulated the systematization of data, research and work in the scientific community in general, in order to understand these phenomena and to propose resilience measures to the populations living in urban spaces.
The intensification of the occurrence of events of a climatological and hydrological nature has presented a relationship with population growth and occupation in a disorderly manner in the big metropolises, above all by the lack of urban planning and public housing policies control in the advance of irregular occupations in areas of risk and environmental preservation (PEREIRA et al., 2020, COELHO et al., 2022;ROCHA, 2021).Tragedies related to environmental disasters such as floods and landslides have been increasingly observed in the state of Rio de Janeiro (SELUCHI et al., 2016).
The city of Rio de Janeiro, as the second largest metropolis in Brazil, with about 6,211,223 inhabitants, according to the IBGE's estimate for the year 2022, has great challenges related to the management of land use and the decrease of the vulnerability of the population to extreme climatic events, since because of the city's geomorphology, part of the population resides in areas susceptible to mass movements or subject to flooding.
In addition, the 2030 Agenda, with Sustainable Development Goal 13 (SDG 13), states that countries must 'take urgent action to combat climate change and its impacts', due to the consequences of climate change (REYMÃO et al., 2021;VISSIRINI et al., 2023).
In this context, this work aims to evaluate the trends of extreme precipitation and temperature event indices, using data from the National Meteorology Institute (INMET) stations and the RClimdex software.Thus, it is hoped to finally subsidize mechanisms and public policies that aim to minimize the vulnerability of the population, besides contributing to the elaboration of resilience measures for the city of Rio de Janeiro.

THEORETICAL FRAME
According to Guedes et al. (2012), when analyzing the history of humanity, extreme climatic events are identified from their beginnings, which directly influenced the migration of the population across the continents.Currently, due to population growth and increased urbanization, climate change has intensified extreme events with increasingly severe impacts on the population (GUEDES; GENOVEZ; VILARINO, 2012;DIAS, 2014).
In the city of Rio de Janeiro, the extreme events are related to some critical aspects, according to Egler and Gusmão (2014)These are: the rising sea levels, strong winds and heavy rains, a combination of these aspects and geomorphological characteristics and high population density, can generate devastating effects, in order to increase the degree of vulnerability to climate change in the city.
Noble and Marengo (2017) they studied the consequences of climate change in the metropolitan region of Rio de Janeiro, and demonstrated the increase in the intensity of the rains and outflows over the last fifty years in the region, which is also to be seen in the southeastern region of Brazil.
In order to identify climate change, 27 climate change indicators have been created by a team of experts in Climate Change Detection, Monitoring and Indices (ETCCDMI), and these indicators are composed of climate indicators of extremes of temperature and precipitation, which are adopted globally and allow a detailed analysis of climate change to be carried out.Of the 27 climate indices defined, some can be applied to medium latitudes, while others in the tropics and there are those that can be adopted for any locality (SANTOS, 2006).
According to Santos (2019), the climate extremes indices proposed by ETCCDMI collaborate with the identification of the dominant modes referring to climate changes, providing the information needed to determine the degree of relation of extreme events with atmospheric characteristics and facilitating the understanding of the interaction between the climatological aspects and the acting atmospheric systems.
Also according to Santos (2019), with the analysis of the indices of climate change, it is possible to understand the phenomena that occur on a global, regional and local scale, identifying climate change, allowing an estimate of frequency to be obtained and possible future natural disasters to be forecast, thus allowing a more positive decision making by public managers.
The calculations of the indices proposed by ETCCDMI were carried out in the software RClimdex which is a widely used program for the modeling of climate extremes indices, in order to monitor and detect changes in climate in a global, regional or local way (SOUZA & AZEVEDO, 2012).The integration model allows for independent application with statistical calculations and graphical representation of temperature and precipitation indicators in percentiles with non-homogeneous series (ZHANG E YANG, 2004).Accordingly, the software was used to determine the indices for detecting variabilities and climate changes observed in the city of Rio de Janeiro.

METHODOLOGY
Meteorological data used for this study were obtained from meteorological stations maintained by INMET, located in the city of Rio de Janeiro, with time series between 01/01/1970 and 31/12/2023.Stations were considered to be operating stations, and inactive stations, but with a series of data consisting of them.Figure 1 illustrates the geographical location of the stations used and Table 1 shows the time series and type of each station, whether conventional or automatic, and whether they are still in operation.
After the selection of the stations, the data was processed and consolidated in the R Climdex software, developed by Zhang and Yang ( 2004), as a tool for integrating the R programming language with Climdex AFL 1.0.The extreme precipitation and temperature indices analyzed in this study are presented in Table 2, with the indicator identification, definition and unit.Dereczynski et al. (2013) conducted research on the issue of climate change in the metropolitan region of Rio de Janeiro, and Silva & Dereczynski (2014) presented similar study of climate characterization for the state of Rio de Janeiro.In both studies, an intensification and increase of extreme events were observed in data from stations in the city and in the state of Rio de Janeiro.6 Precipitation is considered one of the most important meteorological variables related to extreme event studies and environmental assessments, so it was chosen to analyze a total of 7 (seven) indices related to this element, such as: PRCPTOT, R10, R20, R95p, R99p, CDD and CWD.In the analysis of the extreme temperature indices were considered: the mean of the maximum temperatures (TMAX mean), minimum temperatures (TMIN mean), hot days (TX90p), hot nights (TN90p) and thermal amplitude (DTR).
In the graphical representation available from the RClimDex, the axis of the abscissae represents the time series and in the axis of the ordered the values of the extreme indices analyzed, being a: • 'continuous line' as an indication of the actual trend of the historical series; • 'dashed line' means the trend estimated over time, which may also indicate a potential change in climate conditions, depending on the index analyzed at a given stage of the period considered, and • 'line with empty circles' as a dispersion of the index results analyzed each year over the historical series in a linear manner.
The statistical analysis obtained from RClimdex is based on the method of least squares, with the statistical parameters generated by the coefficients of determination (R2), as an adjustment of the regressions, with indication of the variation of the climate index as a function of time, that is, the higher the value of R2, the better the linear model is adjusted to the sample analyzed (SOUZA & AZEVEDO, 2012).
The slope of each line indicates the "uncertainty" in estimating a linear regression slope, and the "p-value" tests the null hypothesis that the coefficient is zero, i.e. no effect.A high pvalue (p > 0.05) indicates that the results were not statistically significant (MOREIRA; DA CUNHA; DA COSTA, 2021), however they are not negligible due to the seasonality of the indices analyzed, so as to consider them in the discussion of the results.
Therefore, it can be considered as an output result, RClimdex calculates, in addition to climate change indices, statistical data such as: linear trend calculated by the method of least squares; level of statistical significance of the trend (p value); coefficient of determination (r2) and standard error of estimation; as well as, the graphs of the annual series (SOUZA, 2012).

RESULTS AND DISCUSSIONS
In the graphs below extreme precipitation indices and extreme temperature indices can be observed.All graphs show the value of the squared error or R2 (in percent), the value of p (p-value), the slope estimate and the slope error.From the total annual precipitation of wet days (PRCPTOT), there is a linear real trend of the historical series with indication of stationarity during the period analyzed between 1970 and 2023.The number of days in the year with precipitation greater than or equal to 20 mm (R20) showed a result similar to the PRCPTOT index, with a linear trend without significant changes during the period.The number of days in the year with precipitation greater than or equal to 10 mm (R10) showed a drop from 47 days to 43 days, both for the actual trend of the series and for the estimated trend.In relation to the 95th percentile (R95p) and 99th percentile (R99p), there is an increase in the number of wet days and extremely humid days in the city of Rio de Janeiro.The total annual rainfall greater than the 95th percentile increased from 240 mm to about 400 mm, while for the 99th percentile, the increase was from 60 mm to 180 mm.As for the analysis of extreme temperature indices, there is a trend of falling on hot days with maximum temperature above 90 percentile (TX90p) from 12 days to 8 days, while the number of days in the year when the minimum temperature was higher than 90 percentile (TN90p) increased from 6.5 to 13 days.In addition, the temperature data show a trend of mean reduction of maximum temperatures (TMAXmean) from 29.2°C to 28.6°C.However, it is noted that the average minimum temperature (TMINmean) increased by 0.5°C from 19.4°C to 19.9°C (Graph 10).Another question posed is that the daily average of the daily temperature amplitude (DTR), which is the result of the average of the maximum temperature difference of the minimum temperature, shows a reduction from 9.7°C to 8.6°C.In the data obtained by De Souza (2022), we observed the same trend of reduction of the DTR index, being that the author analyzed two stations in the city of Rio de Janeiro, in different geomorphological domains, one located 300m above sea level, in the Alto da Boa Vista neighborhood and another at about 40m altitude, in the Realengo neighborhood.It should be noted that in most of the world there has also been a reduction in the DTR index (SILVA, 2012).13

FINAL CONSIDERATIONS
It can be concluded that in the analysis of the historical series of 53 years, between 1970 and 2023, of precipitation data and temperature of the 19 weather stations operated by INMET in this period, in the city of Rio de Janeiro, using the software RClimdex it was observed: • the reduction of days with rainfall less than or equal to 10 mm; • the stationality of days with rainfall less than or equal to 20mm; • the trend and significant increase in extreme precipitation from the 95th percentile (R95p) and the 99th percentile (R99p); • and the decrease in temperature amplitude, with the decrease in the mean of the days with maximum temperatures, and the increase in the days and the mean of the minimum temperatures.
It is recommended for future work the insertion of more meteorological stations and data in the modeling, in addition to the deepening of the analysis of other extreme indices of precipitation and temperature proposed by ETCCDMI, to buoy the dynamics and the change in the climate in the city of Rio de Janeiro.

Graph 1 -
Total annual rainfall on wet days (PRCPTOT) Source: Prepared by the author in the software Rclimdex (2024) Graph 2 -Day rainfall greater than 10mm (R10mm) Source: Prepared by the author in the software Rclimdex (2024) Graph 3 -Day rainfall greater than 20mm (R20mm) Source: Prepared by the author in the software Rclimdex (2024)

Graph 4 -
Very wet days -Annual rainfall >95p percentile Source: Prepared by the author in the software Rclimdex (2024) Chart 7 -Consecutive wet days with precipitation > 1mm (CWD) Source: Prepared by the author in the software Rclimdex (2024)

Graph 8 -
Hot Days with Temp Max>90th Percentile (TX90p) Source: Prepared by the author in the software Rclimdex (2024) Graph 9 -Hot Nights with Temp Max>90th Percentile (TN90p) Source: Prepared by the author in the software Rclimdex (2024)

Table 1 -
Data from INMET Stations in the city of Rio de Janeiro