1. Introduction
Disasters result from the combination of natural hazards and exposed population, comprising social, environmental, geological, physical, and economic vulnerability and insufficient response capacity (Freitas et al. 2012; Chmutina and von Meding 2019; United Nations Office for Disaster Risk Reduction 2020). These events engage both natural and social processes, impacting society according to the pattern of interaction between the natural event and the social organization (Mattedi and Butzke 2001). The impacts of disasters such as droughts and floods on the environment can compromise sanitation, water, soil, and food quality, as well as change disease vector cycles, leading to major economic losses (Freitas and Ximenes 2012; Patz et al. 2012; Freitas et al. 2020) and negatively influencing the provision of essential services.
Since the beginning of this century, extreme weather events have become more frequent in Brazil, with alternating periods of intense rainfall and water shortage (Marengo et al. 2009; Cavalcanti 2012). These extreme natural events potentialize the occurrence of disasters, causing impacts of different magnitudes on society, depending on their intensity, and the social and environmental conditions of exposed populations. There are several risks of disasters associated with climate extremes, such as flash floods, windstorms, hail, waterlogging, and lightning storms (Hartmann et al. 2013; Ning et al. 2016), in addition to droughts, which usually have lower immediate destructive power if compared to landslides and floods despite the larger number of affected populations.
Droughts are characterized by the persistence of periods with low pluviosity or its absence, when the decrease in soil moisture is higher than its recharge, which drastically reduces water storages and impacts the hydrological system (Coimbra de Castro 2003; Grigoletto et al. 2016). Floods occur due to the inundation of areas beyond the limits of a water body natural floodplain. The overflow is usually gradual, caused by the persistence of prolonged rainfall in the water basin, where discharge rates are exceeded by total system inputs. Flash floods, however, are characterized by a fast, highly energetic flow originating from intense rainfall events usually occurring over lands with a high natural slope. Waterlogging is characteristic of plain areas where flowing capacity is exceeded, thus generating an elevation of the water table [National Center for Risk and Disaster Management (Centro Nacional de Gerenciamento de Riscos e Desastres 2013)]. The latter having lowest record in the region under study and out of scope of our work.
According to the Center for Studies and Research in Engineering and Civil Defense [Centro Universitário de Estudos e Pesquisas sobre Desastres/Universidade Federal de Santa Catarina (CEPED/UFSC 2012)], the most recurrent environmental disasters in Brazil in the period from 1991 to 2011 originated from climatic events, with drought accounting for 54% of total registered episodes, followed by hydrological disasters (flash floods, floods, and waterlogging) corresponding to 33% of total events. Disasters of storm-related (cyclones, windstorms, storms) represent only 7% of total events while geological events, which comprise landslides and mass movements, correspond to 6%.
Northeast Brazil (NEB) is characterized by the occurrence of extreme drought events (Moura and Shukla 1981; Groisman et al. 2005; Marengo et al. 2016; de Medeiros et al. 2020). Annually, heavy rainfall events are also registered in the region (Calheiros et al. 2006; Oliveira et al. 2013b; Rodrigues et al. 2020b; Oliveira and Lima 2019), usually concentrated in short periods of time as a result of the simultaneous interaction between topography, geographical location, surface characteristics, and the acting of different meteorological systems (Kayano et al. 2009).
Climatic extremes events revealed that water supply systems and the rainfed agricultural sector in NEB are highly sensitive to drought conditions (Marengo et al. 2018; Pereira and Cuellar 2015) and to heavy rainfall, despite the lower frequency of occurrence of the latter (Liebmann et al. 2011; Kouadio et al. 2012). Consequently, rainfall measurement and monitoring play an important role in managing precipitation-related risks.
NEB is known to be a highly vulnerable area to climate factors, particularly its semiarid region (Silva et al. 2013), which is characterized by a warmer and drier climate, and the more frequent occurrence of drought-type disasters. This is one of the most vulnerable regions in the country (de Almeida et al. 2016, 2020) due to its specific characteristics of water availability, aridity, vegetation cover, land use, and social aspects. The main impacts caused by drought in the region are malnutrition, water supply deficiency, agricultural and livestock losses, human migration, forest fires, water quality degradation, and health and poverty problems (Stanke et al. 2013).
Intense rainfall events have sporadic periodicity and can cause hydrological and meteorological disasters with damages comparable to those generated by drought disasters (Camarinha and Debortoli 2015; Moura et al. 2016; Debortoli et al. 2017), especially when occurring in densely populated cities that usually present areas with inadequate housing conditions and higher social vulnerability. Hydrometeorological disasters triggered by heavy rainfall events (rainfall equal to or greater than 60.0 mm day−1) and extreme rainfall events (rainfall equal to or greater than 100.0 mm day−1) are registered more frequently in the coastal cities of the northern NEB region, such as Fortaleza (Zanella et al. 2009; Olímpio et al. 2013), as well as in coastal cities of the eastern sector of NEB, such as João Pessoa, Recife, Olinda, Maceió, and Natal (Souza et al. 2012; Nóbrega et al. 2014). In these cities, extreme rainfall events are known to have caused geological disasters such as landslides, associated with irregular urban occupation, mainly precarious housing located on slopes and hills.
The impacts of environmental disasters on populations show that communities are affected unequally and in different ways, directly and indirectly, with effects that vary from short to long term, depending on the characteristic of the event and the socioeconomic and environmental vulnerability of the territory (Alderman et al. 2012). Thus, preventive measures can reduce damage caused by these phenomena or at least minimize them, given that planning, prior studies, and risk assessment are carried out. In this case, prevention starts by understanding the factors and processes that trigger natural phenomena. Similarly, prevention can provide adequate conditions for the improvement of societies’ resilience with regard to such events (Marcelino 2008).
In this context, the objective of this study was to estimate the relationship between natural disasters and climate extreme events, sociosanitary and demographic conditions in the Northeast region of Brazil during the period from 1993 to 2013. To this end we considered only the most frequent events in NEB, which are droughts, flash floods, and floods.
2. Material and methods
a. Study area
NEB is located at the tropical zone between 1°N and 18°S and 34.5° and 48.5°W, in the extreme northeast of South America. It is contiguous in the north and east bordering the Atlantic Ocean. It has an area of approximately 1 558 196 km2, corresponding to 21.25% of the national territory, and a population of 53 081 950 inhabitants (Instituto Brasileiro de Geografia 2010). The region encompasses the states of Maranhão (MA), Piauí (PI), Ceará (CE), Rio Grande do Norte (RN), Paraíba (PB), Pernambuco (PE), Alagoas (AL), Sergipe (SE), and Bahia (BA), geographically divided into 42 mesoregions (Fig. 1) that are the observational units in this study.
Geographical division of Northeast Brazil into mesoregions.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0132.1
Mesoregions were grouped according to environmental disasters, climate extremes, sociosanitary and demographic characteristics, in order to highlight common features among them. Thus, three distinct profiles of vulnerability to disasters, as defined by the grade of membership (GoM) fuzzy technique. The classification of the mesoregions according to these profiles was carried out according to the degree of belonging, considering the following predominant characteristics in each of the clusters formed.
There are three predominant types of climate in the region: humid coastal climate (from the coast of Bahia to Rio Grande do Norte); tropical climate (in part of the states of Bahia, Ceará, Maranhão, and Piauí); and semiarid tropical climate (throughout the northeastern inlands). The onset and duration of the wet season differs according to the different subregions: in the eastern part of NEB, it takes place between May and August; in the southern part of the region maximum rainfall occurs in November–December; and in the northern semiarid part of NEB the wet season occurs between February and May (Uvo et al. 1998; Kucharski et al. 2008). The intertropical convergence zone (ITCZ) is the meteorological system that most influences the rainfall pattern in the region, producing rainfall in almost the entire NEB when it reaches its southernmost seasonal position (Oyama and Nobre 2004; Hastenrath 2006). Upper-tropospheric cyclonic vortices, easterly waves disturbances and squall lines may also cause heavy rainfall at any time of the year (Mishra et al. 2001; Gomes et al. 2015; Pereira de Oliveira and Oyama 2015). Furthermore, the southern state of Bahia is also affected by the propagation of frontal systems and the South Atlantic convergence zone (SACZ) (Tomaziello et al. 2015; Paredes-Trejo et al. 2017).
The region is characterized by a remarkable interannual and seasonal variability in rainfall (Zhou and Lau 2001; Kayano and Andreoli 2004; Barbosa et al. 2006). This characteristic favors the occurrence of prolonged droughts that cause economic impacts on the population (Coêlho et al. 2004; Blunden and Arndt 2016; Paredes-Trejo et al. 2015). The average annual rainfall from the coast of Bahia to Rio Grande do Norte is 2000 mm; in part of the states of Bahia, Ceará, Maranhão, and Piauí, it ranges between 1000 and 1200 mm; and in the entire semiarid region it is usually less than 500 mm on average (Da Silva 2004; Alvares et al. 2013; Oliveira et al. 2013a; Rao et al. 2016; Oliveira et al. 2017). Furthermore, the region is characterized by high reference evapotranspiration rates, which intensifies the occurrence of soil water deficits and, consequently, drought, as previously described by Cabral Júnior and Bezerra (2018).
b. Data
Data on disaster occurrences in the Northeast region of Brazil were obtained from CEPED of the Federal University of Santa Catarina, published in the Brazilian Atlas of Natural Disasters for each federation unit for the period from 1991 to 2012. Data for subsequent years were obtained from the Integrated Disaster Information System (S2ID), managed by the National Secretariat for Civil Protection and Defense (SEDEC). These datasets were obtained at the municipal level and then aggregated into mesoregions.
The meteorological data used in this study were retrieved from the joint project between the University of Texas (United States) and the Federal University of Espírito Santo (Brazil). The method used to produce the database was described by Xavier et al. (2016). This dataset is arranged in a regular grid of 0.25° × 0.25° and cover the entire Brazilian territory. Among the available variables, only precipitation (mm) data were used. We initially aimed to select one grid point to represent each mesoregion. However, given their dimensions and different meteorological conditions, one point was not sufficient to adequately describe their characteristics. Thus, in order to provide a more comprehensive characterization of each mesoregion, we selected one grid point in each municipality composing each analyzed unit. Then, a synthetic time series was created for each mesoregion based on the daily mean values between these selected grid points. These synthetic time series were then used in the calculation of the extreme indices for each mesoregion.
Rainfall extreme indices were obtained from the Climdex, developed by Zhang and Yang (2004). For this purpose, we used the methodology proposed by Zhang et al. (2005) and Haylock et al. (2006) to assist in the monitoring and detection of climate change in local studies, which was developed for the computational language of the R software and is freely available for download. Thus, we analyzed six rainfall extreme indices: maximum number of consecutive days with precipitation below 1 mm [consecutive dry days (CDD)], number of consecutive days with precipitation greater than or equal to 1 mm [consecutive wet days (CWD)], maximum amount of precipitation on 1 day (RX1day), maximum amount of precipitation on five consecutive days (RX5day), annual precipitation exceeding the 95th percentile (R95p), and total annual precipitation on wet days (PRECPTOT).
Data on sociosanitary and urban infrastructure conditions were obtained from the 2010 Demographic Survey, conducted by the Brazilian Institute of Geography and Statistics [Instituto Brasileiro de Geografia (IBGE)], and contained the percentage of households with inadequate water supply, sewage, and waste collection services.
Because of the limitations inherent to the different datasets that were used, we considered the period from 1993 to 2013, assuring the compatibility of meteorological and disaster datasets and the use of the proposed statistical methods.
c. Method
1) Principal components analysis
The principal components analysis (PCA) technique is a statistical tool usually applied to fit models in which explanatory variables tend to be highly correlated. In these cases, the methodological procedure consists of an orthogonal transformation (Johnson and Wichern 2002; Mingoti 2005), in which the possibly correlated variables are converted into a new set of linearly independent variables, called principal components (PC). PC are linear combinations of the original variables and can be interpreted as such. The number of retained components is less than or equal to the number of original variables, where the first PC explains most of total data variance. Thus, this technique was used to eliminate correlations between explanatory variables of the used dataset, avoiding data multicollinearity and improving the adjustment and interpretation of the proposed statistical models.
The principal component analysis was applied to the dataset by considering the triannual mean values of the meteorological variables and the sociosanitary and demographic variables provided by the Demographic Census of 2010.
To detect potential multicollinearity in the analyzed data, the correlation matrix was observed and the linear dependence between independent variables was verified. For this purpose, we calculated the condition number K as the ratio between the highest and lowest eigenvalues of the correlation matrix of the independent variables. Multicollinearity was assessed as follows: K < 100 (no multicollinearity), 100 < K < 1000 (moderate–strong multicollinearity), and K > 1000 (severe multicollinearity) (Montgomery et al. 2012).
2) Generalized additive models for location, scale, and shape
On the other hand, disaster counts for the events of drought, flash floods and floods imply different hierarchical levels (mesoregion, triennium, and profile), which are correlated with each other. In addition, since this study comprises longitudinal data, mixed-effect models emerge as an option that aims to describe the dependency structures of correlated data by including relevant random effects (Diggle et al. 2002). In this context, the correlation between repeated measures is the result of an unmeasurable heterogeneity but is described by the variability of random effects. The inclusion of random effects in count data models also accommodates data overdispersion.
To establish an initial notation, we consider each response variable as
3. Results and discussion
NEB accounted for a total of 14 762 drought, flash flood, and flood events from 1993 to 2013. Drought events corresponded to 81.1% of the occurrences, followed by flash floods (11.1%) and floods (7.8%). It is worth mentioning that events of other natures also occurred less frequently and were not considered in the study, such as waterlogging, mass movements, windstorms, forest fires, and erosion.
With regard to the distribution of the most frequent events in the nine NEB states (Fig. 2), Bahia comprised the mesoregions with the highest number of records, accounting for 3051 events (20.7%), followed by Paraíba with 2342 (15.9%) and Piauí with 2047 (13.9%), respectively. The fewest number of events were registered in the states of Maranhão 3.4% (505 events), Sergipe 4.2% (617 events), and Alagoas 5.3% (787 events). Studies developed by de Almeida et al. (2016, 2020) that aimed at the construction of indicators of risk to natural disasters at the national level, reported that NEB presents an expressive number of municipalities (52%) where there is a very high risk to disasters.
Register of occurrence of the most frequent environmental disasters in NEB in the period 1993–2013.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0132.1
Figures 3a–g shows the occurrence of disasters classified as drought in NEB in each triennium comprised in the period from 1993 to 2013. It can be seen that the mesoregions with the most frequent occurrences of drought are located in the center of the region, thus corresponding to the delimitations of the Brazilian semiarid region. This is the region that is most affected by events caused by lack of rainfall because of its climate and environmental characteristics (Marengo et al. 2016, 2018).
(a)–(g) Triennial and (h) annual occurrence of drought in the mesoregions of NEB in the period 1993–2013.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0132.1
The fewest drought registers are found in mesoregions comprised by the state of Maranhão and those located in the eastern NEB, which correspond to the most humid areas. In studies developed by Reboita et al. (2010), rainfall rates larger than 2000 mm yr−1 are observed in the northern part of NEB and rates of approximately 1500 mm yr−1 are observed in the east. As previously mentioned, the ITCZ is the main atmospheric system associated with precipitation in the region, while the occurrence of squall lines, mesoscale convective complexes, easterly waves disturbances, and upper-tropospheric cyclonic vortices should also be highlighted. It is worth noting that the Maranhão state is characterized by a warm, humid climate mainly due to its geographical location and the acting of atmospheric systems such as the ITCZ (Nascimento et al. 2017).
A low frequency of drought events was observed in the southern part of the state of Ceará during the analyzed trienniums, although it is located in the semiarid region and presents dry climate conditions. This different behavior is possibly associated with the number of reservoirs found in the region (Agência Nacional de Águas 2017), which provides greater resilience and capacity to cope with events caused by the irregularity of rainfall, thus leading to fewer drought disasters despite the semiarid conditions.
Figure 3h shows the annual frequency of drought events in NEB comprising the 21 years of analysis. A considerable number of drought events can be observed in NEB region. The years with the most registers of drought episodes were 1993, 1998, 2002, 2003, 2004, 2005, 2006, and 2007, which coincides with years when El Niño–Southern Oscillation acted. Grimm (2003), Rodrigues et al. (2011), and Cai et al. (2020) also observed that episodes of El Niño associated with sea surface temperature (SST) anomalies in the Pacific Ocean alter the ocean–atmosphere circulation patterns, inhibiting convective activity over NEB through the anomalous circulation pattern of the descending branch of the Walker cell.
El Niño is not the only oceanic phenomenon affecting drought conditions in NEB. Differences in SST anomalies in the North and South Atlantic Oceans are also worth mentioning, which is known as the Atlantic dipole (Moura and Shukla 1981; Amorim et al. 2014). The positive phase of the Atlantic dipole influences the northward displacement of the ITCZ, which directly interferes with NEB rainfall regime through changes in circulation patterns (Kayano et al. 2013).
The peak of the drought occurrence time series is observed in 2012, with an extreme situation that is known to have lasted until 2016 and was associated with different oceanic conditions that changed the patterns of atmospheric circulation, thus causing strong anomalies in rainfall rates (Marengo et al. 2017; de Medeiros et al. 2020). The extreme conditions significantly impacted economic activities and the living conditions of the population (Brito et al. 2018). The event was considered the most severe drought to have occurred in the region in recent years (Alvalá et al. 2019; Cunha et al. 2018).
Flash floods are more occasional events, triggered by torrential and/or concentrated rain events, thus occurring at shorter time scales but with a highly destructive power when interacting with the population of locations with precarious infrastructure. This fact aggravates the vulnerability of the population exposed to this type of hazard.
Figures 4a–g shows the occurrence of flash flood events in NEB, considering each triennium comprised in the period from 1993 to 2013. The period from 2002 to 2004 presented the most significant registers (Fig. 3d), with a high frequency of events in 2004 (more than 400), followed by the period from 2008 to 2010 (Fig. 3f), with 2009 standing out. Note that a considerable number of mesoregions did not register flash flood events in the trienniums of 1999–2001 and 2005–07, which were the years with the lowest number of cases in the analyzed period.
As in Fig. 3, but for flash floods.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0132.1
The mesoregions located at the states of Bahia, Pernambuco, and Paraíba stood out with higher occurrences of flash flood events, which further increased in the anomalous trienniums in the time series. de Abreu et al. (2020) reported that these regions have higher altitude and altimetry, which favors the occurrence of flash floods. Souza and Azevedo (2012) and Silva (2014) observed higher registers of intense rainfall (rainfall equal to or greater than 60 mm day−1) and extreme rainfall (rainfall equal to or greater than 100 mm day−1) events in these regions, which potentialize the occurrence of flash floods. Regions to the south of the state of Bahia, where flash floods occur in practically all the analyzed trienniums, are influenced by the climate conditions of the southeast region of the country, in addition to the direct influence of atmospheric systems such as frontal systems and the SACZ (Grimm 2011; de Carvalho and Cavalcanti 2016).
In Fig. 4h one can observe the annual frequency of flash flood events in NEB. It can be seen that these events occur throughout all the data series, with the exception of 1993, which was influenced by the weather conditions associated with the extreme drought established since 1991 due to a strong El Niño event (Rao et al. 1995; Oliveira et al. 2010). When compared to the occurrence of drought events, flash flood occurrence is much less frequent in absolute terms since the overall climate conditions in NEB favor rainfall deficit (Cabral Júnior and Bezerra 2018). The years considered atypical throughout the series are 2004, 2009, and 2010, which are associated with macro-, meso-, and microscale atmospheric systems (Orlanski 1975) that favor the occurrence of rainfall in NEB region at different time scales.
In Figs. 5a–g, the spatial–temporal distribution of flood events in the Northeast region from 1993 to 2013 can be observed. It is worth mentioning that floods differ from flash floods as they are influenced by the gradual persistence of above-average rainfall. The first three analyzed trienniums (Figs. 4a–c) exhibited the lowest overall frequencies of these events, with isolated events occurring in certain mesoregions.
As in Fig. 3, but for floods.
Citation: Weather, Climate, and Society 13, 3; 10.1175/WCAS-D-20-0132.1
The first triennium with high records of flood events was 2002–04 (Fig. 4d), with 2004 accounting for the largest number of cases throughout the entire extension of NEB, with particularly high frequencies in the mesoregions of the Rio Grande do Norte state. The triennium 2008–09 also stood out (Fig. 4f) with many flood registers throughout all states. However, the highest concentration of events in this triennium was observed in mesoregions located in the northern part of NEB region, particularly in the states of Maranhão, Piauí, Ceará, Rio Grande do Norte, and Paraíba. In a study developed by Rodrigues et al. (2020a), which assessed daily rainfall in NEB, the northern portion of the region presented the highest monthly rainfall averages between January and May, with a peak in March of approximately 200 mm, in addition to larger annual accumulated rainfall, which was influenced by the positioning of the ITCZ and mesoscale systems (Palharini and Vila 2017).
Figure 5h shows the annual distribution of flood events in NEB. The anomalous years were the same as in the flash floods time series, with 2004, 2008, and 2009 standing out with the highest records. As previously discussed, the excessive rainfall that potentialized the occurrence of these types of events in the mentioned years originated from specific atmospheric systems that acted over NEB and caused high amounts of rainfall.
To estimate the risk of occurrence of events associated with climate and sociosanitary variables, the generalized additive model for location, scale, and shape was used. We verified that the explanatory variables of this study showed signs of moderate to strong multicollinearity (K = 152.66), which would compromise the estimation of the model parameters. Thus, PCA was performed and retrieved five new variables (PC) that are not correlated with each other (Table 1). These components explain 94% of the accumulated variance of the 10 climate, sociosanitary, and demographic explanatory variables.
Principal components analysis of the climate, sociosanitary, and demographic variables according to the mesoregions of NEB. Values in boldface type are the highest percentages of factor loading contributions to the respective variables in the construction of the principal components.
The five components have the following characteristics: PC1 is characterized by excessive rainfall, since it comprises R95p, RX1day, and RX5day; PC2 is characterized by highly urbanized regions with inadequate sewage and waste collection; PC3 is characterized by wet conditions, since it comprises CWD and PRECPTOT; PC4 corresponds to inadequacy in water supply; PC5 represents the maximum number of CDD.
Subsequently, generalized mixed linear models were adjusted in order to identify the relationship between the principal components and the number of environmental disaster events in NEB. Thus, four distinct distributions were tested for the response variables: Poisson, double Poisson, negative binominal (type I), and negative binomial (type II). The best distribution for the response variables was selected based on the residual analysis criteria for each of the proposed models. The models adjusted with the NB type-II distribution were the ones that best fitted the data. Therefore, Table 2 shows the parameters estimated with the generalized mixed linear model using the type-II NB distribution for the different analyzed disaster types.
Parameters estimation for the generalized additive model for location, scale, and shape of drought, flash flood, and flood events in NEB. Asterisks indicate the probability of the distribution in evidencing the association of natural disasters with the predictor variables, that is, the principal components analyzed.
With regard to drought events, the estimated model detected statistically significant associations with the components PC1, PC2, PC3, and PC4. PC1 and PC3 were negatively associated with drought, that is, the higher the rainfall extremes, the more consecutive wet days, and the higher annual rainfall for the region, the lowest the number of drought events. Each increase in PC1 is associated with s a reduction in drought events of approximately 28.5%, while the increase in PC3, which expresses the volume of accumulated rainfall in wet days, leads to a decrease of 20.4% in drought occurrence.
PC2 and PC4 positively contribute to the increase in the number of drought registers, that is, the greater the number of housings with inadequate sewage, waste collection and water supply, and the greater the percentage of urban population, the more drought disasters are notified. Each unit added in the inadequacies of the sanitary conditions and urban population of NEB corresponds to an average increase of 18.5% and 14.9% in the notification of drought-related disasters, respectively. In general, the simultaneous unit increase in all covariables results in a 19% decrease in drought events in the region.
With the exception of PC5, all other components positively contribute to the number of flash flood events. That is, the greater the rainfall extremes, number of domiciles with inadequate sewage and waste collection, percentage of urban population, consecutive wet days, and annual rainfall, the higher the number of flash flood cases in NEB. The contribution of PC1 to the increase in flash flood occurrence is 29.2%; for PC2 and PC3 it is 23.2% and 42.8%, respectively. The number of consecutive dry days (CP5) negatively contributes to the occurrence of flash floods, which is expected since a decrease of 73.3% in the occurrence of flash floods was observed with a larger number of consecutive dry days. The combined increase in the analyzed explanatory variables lead to an increase of 40% in the occurrence of flash floods in NEB.
For disasters classified as flood, only the components representing climate variables were significant for the model. CP1 and CP3 positively contribute to the number of flood events, that is, wet consecutive days and annual accumulated rainfall tend to increase the frequency of flood disasters, contributing individually with an increase of 69.7% and 37.0%, respectively. In contrast, PC5 contributes negatively by reducing in 0.3% the occurrence of floods. The combined increase in covariables results in a 57% increase in the number of flood events in NEB region. Last, note that the results described in the previous paragraphs represent an average assessment of the model’s behavior. More specific results for each mesoregion can be interpreted by analyzing the associated random effect of each model. In all adjusted mixed models, the random effect variance is not negligible. The estimation of the variance of the random effects represents the differences between NEB mesoregions.
Excessive rainfall in NEB is extremely important for the region’s development, since it suffers from critical rainfall deficits that significantly affect the population that heavily depends on subsistence agriculture. However, the social and environmental conditions of the cities can also potentialize the occurrence of hydrometeorological disasters associated with excessive rainfall. In a study developed by Oliveira and Lima (2019) that analyzed the period of return of extreme events in the capital cities of NEB, the authors observed the occurrence of more frequent and more intense events in the wet season. However, these locations present a different behavior whether they are situated in the northern (São Luiz, Teresina, and Fortaleza) or coastal (Natal, João Pessoa, Recife, Maceió, Aracaju, and Salvador) portions of NEB.
The inadequate sewage, waste collection and water supply conditions of households significantly contribute to enhance the exposure to the occurrence of droughts and flash floods. The precarious sanitary conditions of the population contribute to increase the risks associated with the episodes of disasters in NEB, which is further influenced by high percentages of urban population. Silva (2009) reported that the urbanization process is one of the main aspects that influence disasters resulting from excessive rainfall in urban areas, often caused by the occupation of risk areas by the population, which is associated with increased surface runoff, floodplain occupation and social vulnerability. In general, these areas are not covered by impact prevention or mitigation plans and therefore are severely exposed to these risks. Xavier et al. (2014) emphasized the importance of the combination of social and economic processes to the occurrence of disasters.
Rodrigues and Gouveia (2013) reported that mass movements, flash floods, water resources pollution and soil contamination are the main environmental impacts resulting from the urbanization process. In addition, flash flood events pose a series of threats such as the contamination of water sources and decreased agricultural productivity, especially when referring to vulnerable regions such as NEB, which usually struggle to face and recover from the adverse effects of these events (Silva et al. 2013).
By analyzing the mean estimates of the proposed models for each of the analyzed disaster classification (Table 3), one can observe a certain variability between the profiles considered in the study. The mesoregions classified as having an amorphous profile stood out with a very specific characteristic for the three types of studied disasters. For the other analyzed profiles, one can identify that drought events in profile 1 mesoregions presented an average count of 60 events (59.96). Projections developed by Spinoni et al. (2014) and the analysis of extreme events carried out by Da Silva et al. (2019a) indicated an increase in temperature and reduction in precipitation with an increasing trend for the number of consecutive dry days in the region.
Mean estimates of the profiles of vulnerability to environmental disasters in NEB, retrieved by the generalized additive model for location, scale, and shape of drought, flash flood, and flood events. Here, LT and UT respectively indicate the lower and upper limits of the 95% confidence interval.
With regard to flash floods and floods, profile 2 showed higher averages with approximately nine and five cases (9.13 and 4.62), respectively. Studies developed by Da Silva et al. (2019b) and Rodrigues et al. (2020b) indicated a growing trend in extreme rainfall events in the coastal portion of NEB. These authors reported that these types of events are becoming more frequent and more intense, with 178 mm day−1 mean accumulated values in a 2-yr return period. Note that all estimates from the model are within the 95% confidence interval (CI95%).
The mesoregions classified as profile 2 and profile 3 presented greater risk associated with the events triggered by excessive rainfall, which can be observed by the estimations retrieved with the models for the flash floods and floods events. The eastern coast and northern portion of NEB, covering the state of Maranhão, are areas that present higher rainfall amounts and, consequently, more frequent episodes of rainfall extremes (Oliveira et al. 2013b, 2017; Rodrigues et al. 2020b). Profile 2 and profile 3 differ with regard to sociosanitary aspects, since profile 2 has the best sanitary conditions, which leads to better capacity to cope with the events. Furthermore, profile 2 mesoregions encompass the largest urban agglomerations in NEB, including eight capital cities.
4. Conclusions
NEB is predominantly characterized by the frequent occurrence of droughts, which affect most part of the population although it is not an environmental disaster with highly destructive power. Events caused by excessive rainfall such as flash floods and floods are also recurrent, although much less common than droughts. The remarkable climate variability in NEB potentialize the impacts of these events due to extreme climate and weather conditions. However, climate extreme events in their own are not sufficient to characterize a disaster. Thus, the social and environmental conditions in which the populations are inserted also influence the occurrence of disasters.
Generalized additive models of location, scale, and shape proved to be effective in capturing the relationship between the occurrence of disasters and the climate, sociosanitary, and demographic conditions of NEB. Its applicability seems advantageous for taking into consideration the overdispersion and the nature of the probability distribution of the count of drought, flash flood, and flood events (dependent variables). This type of modeling allows the inclusion of random effects, which take into account the specific variability of each of the profiles formed on the basis of the disasters, sociosanitary, and climate characteristics of the mesoregions. Thus, these types of model emerge as an innovative method to estimate the risk of occurrence of events, considering the associated factors.
Through the proposed models, statistically significant associations were found between disasters registers and climate, sociosanitary, and demographic conditions. Conditions that favor the occurrence of excessive rainfall were significantly associated with all analyzed disaster types. They negatively contributed to the occurrence of water shortage events (droughts) and positively contributed to the occurrence of flash floods and floods, which are triggered by excessive rainfall. Sociosanitary conditions were also important parameters in the analyzed models, with nonsignificant results only for the flood disaster model.
Therefore, we expect that the results of this research can elucidate issues that will help in the development of public policies with regard to the prevention, mitigation, preparedness, response, and recovery from impacts of environmental disasters on the population of NEB.
Data availability statement
All data on the occurrences of natural disasters used during this study are openly available on the Integrated Disaster Information System (https://s2id.mi.gov.br/). Precipitation information comes from Xavier et al. (2016), and the sociohealth conditions microdata are from the Brazilian Institute of Geography and Statistics (IBGE).
REFERENCES
Agência Nacional de Águas, 2017: Reservatórios do semiárido Brasileiro: Hidrologia, balanço hídrico e operação (Reservoirs in the Brazilian semiarid region: Hydrology, water balance and operation). ANA Brasília–DF Annex B Rep., 88 pp., https://www.ana.gov.br/noticias/estudo-reservatorios/reservatorios-do-semiarido-brasileiro_hidrologia-balanco-hidrico-e-operacao-1.pdf.
Alderman, K., L. R. Turner, and S. Tong, 2012: Floods and human health: A systematic review. Environ. Int., 47, 37–47, https://doi.org/10.1016/j.envint.2012.06.003.
Alvalá, R. C., A. P. Cunha, S. S. Brito, M. E. Seluchi, J. Á. Marengo, O. L. Moraes, and M. A. Carvalho, 2019: Drought monitoring in the Brazilian semiarid region. An. Acad. Bras. Ciênc., 91, e20170209, https://doi.org/10.1590/0001-3765201720170209.
Alvares, C. A., J. L. Stape, P. C. Sentelhas, J. L. de Moraes Gonçalves, and G. Sparovek, 2013: Köppen’s climate classification map for Brazil. Meteor. Z., 22, 711–728, https://doi.org/10.1127/0941-2948/2013/0507.
Amorim, A. C. B., R. R. Chaves, and C. M. Santos e Silva, 2014: Influence of the tropical Atlantic Ocean’s sea surface temperature in the eastern Northeast Brazil precipitation. Atmos. Climate Sci., 4, 874–883, https://doi.org/10.4236/acs.2014.45077.
Barbosa, H. A., A. R. Huete, and W. E. Baethgen, 2006: A 20-year study of NDVI variability over the northeast region of Brazil. J. Arid Environ., 67, 288–307, https://doi.org/10.1016/j.jaridenv.2006.02.022.
Blunden, J., and D. S. Arndt, 2016: State of the climate in 2015. Bull. Amer. Meteor. Soc., 97 (8), Si–S275, https://doi.org/10.1175/2016BAMSStateoftheClimate.1.
Brito, S. S. B., A. P. M. Cunha, C. C. Cunningham, R. C. Alvalá, J. A. Marengo, and M. A. Carvalho, 2018: Frequency, duration and severity of drought in the semiarid Northeast Brazil region. Int. J. Climatol., 38, 517–529, https://doi.org/10.1002/joc.5225.
Cabral Júnior, J. B., and B. G. Bezerra, 2018: Análises da evapotranspiração de referência e do índice de aridez para o Nordeste do Brasil (Analyzes of reference evapotranspiration and aridity index for Northeastern Brazil). Rev. Geoci. Nordeste, 4, 71–89.
Cai, W., and Coauthors, 2020: Climate impacts of the El Niño–Southern Oscillation on South America. Nat. Rev. Earth Environ., 1, 215–231, https://doi.org/10.1038/s43017-020-0040-3.
Calheiros, A. J., L. C. B. Molion, J. C. M. Vaz, and R. S. Tenório, 2006: Um evento de precipitação extrema sobre a costa leste do Nordeste do Brasil (An extreme rainfall event on the east coast of Northeast Brazil). Congresso Brasileiro de Meteorologia XIV, Florianópolis, Brazil, Brazilian Meteorological Society, http://mtc-m16b.sid.inpe.br/col/sid.inpe.br/mtc-m17@80/2006/12.26.11.56/doc/Calheiros.Evento.pdf.
Camarinha, P. I. M., and N. S. A. Debortoli, 2015: Brazilian vulnerability index to natural disasters of drought in the context of climate change. 2010 Fall Meeting, San Francisco, CA, Amer. Geophys. Union, Abstract NH33C-1932.
Cavalcanti, I. F. A., 2012: Large scale and synoptic features associated with extreme precipitation over South America: A review and case studies for the first decade of the 21st century. Atmos. Res., 118, 27–40, https://doi.org/10.1016/j.atmosres.2012.06.012.
Centro Nacional de Gerenciamento de Riscos e Desastres, 2013: Anuário brasileiro de desastres naturais (Brazilian yearbook of natural disasters). Ministério da Integração Nacional, Secretaria Nacional de Proteção e Defesa Civil Rep., 75 pp., https://antigo.mdr.gov.br/images/stories/ArquivosDefesaCivil/ArquivosPDF/publicacoes/Anurio-Brasileiro-de-Desastres-Naturais-2013.pdf.
CEPED/UFSC, 2012: Atlas Brasileiro de Desastres Naturais 1991 a 2010 (Brazilian Atlas of Natural Disasters 1991 to 2010). Volume Brasil, 94 pp.
Chmutina, K., and J. A. von Meding, 2019: Dilemma of language: “Natural disasters” in academic literature. Int. J. Disaster Risk Sci., 10, 283–292, https://doi.org/10.1007/s13753-019-00232-2.
Coimbra de Castro, A. L., J. G. Adair, and J. S. Mocellin, 2004: Psychological responses to drought in northeastern Brazil. Rev. Interam. Psychol., 38, 95–103.
Castro, A. L. C, 2003: Manual de desastres: Desastres naturais (Disaster handbook: Natural disasters). Ministério da Integração Nacional Rep., 182 pp., https://www.livrosgratis.com.br/ler-livro-online-41339/manual-de-desastres--desastres-humanos--de-natureza-tecnologica.
Cunha, A. P. M. A., J. Tomasella, G. G. Ribeiro-Neto, M. Brown, S. R. Garcia, S. B. Brito, and M. A. Carvalho, 2018: Changes in the spatial–temporal patterns of droughts in the Brazilian Northeast. Atmos. Sci. Lett., 19, e855, https://doi.org/10.1002/asl.855.
Da Silva, P. E., C. M. Santos e Silva, M. H. C. Spyrides, and L. D. M. B. Andrade, 2019a: Análise de índices de extremos climáticos no Nordeste e Amazônia Brasileira para o período entre 1980 a 2013 (Analysis of extreme climate indices in the Northeast and the Brazilian Amazon for the period between 1980 and 2013). Anu. Inst. Geoci., 42, 137–148, https://doi.org/10.11137/2019_2_137_148.
Da Silva, P. E., C. M. Santos e Silva, M. H. C. Spyrides, and L. D. M. B. Andrade, 2019b: Precipitation and air temperature extremes in the Amazon and Northeast Brazil. Int. J. Climatol., 39, 579–595, https://doi.org/10.1002/joc.5829.
Da Silva, V. P. R., 2004: On climate variability in Northeast of Brazil. J. Arid Environ., 58, 575–596, https://doi.org/10.1016/j.jaridenv.2003.12.002.
de Abreu, L. P., W. A. Gonçalves, E. V. Mattos, and R. I. Albrecht, 2020: Assessment of the total lightning flash rate density (FRD) in Northeast Brazil (NEB) based on TRMM orbital data from 1998 to 2013. Int. J. Appl. Earth Obs. Geoinf., 93, 102195, https://doi.org/10.1016/j.jag.2020.102195.
de Almeida, L. Q., T. Welle, and J. Birkmann, 2016: Disaster risk indicators in Brazil: A proposal based on the world risk index. Int. J. Disaster Risk Reduct., 17, 251–272, https://doi.org/10.1016/j.ijdrr.2016.04.007.
de Almeida, L. Q., A. M. S. de Araujo, T. Welle, and J. Birkmann, 2020: DRIB Index 2020: Validating and enhancing disaster risk indicators in Brazil. Int. J. Disaster Risk Reduct., 42, 101346, https://doi.org/10.1016/j.ijdrr.2019.101346.
Debortoli, N. S., P. I. M. Camarinha, J. A. Marengo, and R. R. Rodrigues, 2017: An index of Brazil’s vulnerability to expected increases in natural flash flooding and landslide disasters in the context of climate change. Nat. Hazards, 86, 557–582, https://doi.org/10.1007/s11069-016-2705-2.
de Carvalho, L. M. V., and I. F. A. Cavalcanti, 2016: The South American monsoon system (SAMS). The Monsoons and Climate Change, L. de Carvalho and C. Jones, Eds., Springer, 121–148.
de Medeiros, F. J., C. P. de Oliveira, and R. R. Torres, 2020: Climatic aspects and vertical structure circulation associated with the severe drought in Northeast Brazil (2012–2016). Climate Dyn., 55, 2327–2341, https://doi.org/10.1007/s00382-020-05385-1.
Diggle, P., P. J. Diggle, P. Heagerty, K. Y. Liang, P. J. Heagerty, and S. Zeger, 2002: Analysis of Longitudinal Data. Oxford University Press, 633 pp.
Freitas, C. M. D., and E. F. Ximenes, 2012: Enchentes e saúde pública: Uma questão na literatura científica recente das causas, consequências e respostas para prevenção e mitigação (Floods and public health: A review of the recent scientific literature on the causes, consequences and responses to prevention and mitigation). Cienc. Saude Colet., 17, 1601–1616, https://doi.org/10.1590/S1413-81232012000600023.
Freitas, C. M. D., M. L. D. Carvalho, E. F. Ximenes, E. F. Arraes, and J. O. Gomes, 2012: Vulnerabilidade socioambiental, redução de riscos de desastres e construção da resiliência: Lições do terremoto no Haiti e das chuvas fortes na Região Serrana, Brasil (Socio-environmental vulnerability, disaster risk reduction and resilience building: Lessons from the earthquake in Haiti and heavy rains in the highlands, Brazil). Cienc. Saude Colet., 17, 1577–1586, https://doi.org/10.1590/S1413-81232012000600021.
Freitas, C. M. D., I. V. D. M. Silva, D. R. Xavier, E. L. Silva, and C. Barcellos, 2020: Desastres naturais e seus custos nos estabelecimentos de saúde no Brasil no período de 2000 a 2015 (Natural disasters and their costs in health facilities in Brazil from 2000 to 2015). Cad. Saude Publica, 36, e00133419, https://doi.org/10.1590/0102-311x00133419.
Gomes, H. B., T. Ambrizzi, D. L. Herdies, K. Hodges, and P. F. Pontes da Silva, 2015: Easterly wave disturbances over Northeast Brazil: An observational analysis. Adv. Meteor., 2015, 176238, https://doi.org/10.1155/2015/176238.
Grigoletto, J. C., and Coauthors, 2016: Gestão das ações do setor saúde em situações de seca e estiagem (Management of health sector actions in drought and drought situations). Ciênc. Saúde Colet., 21, 709–718, https://doi.org/10.1590/1413-81232015213.26212015.
Grimm, A. M., 2003: The El Niño impact on the summer monsoon in Brazil: Regional processes versus remote influences. J. Climate, 16, 263–280, https://doi.org/10.1175/1520-0442(2003)016<0263:TENIOT>2.0.CO;2.
Grimm, A. M., 2011: Interannual climate variability in South America: Impacts on seasonal precipitation, extreme events, and possible effects of climate change. Stochastic Environ. Res. Risk Assess., 25, 537–554, https://doi.org/10.1007/s00477-010-0420-1.
Groisman, P. Ya., R. W. Knight, D. R. Easterling, T. R. Karl, G. C. Hegerl, and V. N. Razuvaev, 2005: Trends in intense precipitation in the climate record. J. Climate, 18, 1326–1350, https://doi.org/10.1175/JCLI3339.1.
Hartmann, D. L., and Coauthors, 2013: Observations: Atmosphere and surface. Climate Change 2013: The Physical Science Basis, T. F. Stocker et al., Eds., Cambridge University Press, 159–254.
Hastenrath, S., 2006: Circulation and teleconnection mechanisms of Northeast Brazil droughts. Prog. Oceanogr., 70, 407–415, https://doi.org/10.1016/j.pocean.2005.07.004.
Haylock, M. R., and Coauthors, 2006: Trends in total and extreme South American rainfall in 1960–2000 and links with sea surface temperature. J. Climate, 19, 1490–1512, https://doi.org/10.1175/JCLI3695.1.
Instituto Brasileiro de Geografia, 2010: Estados: População (States: Population). IBGE, http://censo2010.ibge.gov.br/apps/atlas/.
Johnson, R. A., and D. W. Wichern, 2002: Applied Multivariate Statistical Analysis. Prentice Hall, 794 pp.
Kayano, M. T., and R. V. Andreoli, 2004: Decadal variability of northern Northeast Brazil rainfall and its relation to tropical sea surface temperature and global sea level pressure anomalies. J. Geophys. Res., 109, C11011, https://doi.org/10.1029/2004JC002429.
Kayano, M. T., C. P. de Oliveira, and R. V. Andreoli, 2009: Interannual relations between South American rainfall and tropical sea surface temperature anomalies before and after 1976. Int. J. Climatol., 29, 1439–1448, https://doi.org/10.1002/joc.1824.
Kayano, M. T., R. V. Andreoli, and R. A. F. Souza, 2013: Relations between ENSO and the South Atlantic SST modes and their effects on the South American rainfall. Int. J. Climatol., 33, 2008–2023, https://doi.org/10.1002/joc.3569.
Kouadio, Y. K., J. Servain, L. A. Machado, and C. A. Lentini, 2012: Heavy rainfall episodes in the eastern Northeast Brazil linked to large-scale ocean–atmosphere conditions in the tropical Atlantic. Adv. Meteor., 2012, 369567, https://doi.org/10.1155/2012/369567.
Kucharski, F., D. Polzin, and S. Hastenrath, 2008: Teleconnection mechanisms of Northeast Brazil droughts: Modeling and empirical evidence. Rev. Bras. Meteor., 23, 115–125, https://doi.org/10.1590/S0102-77862008000200001.
Liebmann, B., G. N. Kiladis, D. Allured, C. S. Vera, C. Jones, L. M. V. Carvalho, I. Bladé, and P. L. M. Gonzáles, 2011: Mechanisms associated with large daily rainfall events in Northeast Brazil. J. Climate, 24, 376–396, https://doi.org/10.1175/2010JCLI3457.1.
Marcelino, E. V., 2008: Desastres naturais e geotecnologias: Conceitos básicos (Natural disasters and geotechnologies: Basic concepts). INPE Rep., 40 pp., http://mtc-m16c.sid.inpe.br/col/sid.inpe.br/mtc-m18@80/2008/07.02.16.22/doc/publicacao.pdf.
Marengo, J. A., R. Jones, L. M. Alves, and M. C. Valverde, 2009: Future change of temperature and precipitation extremes in South America as derived from the PRECIS regional climate modeling system. Int. J. Climatol., 29, 2241–2255, https://doi.org/10.1002/joc.1863.
Marengo, J. A., A. P. Cunha, and L. M. Alves, 2016: A seca de 2012-15 no semiárido do Nordeste do Brasil no contexto histórico (The 2012-15 drought in the semiarid region of Northeast Brazil in the historical context). Climanálise, 3 (1), 1–6.
Marengo, J. A., R. R. Torres, and L. M. Alves, 2017: Drought in Northeast Brazil—Past, present, and future. Theor. Appl. Climatol., 129, 1189–1200, https://doi.org/10.1007/s00704-016-1840-8.
Marengo, J. A., L. M. Alves, R. Alvala, A. P. Cunha, S. Brito, and O. L. Moraes, 2018: Climatic characteristics of the 2010-2016 drought in the semiarid Northeast Brazil region. An. Acad. Bras. Cienc., 90, 1973–1985, https://doi.org/10.1590/0001-3765201720170206.
Mattedi, M. A., and I. C. Butzke, 2001: A relação entre o social e o natural nas abordagens de hazards e de desastres (The relation between the social and the natural in the approach of hazards and disasters). Ambiente Soc., 4, 93–114, https://doi.org/10.1590/S1414-753X2001000900006.
Mingoti, S. A., 2005: Análise de Dados Através de Métodos Estatísticos Multivariados: Uma Abordagem Aplicada (Data Analysis Using Multivariate Statistics Methods: An Applied Approach). UFMG, 297 pp.
Mishra, S. K., V. B. Rao, and M. A. Gan, 2001: Structure and evolution of the large-scale flow and an embedded upper-tropospheric cyclonic vortex over Northeast Brazil. Mon. Wea. Rev., 129, 1673–1688, https://doi.org/10.1175/1520-0493(2001)129<1673:SAEOTL>2.0.CO;2.
Montgomery, D. C., E. A. Peck, and G. G. Vining, 2012: Introduction to Linear Regression Analysis. Vol. 821, John Wiley and Sons, 704 pp.
Moura, A. D., and J. Shukla, 1981: On the dynamics of droughts in Northeast Brazil: Observations, theory, and numerical experiments with a general circulation model. J. Atmos. Sci., 38, 2653–2675, https://doi.org/10.1175/1520-0469(1981)038<2653:OTDODI>2.0.CO;2.
Moura, M. O., C. Cunico, R. S. Nóbrega, and C. C. Duarte, 2016: Desastres hidrometeorológicos na região nordeste do Brasil: Distribuição espaço-temporal dos reconhecimentos de estado de calamidade pública (Hydrometeorological disasters in the northeastern region of Brazil: Spatio-temporal distribution of the state of public calamity recognition). Cad. Geogr., 26, 259–271, https://doi.org/10.5752/p.2318-2962.2016v26nesp2p259.
Nascimento, F. C. A., C. C. Braga, and F. R. C. D. Araujo, 2017: Análise estatística dos eventos secos e chuvosos de precipitação do Estado do Maranhão (Statistical analysis of dry and rainy event of Maranhão rainfall). Rev. Bras. Meteor., 32, 375–386, https://doi.org/10.1590/0102-77863230005.
Ning, S., J. Wang, J. Jin, and H. Ishidaira, 2016: Assessment of the latest GPM-era high-resolution satellite precipitation products by comparison with observation gauge data over the Chinese mainland. Water, 8, 481, https://doi.org/10.3390/w8110481.
Nóbrega, J. N. D., C. A. Dos Santos, O. M. Gomes, B. G. Bezerra, and J. I. B. D. Brito, 2014: Eventos extremos de precipitação nas mesorregiões da Paraíba e suas relações com a TSM dos oceanos tropicais (Extreme precipitation events in the mesoregions of Paraíba and its relationship with the tropical oceans SST). Rev. Bras. Meteor., 29, 197–208, https://doi.org/10.1590/S0102-77862014000200005.
Olímpio, J. L. S., P. M. B. Vieira, M. E. Zanella, and M. C. L. Sales, 2013: Episódios pluviais extremos e a vulnerabilidade socioambiental do município de Fortaleza: O episódio do dia 27/03/2012 (Extreme pluvial episodes and the socio-environmental vulnerability of the municipality of Fortaleza: The episode of the day 03/27/2012). Geo UERJ, 1, 181–206, https://doi.org/10.12957/geouerj.2013.5074.
Oliveira, D. H. M. C., and K. C. Lima, 2019: What is the return period of intense rainfall events in the capital cities of the northeast region of Brazil? Atmos. Sci. Lett., 20, e934, https://doi.org/10.1002/asl.934.
Oliveira, G. C. S., K. M. A. S. Alves, and R. S. Nóbrega, 2010: Sistemas atmosféricos dinâmicos de grande escala responsáveis pela ocorrência de chuva no Nordeste Brasileiro (Large-scale dynamic atmospheric systems responsible for the occurrence of rain in Northeast Brazil). II Workshop de Mudanças Climáticas e Recursos Hídricos de Pernambuco, Recife, Brazil, Mudanças Climáticas e Recursos Hídricos, 1–8.
Oliveira, P. T., C. M. Santos e Silva, and K. C. Lima, 2013a: Linear trend of occurrence and intensity of heavy rainfall events on Northeast Brazil. Atmos. Sci. Lett., 15, 172–177, https://doi.org/10.1002/asl2.484.
Oliveira, P. T., K. C. Lima, and C. M. Santos e Silva, 2013b: Synoptic environment associated with heavy rainfall events on the coastland of Northeast Brazil. Adv. Geosci., 35, 73–78, https://doi.org/10.5194/adgeo-35-73-2013.
Oliveira, P. T., C. M. Santos e Silva, and K. C. Lima, 2017: Climatology and trend analysis of extreme precipitation in subregions of Northeast Brazil. Theor. Appl. Climatol., 130, 77–90, https://doi.org/10.1007/s00704-016-1865-z.
Orlanski, I., 1975: A rational subdivision of scales for atmospheric processes. Bull. Amer. Meteor. Soc., 56, 527–530, https://doi.org/10.1175/1520-0477-56.5.527.
Oyama, M. D., and C. A. Nobre, 2004: Climatic consequences of a large-scale desertification in Northeast Brazil: A GCM simulation study. J. Climate, 17, 3203–3213, https://doi.org/10.1175/1520-0442(2004)017<3203:CCOALD>2.0.CO;2.
Palharini, R. S. A., and D. A. Vila, 2017: Climatological behavior of precipitating clouds in the northeast region of Brazil. Adv. Meteor., 2017, 5916150, https://doi.org/10.1155/2017/5916150.
Paredes-Trejo, F. J., H. Barbosa, and E. Guevara, 2015: Análisis espacial y temporal de las sequias en el Nordeste de Brasil (Spatial and temporal analysis of droughts in Northeast Brazil). Agriscientia, 32 (1), 1–14, https://doi.org/10.31047/1668.298x.v32.n1.16552.
Paredes-Trejo, F. J., H. A. Barbosa, and T. L. Kumar, 2017: Validating CHIRPS-based satellite precipitation estimates in Northeast Brazil. J. Arid Environ., 139, 26–40, https://doi.org/10.1016/j.jaridenv.2016.12.009.
Patz, J., and Coauthors, 2012: Our planet, our health, our future. Human health and the Rio Conventions: Biological diversity, climate change and desertification. WHO Rep., 64 pp., https://www.who.int/globalchange/publications/reports/health_rioconventions.pdf.
Pereira, G. R., and M. D. Z. Cuellar, 2015: Conflitos pela água em tempos de seca no Baixo Jaguaribe, Estado do Ceará (Water conflicts in times of drought in the Lower Jaguaribe, State of Ceará). Estud. Avanç., 29, 115–137, https://doi.org/10.1590/S0103-40142015000200008.
Pereira de Oliveira, F., and M. D. Oyama, 2015: Antecedent atmospheric conditions related to squall-line initiation over the northern coast of Brazil in July. Wea. Forecasting, 30, 1254–1264, https://doi.org/10.1175/WAF-D-14-00120.1.
Rao, V. B., K. Hada, and D. L. Herdies, 1995: On the severe drought of 1993 in north-east Brazil. Int. J. Climatol., 15, 697–704, https://doi.org/10.1002/joc.3370150608.
Rao, V. B., S. H. Franchito, C. M. Santo, and M. A. Gan, 2016: An update on the rainfall characteristics of Brazil: Seasonal variations and trends in 1979–2011. Int. J. Climatol., 36, 291–302, https://doi.org/10.1002/joc.4345.
Reboita, M. S., M. A. Gan, R. P. D. Rocha, and T. Ambrizzi, 2010: Regimes de precipitação na América do Sul: Uma revisão bibliográfica (Precipitation regimes in South America: A bibliographic review). Rev. Bras. Meteor., 25, 185–204, https://doi.org/10.1590/S0102-77862010000200004.
Rigby, R. A., and D. M. Stasinopoulos, 2005: Generalized additive models for location, scale and shape. J. Roy. Stat. Soc., 54C, 507–554, https://doi.org/10.1111/j.1467-9876.2005.00510.x.
Rigby, R. M., and D. M. Stasinopoulos, 2008: Generalized additive models for location scale and shape (GAMLSS) in R. J. Stat. Software, 23, 1–46, https://www.jstatsoft.org/article/view/v023i07.
Rodrigues, C., and I. M. C. Gouveia, 2013: Importância do fator antrópico na redefinição de processos geomorfológicos e riscos associados em áreas urbanizadas do meio tropical úmido: Exemplos na Grande São Paulo (Importance of the anthropic factor in the redefinition of geomorphological processes and associated risks in urban areas of the humid tropical environment: Examples in Greater São Paulo). Processos Erosivos e Recuperação de Áreas Degradadas, 1st ed. São Paulo Oficina de Textos, 66–94.
Rodrigues, D. T., W. A. Gonçalves, M. H. C. Spyrides, and C. M. Santos e Silva, 2020a: Spatial and temporal assessment of the extreme and daily precipitation of the Tropical Rainfall Measuring Mission satellite in Northeast Brazil. Int. J. Remote Sens., 41, 549–572, https://doi.org/10.1080/01431161.2019.1643940.
Rodrigues, D. T., W. A. Gonçalves, M. H. C. Spyrides, C. M. Santos e Silva, and D. O. de Souza, 2020b: Spatial distribution of the level of return of extreme precipitation events in Northeast Brazil. Int. J. Climatol., 40, 5098–5113, https://doi.org/10.1002/joc.6507.
Rodrigues, R. R., R. J. Haarsma, E. J. Campos, and T. Ambrizzi, 2011: The impacts of inter–El Niño variability on the tropical Atlantic and Northeast Brazil climate. J. Climate, 24, 3402–3422, https://doi.org/10.1175/2011JCLI3983.1.
Silva, J., 2009: Mapeamento de inundações no Brasil: Proposta de gestão ambiental através de um sistema de informações geográficas (Flood mapping in Brazil: Proposal for environmental management through a geographic information system). IX Seminário de Pós-Graduação em Geografia, Rio Claro, Brazil, UNESP, http://www.ambiente-augm.ufscar.br/uploads/A2-045.pdf.
Silva, N. T., 2014: Precipitações diárias intensas na cidade de João Pessoa, Paraíba (Intense daily rainfall in the city of João Pessoa, Paraíba). Universidade Federal da Paraíba Departamento de Geociências Centro de Ciências Exatas e da Natureza Rep., 70 pp., https://repositorio.ufpb.br/jspui/bitstream/123456789/679/1/NTS08082014.pdf.
Silva, V. M. D. A., M. D. C. M. Patrício, V. H. D. A. Ribeiro, and R. M. D. E. Medeiros, 2013: O desastre seca no Nordeste Brasileiro (The drought disaster in Northeast Brazil). Polêmica, 12, 284–293.
Souza, W. M., and P. V. Azevedo, 2012: Índices de detecção de mudanças climáticas derivados da precipitação pluviométrica e das temperaturas em recife-PE (Detection indexes derived from climate change rainfall and temperatures in recife-PE). Rev. Bras. Geogr. Fís., 5, 143–159, https://doi.org/10.26848/rbgf.v5i1.232793.
Souza, W. M., P. V. de Azevedo, and L. E. de Araújo, 2012: Classificação da precipitação diária e impactos decorrentes dos desastres associados às chuvas na cidade do Recife-PE (Classification of daily precipitation and impacts from disasters associated with rainfall in the city of Recife-PE). Rev. Bras. Geogr. Fís., 5, 250–268, https://doi.org/10.26848/rbgf.v5i2.232788.
Spinoni, J., G. Naumann, H. Carrao, P. Barbosa, and J. Vogt, 2014: World drought frequency, duration, and severity for 1951–2010. Int. J. Climatol., 34, 2792–2804, https://doi.org/10.1002/joc.3875.
Stanke, C., M. Kerac, C. Prudhomme, J. Medlock, and V. Murray, 2013: Health effects of drought: A systematic review of the evidence. PLOS Curr., 5, 5, https://doi.org/10.1371/currents.dis.7a2cee9e980f91ad7697b570bcc4b004.
Stasinopoulos, M. D., R. A. Rigby, G. Z. Heller V. Voudouris, and F. De Bastiani, 2017: Flexible Regression and Smoothing: Using gamlss in R. CRC Press, 572 pp.
Tomaziello, A. C. N., L. M. Carvalho, A. W. Gandu, 2015: Intraseasonal variability of the Atlantic intertropical convergence zone during austral summer and winter. Climate Dyn., 47, 1717–1733, https://doi.org/10.1007/s00382-015-2929-y.
United Nations Office for Disaster Risk Reduction, 2020: Terminology. Accessed 14 August 2020, https://www.undrr.org/terminology.
Uvo, C. B., C. A. Repelli, S. E. Zebiak, and Y. Kushnir, 1998: The relationships between tropical Pacific and Atlantic SST and Northeast Brazil monthly precipitation. J. Climate, 11, 551–562, https://doi.org/10.1175/1520-0442(1998)011<0551:TRBTPA>2.0.CO;2.
Winkelmann, R., 1995: Duration dependence and dispersion in count-data models. J. Bus. Econ. Stat., 13, 467–474, https://doi.org/10.1080/07350015.1995.10524620.
Winkelmann, R., 2008: Econometric Analysis of Count Data. Springer Science & Business Media, 316 pp.
Xavier, A. C., C. W. King, and B. R. Scanlon, 2016: Daily gridded meteorological variables in Brazil (1980-2013). Int. J. Climatol., 36, 2644–2659, https://doi.org/10.1002/joc.4518.
Xavier, D. R., C. Barcellos, H. D. S. Barros, M. D. A. F. M. Magalhães, V. P. D. Matos, and M. D. M. Pedroso, 2014: Organização, disponibilização e possibilidades de análise de dados sobre desastres de origem climática e seus impactos sobre a saúde no Brasil (Organization, availability and possibilities of data analysis on disasters of climatic origin and their impacts on health in Brazil). Cienc. Saude Colet., 19, 3657–3668, https://doi.org/10.1590/1413-81232014199.00992014.
Zanella, M. E., M. C. L. Sales, and N. J. A. Abreu, 2009: Análise das precipitações diárias intensas e impactos gerados em Fortaleza, CE (analysis of intense daily precipitations and impacts generated in Fortaleza, CE). GEOUSP Espaço Tempo, 13, 53–68, https://doi.org/10.11606/issn.2179-0892.geousp.2009.74112.
Zhang, X., and F. Yang, 2004: RClimDex (1.0) user manual. Environment Canada Climate Research Branch Doc., 22 pp.
Zhang, X., G. Hegerl, F. W. Zwiers, and J. Kenyon, 2005: Avoiding inhomogeneity in percentile-based indices of temperature extremes. J. Climate, 18, 1641–1651, https://doi.org/10.1175/JCLI3366.1.
Zhou, J., and K. M. Lau, 2001: Principal modes of interannual and decadal variability of summer rainfall over South America. Int. J. Climatol., 21, 1623–1644, https://doi.org/10.1002/joc.700.
