ARTIGO 02.pdf
Documento PDF (8.3MB)

Exibir conteúdo Baixar

                    Climate Dynamics (2020) 55:2327–2341
https://doi.org/10.1007/s00382-020-05385-1

Climatic aspects and vertical structure circulation associated
with the severe drought in Northeast Brazil (2012–2016)
Felipe Jeferson de Medeiros1

· Cristiano Prestrelo de Oliveira1,2 · Roger Rodrigues Torres3

Received: 16 March 2020 / Accepted: 13 July 2020 / Published online: 18 July 2020
© Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract
This study provides an updated dynamical analysis of the observed climate conditions during the longest, most severe and
continuous drought (2012–2016) ever recorded in Northeast Brazil (NEB). This investigation was carried out based on
temperature, precipitation, sea surface temperature, outgoing long-wave radiation, and ERA-Interim reanalysis data from
1981 to 2016. This was done by first looking for the area of the NEB where the drought was most severe, and then calculating the main rainy season anomalies and atmospheric parameters from 2012 to 2016 in relation to the climatological
conditions (1981–2010). Results show this drought event was influenced mainly by two distinct oceanic conditions that
induced anomalous atmospheric circulation patterns. The observed negative precipitation anomalies were firstly related to
the northward displacement of the Intertropical Convergence Zone (ITCZ), followed by an anomalous upward motion over
the Western Amazon, as well as the anomalous precipitation deficit along the South Atlantic Convergence Zone, that was
slightly displaced southward of the climatological position. From 2015 to 2016 the subsidence over NEB Brazil was linked
to the descending branch of the Walker circulation, as well as to the strong anomalous flow induced by convection in the
ITCZ over Africa and Gulf of Mexico.
Keywords Drought dynamics · Precipitation · Climate variability · ENSO · Subsidence

1 Introduction
Northeast Brazil (NEB) covers an area of 1588.196 km2,
of which 1006.654 km2 is part of the so-called Brazilian
semiarid zone, the location of 1171 municipalities and ~ 26
million inhabitants, considering only those included in the
Northeast Brazil states (Condel 2017). According to Article 2 of Resolution 107/2017 from the Ministry of National
Integration (Brasil 2017), municipalities in semiarid zones
have at least one of the following criteria: (i) average annual
rainfall of 800 mm or less; (ii) Thorthwaite aridity index
* Felipe Jeferson de Medeiros
felipetkd_@hotmail.com
1

Graduate Program in Climate Sciences, Federal
University of Rio Grande do Norte (UFRN), Natal,
Rio Grande do Norte, Brazil

2

Department of Atmospheric and Climate Sciences, Federal
University of Rio Grande do Norte (UFRN), Natal,
Rio Grande do Norte, Brazil

3

Natural Resources Institute, Federal University of Itajubá
(UNIFEI), Itajubá, Minas Gerais, Brazil

equal to or less than 0.50; and (iii) daily percentage of water
deficit equal to or greater than 60%, considering every day
of the year. Therefore, the NEB is characterized by rainfall scarcity and considered a region vulnerable to climate
extremes, mainly those related to droughts conditions.
Projected climate change scenarios indicate that dry spells
events become worst in the near and long-term future, with
decrease in water availability for irrigated agriculture and
human use owing to reductions in precipitation and increases
in evapotranspiration and air temperature (Marengo et al.
2016). In NEB region, droughts events affect more people
than any other natural hazard owing to their large scale and
long-lasting nature (Marengo et al. 2019). In this context,
several major droughts have been recorded and studied
in NEB. From 1603 to 2016, 34 years of severe drought
occurred, classified in 15 events as follows: 1614, 1723–24,
1776–77, 1790–94, 1824–25, 1877–79, 1900, 1915, 1919,
1941–44, 1958, 1982–83, 1992–93, 1997–98 and 2012–16
(Mossman 1919; Ferraz 1950; Hastenrath and Heller 1977;
Rao et al. 1995; Brito et al. 2017; Cunha et al. 2018; Medeiros et al. 2020b).

13

Vol.:(0123456789)

2328

The NEB region presents another adverse climatic factor: the rainfall distribution is not uniform. It varies from
year to year and from one sub-region with homogeneous
precipitation to others (Oliveira et al. 2017; Medeiros et al.
2018). Among the regions most affected by the spatial and
temporal variability of precipitation, the Northern Northeast
Brazil (NNEB) is the most vulnerable to droughts (Moura
and Shukla 1981; Marengo et al. 2017; Cunha et al. 2018).
It is estimated that in the drought of 2012–2016, approximately 85% of the region’s reservoirs contained less than
25% of their capacity (Getirana 2016; ANA 2017), which
caused negative reflections on activities like agriculture,
tourism and recreation, electricity generation, urban water
supply and transportation (Wilhite et al. 2014). This drought
affected almost 33.4 million people, and caused an estimated
damage of approximately $US 30 billion (Marengo et al.
2017). Continuous drought not only directly affects people’s
water and food security, but may also change surface landscapes (such as land cover and soil properties), and further
affect the processes of runoff yield and flow concentration,
leading to a series of ecological and environmental problems
(Pereira et al. 2014).
In NEB drought episodes are often associated with
large-scale phenomena such as the El Niño-Southern
Oscillation (ENSO) (Ropelewski and Halpert 1987;
Grimm 2003; Tedeschi et al. 2015; Cai et al. 2020) and
Sea Surface Temperature (SST) anomalies in the tropical
Atlantic (Moura and Shukla 1981; De Souza et al. 2005;
Andreoli and Kayano 2004; Hastenrath 2006; Amorim
et al. 2014; Foltz et al. 2019). These phenomena produce
changes in the atmospheric-oceanic system that influence
the NEB climate. During El Niño (EN) episodes, there is
a tendency to inhibition of convective activity over NEB,
associated with an anomalous Walker circulation with its
descending branch over NEB and surrounding regions,
which leads to a drier than normal NNEB rainy season
(Chaves and Cavalcanti 2001; Hastenrath 2006; Rodrigues
et al. 2011). On the other hand, during La Niña episodes,
the anomalous Walker circulation becomes inverted, which
results in an enhanced convective process over NEB, with
associated higher precipitation (Grimm 2004; Manatasa
and Mukwada 2017). However, it is important to highlight
that although the large majority of severe drought events in
NEB region is associated with the occurrence of El Niño,
this is not always the case (Kane 1997). In relation to the
Atlantic Ocean, droughts in NEB are observed when the
simultaneous existence of a heat source to the north and
a cold sink to the south of the Equator is present (Nobre
and Shukla 1996). Such meridional gradient hydrostatically controls the sea level pressure and wind pattern over
the equatorial Atlantic, setting a scene for an anomalously
steep interhemispheric northward SST gradient, stronger
southerly wind component, and thus anomalous northward

13

F. J. de Medeiros et al.

migration of the ITCZ and consequently deficient NEB
rainfall (Curtis and Hastenrath 1995; Kayano et al. 2013).
Others teleconnection patterns, such as the Madden–Julian
Oscillation (MJO), also influence the circulation and precipitation in northern South America (Oliveira et al. 2016;
Valadão et al. 2017). Rodrigues et al. (2019) showed that
the tropical convection and the related wind convergence
in the eastern Indian Ocean associated with the passage
of the MJO leaded to atmospheric blocking over the
western South Atlantic as part of planetary Rossby wave
train, which, in turn, resulted in one of the worst drought
(2013–2014) recorded in eastern South America.
In particular, the beginning (2011–2012) of the drought
event in NEB (2012–2016) has been documented as not
related to El Niño or warm SST over the tropical North
Atlantic (Rodrigues and McPhaden 2014; Jiménez-Muñoz
et al. 2019). Rodrigues and McPhaden (2014) showed that
NEB rainy season (March–May) of 2012 was preceded
by negative SST anomalies in the central Pacific Ocean,
whereas in the Atlantic Ocean neutrality conditions dominated during 2011–2012. Nevertheless, below-average precipitation over NEB was verified, being the austral autumn
of 2012 the driest between 1961 and 2016 (Marengo et al.
2017; Medeiros et al. 2020b). During 2012–2016 many
drought episodes were detected not only in NEB, but in the
other regions of Brazil, such as Amazon (Jiménez-Muñoz
et al. 2016, 2019; Aragão et al. 2018) and Southeast (Coelho
et al. 2015, 2016), and worldwide (Griffin and Anchukaitis
2014; Funk et al. 2016; Baudoin et al. 2017; Lin et al. 2017;
Stojanovic et al. 2018; Mo and Letternmaier 2018).
Specifically, for the recent drought in NEB, several studies have been conducted (Pereira et al. 2014; Rodrigues and
McPhaden 2014; Cunha et al. 2015, 2018, 2019; Marengo
et al. 2016, 2017; Getirana 2016; Martins et al. 2017; Brito
et al. 2017; Azevedo et al. 2018; Barbosa et al. 2019). However, these studies do not provide a comprehensive background to some atmospheric dynamics and oceanic aspects
related to the observed climate variability in NEB during
the whole period (2012–2016). The exception is the paper of
Marengo et al. (2017). Nevertheless, the dynamical analyses
are restricted to sea surface temperature and Walker circulation. Thus, there is still a need for further studies to shed
light into advance the understanding of the regional-scale
mechanisms and atmospheric teleconnection patterns associated with the causes of the NEB drought during 2012–2016.
In this sense, given that this drought episode was the longest continuous dry spell on the NEB historical record (1583
onwards), and affected a large area with significant impacts
for population, as well as economic activities (Marengo et al.
2016; Brito et al. 2017; Barbosa et al. 2019), the aim of
this paper is to investigate the vertical structure circulation
associated with the intense and persistent drought event in
the Northeast Brazil during 2012–2016, which complements

Climatic aspects and vertical structure circulation associated with the severe drought in…

the large scale circulation patterns discussed by Marengo
et al. (2017).

2 Materials and methods
2.1 Observational data
The observational dataset used in this study was provided
by Xavier et al. (2016) and Gadelha et al. (2019). They
provide daily information about the following weather
variables: rainfall, maximum and minimum temperature,
wind speed and direction, relative humidity, solar radiation and evapotranspiration. This dataset was built using
3625 rain gauges and 735 weather stations covering all of
Brazil’s territory for the 1980–2016 period, being arranged
in a regular 0.25° × 0.25° grid (Xavier et al. 2016). For the
NNEB region, the precipitation dataset considered 446 rain
gauges, collected from different sources such as conventional (57) and automatic (78) rain gauges from “Instituto
Nacional de Meteorologia” (INMET), and conventional
rain gauges (311) from the “Agência Nacional de Águas”
(ANA). To investigate the 2012–2016 drought event and
its impacts, the variables used were precipitation, maximum and minimum temperatures. In this sense, to obtain
the accumulated precipitation anomalies we first calculate
the annual climatology based on the period 1981–2010 and
then annual anomalies from 2012 to 2016 are obtained by
subtracting them from the climatological period. For the seasonal analyzes similar procedure was adopted. We calculate
the seasonal climatologies based on the period 1981–2010
and then anomalies are obtained by subtracting them from
each season during the 1980–2016 period. Although Cunha
et al. (2019) have showed that between 2011 to 2017 drought
events were severe and widespread over Brazil, we restrict
our analysis for the period 2012–2016 due to Marengo et al.
(2017) have shown that the rainfall over the NEB Brazil in
2011 was above average, as well as the observational data
used (Xavier et al. 2016) be available until 2016. Furthermore, according to Medeiros et al. (2020a) the precipitation anomalies observed in the northern sector of the semiarid region in 2017 was close to the climatological average
(− 17.0 mm).

2.2 Atmospheric and oceanic reanalysis data
The meteorological fields of monthly means of zonal and
meridional wind, temperature, specific humidity, and the
vertical motion (omega) from 1981 to 2016 of the ERAInterim reanalyze, provided by the European Center for
Medium-Range Weather Forecasting (ECMWF) in a
1.5° × 1.5° grid were used (Dee et al. 2011). These variables are used from the following pressure levels: 1000,

2329

975, 950, 925, 900, 875, 850, 825, 800, 775, 750, 700, 650,
600, 550, 500, 450, 400, 350, 300, 250, 225, 200, 150, 125
and, 100 hPa. The Outgoing Long-Wave Radiation (OLR)
dataset (Liebmann and Smith 1996) is used to investigate
the regions with deep convection and was obtained from
the Climate Diagnostics Center of the National Oceanic and
Atmospheric Administration (NOAA-CDC) in a 2.5° × 2.5°
grid.
To analyze ocean conditions, we used the NOAA
Extended Reconstructed Sea Surface Temperature (ERRST)
v3b dataset (Smith et al. 2008). These data consist of reconstructed monthly sea-surface temperature for the whole
globe on a 2.0° × 2.0° grid. However, to identify the atmospheric circulation pattern associated with NEB drought
conditions, several studies proposed the evaluation of both
ENSO and SST gradient in the tropical Atlantic in order to
know which mode is most important to NEB interannual
rainfall variability (Moura and Shukla 1981; Hastenrath
2012; Kayano et al. 2013). Therefore, to identify and evaluate the intensity of ENSO events the Oceanic Niño Index
(ONI) are applied, and to investigate the role of the tropical
Atlantic, the cross-equatorial sea surface temperature gradient (GRAD) are select. Only the months from March to May
(MAM) are considered because the main focus is the NNEB
rainy season related to those ocean’s conditions.
The ONI categorizes warm (El Niño) and cold (La Niña)
phases of the ENSO according to the three-month running
mean of Sea Surface Temperature Anomaly (SSTA) (based
on centered 30-year base periods updated every 5 years) in
the Niño 3.4 region (5° N–5° S, 120° W–170° W). A given
period is defined as an El Niño (La Niña) event if the ONI
value is higher (lower) than a + 0.5 °C (− 0.5 °C) threshold
for at least five consecutive overlapping seasons of the year.
Based on the analysis of the Climate Prediction Center—
CPC/NOAA (https​://origi​n.cpc.ncep.noaa.gov/produ​cts/
analy​sis_monit​oring​/ensos​tuff/ONI_v5.php), the oceanic
conditions in the eastern equatorial Pacific during the NEB
rainy season (MAM) in 2012, 2013 and 2014 were neutral,
whereas in 2015 and 2016, an El Niño event occurred. In
addition, during the 3-month average from November 2015
to January 2016 the ONI value had peaked at 2.6 °C, representing the highest anomaly since 1981. Thus, we compare
the composite of the rainy seasons in 2012–2014 with the
composite of the 20 neutral years observed in the period
1981–2010 (1981, 1984, 1986, 1988, 1990, 1991, 1993,
1994, 1995, 1996, 1997, 2001, 2002, 2003, 2004, 2005,
2006, 2007, 2009, 2010). The same analysis was performed
with the rainy seasons of the years 2015 and 2016 that
was compared to the 5 El Niño years observed between
1981–2010 (1982, 1983, 1987, 1992, 1998). Therefore,
30 years are used in this study to perform the dynamics
analysis of the ocean conditions and atmospheric circulation
during the drought period. These analyses will allow a better

13

2330

F. J. de Medeiros et al.

understanding of the atmospheric characteristics observed
during the neutral period (2012–2014) and ENSO positive
phase (2015–2016).
In order to investigate the GRAD mode in more detail,
monthly SST time series obtained from the CPC/NOAA
(https ​ : //www.cpc.ncep.noaa.gov/data/indic ​ e s/) for the
Tropical North Atlantic (TNA) (5°–20° N, 60°–30° W) and
Tropical South Atlantic (TSA) (0°–20° S, 30° W–10° E)
are used. The monthly SSTA averaged in each area were
obtained. Then, the seasonally averaged TNA and TSA time
series were obtained for MAM, which were used to calculate
the GRAD index, defined as TNA-TSA difference. Positive
(negative) GRAD index denotes the northward (southward)
SST gradient over the intertropical Atlantic basin (Servain
1991; De Souza et al. 2005; Amorim et al. 2014).
With the goal of identifying the changes in the regional
Hadley and Walker circulations in the neutral and El Niño
years during 2012–2016, we calculated these circulations
defining the standard bounded of the regional Hadley circulation over Northeast Brazil (90° N–90° S, 30° W–50°
W), and the Walker circulation around the equatorial region
(0°–360°, 10° S–0). This latitudinal region was chosen to
analyze the position and intensity of the ITCZ, that according to Coelho et al. (2004) and Utida et al. (2019) is the
major rain atmospheric system for NEB. The regional
Hadley circulation as well as the anomalous fields of SST,
OLR, velocity potential and divergent wind at 200 hPa was
explore to verify the surface, and the upper and lower level
convergence-divergence over NEB and surroundings regions
associated with the ITCZ position.

2.3 Thermodynamical aspects
To analyze the atmospheric thermodynamics, the potential
temperature (θ), equivalent potential temperature ( 𝜃e ) and
saturated equivalent potential temperature (𝜃es) were calculated also using the ERA-Interim reanalysis. We used the
formulations recommended by Betts and Dugan (1973) and
Bolton (1980).

𝜃 = Tk

(

1000
𝜌

𝜃e = 𝜃 × exp

)0.2854(1−0.28×10−3 r)

[(

𝜃es = 𝜃 × exp

(

)
]
3.376
3
− 0.00254 × r(1 + 0.81 × 10 − r)
TL

Lv rs
Cpd Tk

)

where Tk is the absolute temperature, ρ is the partial pressure of dry air, r and rs are the mixing ration and saturation
mixing for water vapor, TL is the absolute temperature at the

13

lifting condensation level, Lv is the latent heat of vaporization, and Cpd is the specific heat for dry air.

2.4 Statistical analysis of the climatic time series
In order to identify trends in the annual and seasonal time
series of precipitation, maximum and minimum temperature,
the nonparametric Man-Kendall (MK) test (Mann 1945;
Kendall 1975) was applied. This test compares each value
of the time series with the remaining values in sequential
order, counting the number of times that the remaining terms
are greater than the analyzed value. The null hypothesis Ho
of the MK test is that there is no trend of the calculated
time series, while the alternative hypothesis H1 says there is
trend (positive or negative). This method is considered one
of the most complete for identification of changes in time
series and is widely used for trend detection in meteorological series (Oliveira et al. 2014; Ye et al. 2016; Espinoza et al.
2018; Cabral Júnior et al. 2019). All of these statistical tests
are evaluated using significance of at least 99% (p < 0.01).
The null hypothesis Ho should be rejected if the p value is
less than the significance level.

3 Results and discussion
3.1 Rainfall anomaly distribution during the 2012–
2016 Northeast Brazilian drought
Figure 1a shows the spatial distribution of precipitation
over the 30-year period from 1981–2010 in Brazil. During
this period, the mean annual rainfall in NEB ranged from
2000 mm in the Northwest to lower than 600 mm in many
places over the central portion of the northeastern region.
Although the coastland and extreme northwestern regions
exhibited high annual rainfall totals, it can be seen that the
Northeast Brazil region is the driest in the country. In fact,
although different atmospheric systems act in NEB, namely
the Intertropical Convergence Zone, upper tropospheric
cyclonic vortex, easterly wave disturbances, sea and land
breezes, squall lines, front systems, among others (Molion
and Bernado 2002), the rainfall spatial pattern is heterogeneous and characterized by high space–time variability
(Oliveira et al. 2017).
In relation to annual rainfall anomalies during the period
from 2012 to 2016 (Fig. 1b), it can be seen that negative
anomalies were present in all of NEB’s territory, and in
many others parts of Brazil, such as Southeast and Central
Brazil. Therefore, the 2012–2016 NEB drought is part of a
much bigger area where precipitation well below average
prevailed. However, regarding its intensity, the largest precipitation deficits is clearly observed in NNEB (1.2° S–8°
S and 34.4° W–48.25° W). In terms of magnitude, over the

Climatic aspects and vertical structure circulation associated with the severe drought in…

2331

Fig. 1  a Spatial variation of
annual rainfall (mm) over Brazil
during 1981–2010 and b 2012–
2016 accumulated precipitation
anomalies (mm) in relation
to the average climatological
precipitation (1981–2010).
The rectangle in b defines the
Northern Northeast Brazil (area
between 1.2° S and 8° S and
34.4° W–48.25° W)

NEB the anomalies from 2012 to 2016 were − 353.2, − 73.5,
− 165.1, − 327.6 and − 143.7 mm, respectively, while over
the NNEB they were − 425.3, − 201.6, − 162.4, − 363.0
and − 243.1 mm, respectively (not shown). Therefore, the
accumulated precipitation anomalies over the whole period
were 1063.1 mm in NEB and 1395.4 mm in NNEB. Since
the spatial analysis of precipitation anomalies showed that
this drought event was more intense in NNEB, the following
climate characteristics analyses of the precipitation, maximum and minimum temperature, and the outgoing longwave
radiation anomalies are restricted to this area. Figure 1 also
emphasizes how the occurrence of drought events could
be detrimental to the central part of the Northeast Brazil,
considering that an occurrence of negative precipitation
anomalies in an area with climatological low average rainfall (less than 600 mm/year) have a huge impact on water
resources. This impact can be seen in the São Francisco river
basin, which reached on January 2016 just 5% of its volume
capacity (Martins et al. 2017), and in the Armando Ribeiro
Gonçalves and Coremas reservoirs—the most important
reservoirs of the Rio Grande do Norte and Paraiba states,
respectively—which approximately both lost 30% of its volume stored throughout the extreme dry year of 2012 (Medeiros et al. 2018).

3.2 Main rainy season in NNEB
Figure 2a shows the histogram of the monthly average rainfall over the 30-year period from 1981 to 2010 in the NNEB
region. The maximum rainfall occurred between February
and May (71.5%), while July to December concentrated less
than 15% of the annual rainfall. The months of June and
January, on the other hand, were transitional to the rainy and
dry season, respectively. Furthermore, according to Fig. 2a

Fig. 2  Histogram of the monthly average rainfall in NNEB in the a
climatological period (1981–2010) (%) and b a comparison between
the seasonal rainfall distribution during the 1981–2010 period and the
drought event (2012–2016) (mm)

only the months from March to May (MAM) accounted for
more than 55% of the total annual climatological volume,
being the main rainy season in NNEB, mainly due to the
ITCZ reaching its southernmost position in the equatorial

13

2332

South Atlantic around April (Waliser and Gautier 1993;
Drumond et al. 2010).
Comparing the rainfall between the 1981–2010 period
and during the drought event (2012–2016) on a seasonal
basis (Fig. 2b), it can be seen that the austral summer, winter and spring presented similar rainfall values. In fact, the
difference between the climatological average and drought
event is + 29.9 mm in summer, while in the winter and
spring are − 23.9 mm and − 11 mm, respectively. On the
other hand, in the austral autumn, the values are very different. In the climatological period, the average rainfall value
was 742.2 mm, while in the drought event it was 462.2 mm,
representing a water deficit of 37.8%. These characteristics,
associated with that observed and discussed in Sect. 3.1,
highlights how a severe drought event can be to water
resources and consequently for the NEB population, given
that the recompositing of water reservoir used for city supply
and agricultural use is restricted primarily to those months
of the short rainy season (MAM), which was highly affected
during the 5-year drought (Fig. 2b). For example, according
to the Brazilian National Water Agency—ANA, approximately 55% and 25% of the water demand in the NNEB
region (composed by ~ 940 municipalities with 350 water
reservoirs monitored by ANA) are used by the population
and agriculture, respectively (ANA 2018).
In addition, to further detail how the NNEB rainy season
(MAM) was affected from 2012 to 2016, we analyzed the
area average austral autumn 1981–2016 precipitation anomaly time series (Fig. 3a). It can be noted that from 2012 to
2016 the rainfall was − 840.3 mm below average, with the
austral autumn of 2012 standing out with the largest negative
precipitation anomaly of − 317.5 mm with respect to the
1981–2010 climatological mean for the region (573.5 mm).
Cunha et al. (2018) suggested that these anomalies could be
related to warming trends observed in the tropical Atlantic.
Examining the GRAD mode (Fig. 3d), that is strongly correlated to ITCZ position (Gloor et al. 2013; Kayano et al.
2018), positive trend is observed with 0.016 °C (no statistical significance) (Table 1). For the 2012–2016 period,
positive northward SST anomalous gradient in 2012, 2013
and 2016 probably displaced Atlantic ITCZ more to north
than usual, which may lead to less overall precipitation over
NNEB (Fig. 3a). The atmospheric circulation will be discussed further latter (Sect. 3.3) through an analysis of the
Hadley and Walker circulation.
Furthermore, examining the historical times series
of Fig. 3a one can notice that the NNEB region presents
a high rainfall variability. For example, during the years
from 1981 to 1997, the precipitation was below normal
from 1981 to 1983, followed by a period of excess rainfall
between 1984 and 1989, returning to a period of precipitation deficit between 1990 and 1993, and again a period
of excess between 1994 and 1997. However, from 2001

13

F. J. de Medeiros et al.

Fig. 3  Observed average area austral autumn (March–April–May)
time series anomalies during 1981–2016 in Northern Northeast Brazil, defined by the rectangle in Fig. 1b. a Xavier et al. (2016) precipitation anomalies (colored bars) expressed in mm. Temperature
maximum b and minimum c extracted from Xavier et al. (2016)
dataset. d Cross-equatorial sea surface temperature gradient (GRAD)
index, defined as TNA-TSA difference, extracted from the Climate
Prediction Center (CPC), and e outgoing longwave radiation from
the NOAA-CDC reanalysis. The long-dashed line indicates the time
series tendency and the red box highlight the drought event period
(2012–2016). Anomalies are computed with respect to the 1981–
2010 climatological mean

Climatic aspects and vertical structure circulation associated with the severe drought in…

2333

Table 1  Trend statistical of the Mann–Kendall test applied to annual
time series anomalies of precipitation and maximum/minimum temperature, and seasonal time series anomalies (MAM) of precipitation,
maximum/minimum temperature, cross-equatorial sea surface temperature gradient index and outgoing longwave radiation for Northern
Northeast Brazil from 1981 to 2016, at significance level of 1%
Variables
Annual
Precipitation
Maximum temperature
Minimum temperature
Seasonal (MAM)
Precipitation
Maximum temperature
Minimum temperature
GRAD index
Outgoing longwave radiation

Test Z

p value

Significance

− 0.072
0.460
0.486

p > 0.05
p < 0.01
p < 0.01

*
*

− 0.143
0.316
0.414
0.016
0.167

p > 0.05
p < 0.01
p < 0.01
p > 0.05
p > 0.05

*
*

*Trend is significant at p < 0.01

to 2016, twelve out of sixteen rainy seasons experienced
rainfall below normal, indicating the multiannual nature of
this intense drought, as suggest by Marengo et al. (2016).
Despite the prevalence of negative anomalies and trend,
especially in the last 5 years, rainfall in NNEB does not show
either annual (not shown) or seasonal statistical significance
change (Table 1).
Still according to Fig. 3a, the second driest rainy season was 1983, with negative precipitation anomalies of
− 286.3 mm, 9.8% lower than 2012. In relation to intensity and duration, the intensity of the past events was lower
compared to the drought event during 2012–2016, although
negative precipitation anomalies were registered in several
consecutive years (1981–1983, 1990–1993; 2001–2005). For
example, in 1981–1983 and 1990–1993, the MAM anomalies were equal to − 543.6 and − 705.1 mm, respectively. In
the 2001–2005 period, an event with similar duration to the
recent drought, the anomalies in MAM was − 378.7 mm,
i.e., 54.9% lower than 2012 to 2016. Thus, these results
clearly show the severity of the 2012–2016 drought event,
considered the one that affected the largest area in NEB
in the past decades, as previously indicated by Brito et al.
(2017).
In relation to maximum and minimum temperature
anomalies in MAM during the period from 1981 to 2016
(Fig. 3b, c), we found for both variables’ positive trends at
significance level of 1%, with 0.316 °C to maximum temperature, and 0.414 °C to minimum temperature (Table 1).
This behavior is also observed during the annual variation,
with some differences in terms of intensity (not shown).
For the drought event period (2012–2016), the maximum
temperature anomalies were 1.5, 1.3, 0.5, 0.9 and 1.7 °C,
respectively, indicating that the NNEB was afflicted not only

Fig. 4  Mean vertical profiles of potential temperature (θ), equivalent
potential temperature (𝜃e), and saturated equivalent potential temperature (𝜃es) for March–May 2012–2016 at the Northern Northeast Brazil

by negative precipitation anomalies, but also by higher air
temperatures. This warming is primarily caused by increased
surface solar radiation due to lower cloud cover, as indicated by the positive OLR anomalies (Fig. 3e). In addition,
the large difference between the average vertical profile of
potential temperature (θ) and equivalent potential temperature (𝜃e ) (Fig. 4) indicates a dry layer and absence of moisture in the lower and middle atmosphere, contributing to
the unfavorable scenario for convection. Furthermore, the
small decrease of equivalent potential temperature (𝜃e ) and
saturated equivalent potential temperature (𝜃es) confirms on
average the static stability conditions of the atmosphere.
This stable condition was also observed in the climatological
average (not shown), which indicates that climatologically
over the NEB the large-scale dynamical conditions are more
important than the thermodynamics aspects. Regarding the
minimum temperature (Fig. 3c), similar behavior was found,
i.e., from 2012 to 2016 the minimum temperature anomalies also presented a rise in magnitude, with values of 0.1,
0.8, 0.3, 0.3 and 1.0 °C, respectively. This analysis indicates
that during the drought event, there was an increase in the
frequency of hot spells and a decrease in the frequency of
cold nights, which is in accordance with other studies that
indicate this pattern over NEB along the beginning of the
twenty-first century (Bezerra et al. 2018; Silva et al. 2018).
All these changes in temperature and rainfall may significantly threaten not only the water resources, agriculture and
health, but also the ecosystems’ survivability (Cook et al.
2018).

3.3 Large‑scale dynamics
This section investigates the relationship between the Northern Northeast Brazil precipitation and the large-scale ocean

13

2334

and atmospheric patterns in order to identify possible forcing
mechanisms associated with the 2012–2016 drought event.
The first analysis concerns about the sea surface temperature anomalies (Fig. 5), which according to Coelho
et al. (2016) are generally considered a key slowly varying
parameter to understand climate anomalies. The anomalies
(difference in the climatological mean of the ENSO events
between 1981 and 2010) in blue shades represent waters
with below-normal temperature (negative anomaly), and red
shades indicate waters with above-average temperature (positive anomaly). During neutral years (2012–2014) (Fig. 5a),
SST anomalies were above average across most of the Niño
region in the Pacific Ocean, whereas in the Atlantic Ocean,
slightly negative anomalies prevailed. This indicates that
during the years of 2012–2014, the SST in the Niño (Atlantic) region was warmer (colder) than the other neutral years
recorded in the 1981–2010 period (see Sect. 2.2). In particular, these SST anomalies in the Atlantic Ocean did not favor
the southward migration of the ITCZ, since the interhemispheric SST gradient mode between the northern and southern portions was anomalous northward (Fig. 3d) (Moura and
Shukla 1981; Lucena et al. 2011; Amorim et al. 2014). Consequently, the negative precipitation anomalies over NNEB
(Fig. 3a) might be associated with this mode of climate variability, as illustrated by Rodrigues and McPhaden (2014)
which shown that the ITCZ does not migrate southward in
the austral summer of 2012.
During the 2015–2016 El Niño event (Fig. 5b), on the
other hand, negative sea surface temperature anomalies prevailed over the entire range of the Nino SST indices (Niño 1,
2, 3 and 4) in MAM, suggesting that over these regions the
sea surface temperatures were lower compared to previous
El Niño events between 1981 and 2010 (see Sect. 2.2). The
entire 2015–2016 El Niño event was one of the strongest
Fig. 5  a Anomalies of sea
surface temperature (°C) based
on the NOAA ERRST v3b
dataset during a 2012 to 2014
and b 2015 to 2016. Anomalies
are computed with respect to
the composite of 20 neutral
years and five El Niño years that
occurred in MAM according
to the ONI from 1981 to 2010.
The dashed line indicates the
NEB region

13

F. J. de Medeiros et al.

ever recorded (Jiménez-Muñoz et al. 2016), with six consecutive quarters having SSTA values above 2 °C, a pattern
not observed in any of the others events between 1981 and
2010. However, in MAM 2015 El Niño was just beginning
and still weak, and in MAM 2016 El Niño was ending and
also weak. Thus, since the MAM quarter of 2015 and 2016
corresponded to the beginning and end of the El Niño event,
and considering that the atmospheric teleconnection pattern
associated to ENSO signal on precipitation response is well
established only during moderate and strong El Niño (Rodrigues et al. 2011; Oliveira et al. 2018), possibly the prolonged
effected on the negative precipitation anomalies in NNEB
during 2015 and 2016 (Fig. 3a) cannot be exclusively related
to the El Niño, as suggested by Marengo et al. (2017), but
to a jointly influence of another atmospheric teleconnection pattern. Cunha et al. (2018) suggested that the warming
trends observed in the tropical Atlantic Ocean could be the
main source related to this recent severe drought in NEB. In
fact, the positive GRAD index in 2016 (Fig. 3d) indicates
that the TNA warmed anomalously than TSA, which may
lock the ITCZ position to stay northward with the shift leading to less precipitation over NNEB.
The patterns associated with the SSTA become more
evident when examining the austral autumn outgoing longwave anomalies, which are associated with enhanced or suppressed convection (Fig. 6). Negative OLR anomalies are
noted stretching from Venezuela toward Africa (at about 20°
N), and extending NW–SE from the Amazon region to the
Atlantic Ocean, associated with the cloud band of the South
Atlantic Convergence Zone from 2012 to 2014 (Fig. 6a).
Over the Northeast region of Brazil, positive OLR anomalies
are noticed, indicating weaker convection in these areas, in
line with the reduced regional rainfall (Fig. 3a). This OLR
anomaly pattern exposed in Fig. 6a sounds consistent with

Climatic aspects and vertical structure circulation associated with the severe drought in…

2335

Fig. 6  Outgoing longwave
radiation anomalies in W m−2
based on the NOAA-CDC reanalysis during a 2012 to 2014,
and b 2015 to 2016. Anomalies
are computed with respect to
the composite of 20 neutral
years and five El Niño years that
occurred in MAM according
to the ONI from 1981 to 2010.
The dashed line indicates the
NEB region

the Walker type circulation, where subsidence over the NEB
are associated to enhanced convection in the Amazon and
the South Atlantic Convergence Zone, as shown by Gandu
and Silva Dias (1998). In the 2015–2016 period (Fig. 6b),
negative OLR anomalies occurred around 170° E, indicating the presence of an enhanced convection structure acting
as a tropical heat source. To the east of this tropical heat
source, positive OLR anomalies occurred from 170° to 80°
W, indicating the suppressed convection over this region.
This tropical pattern is explained by a Walker circulation
cell with upward vertical motion over the region of negative
OLR, and downward vertical motion over the region of positive OLR, as suggested by Coelho et al. (2016), who found
similar behavior in the same area. Over Northeast Brazil,
despite the low intensity, positive OLR anomalies can be
notice, involving reduction of convection and consequently
precipitation.
As seen in the SSTA (Fig. 5) and the positive and negative regional OLR anomalies (Fig. 6) both the Walker and
Hadley cells can influence the precipitation in NNEB.
Therefore, we examined the characteristics of the vertical
structure of the regional (meridional and zonal) circulations averaged over MAM during neutral (2012–2014) and
positive phase of ENSO (2015–2016). The regional Hadley
circulation composite during neutral events (Fig. 7a) shows
strong ascending motion in the entire troposphere over the
equatorial region, accompanied by the subsidence in the
tropical northern and southern hemisphere. During the neutral years (2012–2014) (Fig. 7c) the pressure-latitude crosssections indicates that the upward motion was weaker than
normal between 5° N and 5° S, while from 20° S to 10° S,
and 10° N–20° N the descending branch of the Hadley cell
is stronger. The weaker upward movement in the equatorial

region and the stronger downward movement in the southern
and northern hemisphere in 2012–2014 caused the predominance of anomalously descending motions over the equatorial region and southern hemisphere latitudes (Fig. 7e).
This pattern helps to explain the positive longwave outgoing
anomalies over Northeast Brazil (Fig. 6a) and the negative
precipitation anomalies in NNEB (Fig. 3a). Similar analyses
carried out by De Souza and Ambrizzi (2002) and Drumond
et al. (2010) also showed that when the descending branch of
the Hadley circulation are observed over the tropical areas
of South America, negative precipitation anomalies are
observed in Northeastern Brazil.
The Hadley circulation composite cell in El Niño years
(Fig. 7b) had the same behavior as in neutral years (Fig. 7a),
but clearly is less intense when compared to the previous
one, as also illustrated by Freitas et al. (2017) and Oliveira
et al. (2018), especially over the equatorial region and in the
tropical northern hemisphere. During 2015–2016 (Fig. 7d)
the upward motion over 0°–5° N is weaker, while from 0°
to 10° S the upward motion is wider and stronger in relation
to previous El Niño years (Fig. 7b). On the other hand, the
downward motion is stronger at the southern and northern
hemisphere. As an outcome, the anomalous vertical motion
indicates stronger subsidence over the northern hemisphere,
and from 20° S to 10° S, while from 10° S to 0° ascending
motion are observed (Fig. 7f). In general, the Hadley circulation is weaker during the 2012–2016 period in the 50°
W–30° W cross section.
Figure 8c, d show the 2012–2016 austral autumn latitudinal cross-section with average vertical motion (omega) in
the 10° S–0° latitude band. Upward motion is noted between
120° E and 130° W, and over the Amazon latitudes, while
downward motion is observed mainly from 120° W to 80°

13

2336

F. J. de Medeiros et al.

Fig. 7  Meridional-vertical
circulation (Hadley) averaged in the longitudinal band
between 50° W and 30° W for
MAM during the a 20 neutral
composite years, b five El Niño
composite years, from c 2012 to
2014, and d 2015 to 2016. The
anomalous Hadley circulation
during 2012–2016 regarding the
neutral and El Niño years are
shown in (e) and (f). The shaded
plot represents vertical motion
(omega) in hPa s−1, which is
superimposed by wind vectors
in m/s

W both in neutral (2012–2014) (Fig. 8c) and El Niño years
(2015–2016) (Fig. 8d). In relation to neutral composite
years (Fig. 8a), the Walker circulation is stronger during
2012–2014, especially over the Western Amazon (Fig. 8e),
which matches with the occurrence of flood events reported
in Amazonia in 2014 (Ovando et al. 2016; Espinoza et al.
2018). In line with these results, Barichivich et al. (2018)
showed that the intensification of the Walker circulation
averaged over Amazonia is highly consistent with the
trend toward increased flooding in Manaus, and the associated increase in basin-wide wet season precipitation. This
upward motion over the Amazon reaches upper levels where
an outflow directed westward and eastward occurs, with the
upper-level eastward flow sinking over the Northeast Brazil and Eastern Atlantic (Fig. 8e), which is consistent with
the negative precipitation anomalies in NNEB from 2012 to
2014 (Fig. 3a).
In El Niño years (2015–2016) (Fig. 8d) upward motion
also occurs over 120° E–130° W and from 60° W to 20°
W, as observed in previous El Niño years (Fig. 8b). However, towards 120° W and 80° W the circulation pattern is
completely reverse, which resulted in stronger descending motion over this region during 2015–2016 (Fig. 8f).

13

On the other hand, over the Amazon the ascending branch
of the Walker circulation is strengthened throughout the
troposphere, while over the NEB and surrounding region
sinking motion are observed at the middle and upper-level,
and ascending motion are seen in the lower troposphere
(Fig. 8e). The ascending motion over the Amazon with the
anomalous outflow directed westward and eastward suggest
that the occurrence of below-normal NNEB rainy season
in 2015–2016 (Fig. 3a) is linked to the descending branch
of the Walker circulation. However, the weak intensity of
the descending branch of the Walker cell over NEB during
MAM indicates that others teleconnections pattern, such as
the MJO propagation and higher latitude forcing, may also
contribute to the descending movement over NEB (Rodrigues et al. 2019).
The analyses of the upper-tropospheric velocity potential
and divergent wind vector anomaly (Fig. 9) corroborate with
the anomalous ascending and descending branches of the
Hadley and Walker cells (Figs. 7e, f and 8e, f) previously
described over NEB. In neutral years (Fig. 9a), there was a
pattern of converging winds near NEB at 200 hPa, consistent with the outgoing longwave radiation anomalies shown
in Fig. 6a, and supporting the earlier described Hadley

Climatic aspects and vertical structure circulation associated with the severe drought in…

2337

Fig. 8  Zonal-vertical circulation (Walker) averaged in the latitudinal
band between 10° S and 0° for MAM during the a 20 neutral composite years, b five El Niño composite years, from c 2012 to 2014, and d
2015 to 2016. The anomalous Walker circulation during 2012–2016

regarding the neutral and El Niño years are shown in (e) and (f). The
shaded plot represents vertical motion (omega) in hPa s−1, which is
superimposed by wind vectors in m/s

circulation with anomalously downward motion between 5°
S and 5° N (Fig. 7e). Furthermore, the sinking motion over
the NEB is also associated with the anomalous flow induced
by convection over the Western Amazon (Fig. 8c, e), and the
slightly displaced of the South Atlantic Convergence Zone
climatological position (Fig. 6a) (Zilli et al. 2018). These
anomalous circulations at high levels enhance the unfavorable scenario for cloud formation over NEB, which are associated with observed negative precipitation anomalies in the
region from 2012 to 2014 (Fig. 3a).
In El Niño years (2015–2016) (Fig. 9b), convergence pattern at upper-level is observed around 25° W, i.e., in a western position than in the neutral years, indicating that in these
surrounding regions low-level (1000–700 hPa) subsidence
movements are present. Part of this subsidence is coherent
with the changes in the Walker circulation associated with
the anomalous strong convection over the Amazon, as also
shown in Fig. 8f. Moreover, the sinking motion over NEB

seems also related to the strong anomalous flow induced
by convection in the ITCZ over Africa and Gulf of Mexico
(Fig. 9b). These patterns are in line with the positive OLR
anomalies in Fig. 6b, and confirms that the negative precipitation anomalies over NNEB during 2015–2016 (Fig. 3a)
could not be exclusively related to El Niño conditions, as
described by Marengo et al. (2016), Cunha et al. (2019),
among others.

4 Conclusion
Understanding the mechanisms associated with drought
events in Northeast Brazil can contribute to subsidizes analyzes of droughts events, as well as in actions to mitigate
their impacts. In this sense, the aim of this study was to
investigate the vertical structure circulation associate with
the intense and persistent drought event occurred in the

13

2338

F. J. de Medeiros et al.

Fig. 9  Velocity potential
anomalies in 106 m2 s (contours
and colors) and divergent wind
anomalies in m s−1 (vectors)
at 200 hPa during the austral
autum from a 2012 to 2014 and
b 2015 to 2016. Anomalies are
computed with respect to the
composite of 20 neutral years
and five El Niño years that
occurred in MAM according
to the ONI from 1981 to 2010.
The dashed line indicates the
NEB region

Northeast Brazil during 2012–2016. To achieve this goal,
we examined historical precipitation records as well as associated global and regional aspects of some atmospheric and
oceanic parameters, along with possible teleconnection patterns associated with this drought.
The precipitation dataset indicated that Northern Northeast Brazil was the region most affected by the drought, with
rainfall deficits greater than 1000 mm from 2012 to 2016.
By evaluating the differences between the mean climatological period (1981–2010) and the drought event (2012–2016)
during the NNEB main rainy season (MAM), we found a
deficit of 37.8% during 2012–2016. This highlight the reason why many water reservoirs across the Northeast region
reached low levels of its volume capacity, given that the
recompositing of these reservoirs are restricted primarily
to those months of the short rainy season. The MAM average precipitation for each year during 2012 to 2016 showed
that 2012 was the driest rainy season, with negative precipitation anomalies of − 317.5 mm, followed by 2016 with
− 184.6 mm. The maximum and minimum temperature
analysis showed that during the severe drought the NNEB
population not only suffered from water scarcity, but also by
the increase in the frequency of hot spells and decrease frequency of cold nights (statistically significant at 1% level).
As the NNEB is one of the most socio-climatic vulnerable
area in the country (Torres et al. 2012; Lapola et al. 2019),
these results may reveal worse conditions with regard to
human thermal comfort, affecting mainly the most vulnerable age groups such as children and the elderly.

13

With respect to the ocean conditions, the ONI showed
that this drought was influenced by two distinct ocean
conditions: neutral (2012–2014) and El Niño years
(2015–2016). According to our analyzes involving the
regional scale and the teleconnection patterns, the negative
precipitation anomalies from 2012 to 2014 was initially
caused by the northward displacement of the ITCZ, as
indicated by the OLR anomalies. In the following years,
our results suggest that it is reasonable to assume that the
upward anomalous motion over the Western Amazon, as
well as the anomalous convection along the South Atlantic Convergence Zone, that was slightly displaced to the
south of its climatological position, induces subsidence
over NEB, being probably the main cause of the drought.
In 2015–2016, despite the occurrence of a strong El
Niño event, the two MAM quarters occurred close to the
beginning and ending of the event. Thus, although the
Walker circulation indicated anomalously sinking motion
over NEB, the weak intensity of the descending branch of
the Walker cell suggest that not only El Niño was related
to the dryness 2015–2016 period. The velocity potential
and divergent wind anomalies in the upper troposphere
(200 hPa) indicated strong anomalous flow over Gulf of
Mexico and Africa, leading to sinking motion in the NEB
and neighboring equatorial Atlantic. Moreover, transient
forcing mechanisms, such as MJO and/or higher latitude
forcing, may also be related to the precipitation deficit,
which will be the subject of future works.

Climatic aspects and vertical structure circulation associated with the severe drought in…

Overall, the results of this study provide the atmospheric
circulation pattern background associated with an intense
and persistent drought event occurred in Northeast Brazil
(2012–2016). This information can contribute to the scientific community and might particularly be useful for state,
regional and national meteorological centers, since the staff
can use our findings to improve the climate monitoring tasks
and the seasonal rainfall forecasting.
Acknowledgements This study was financed in part by Coordenação
de Aperfeiçoamento de Pessoa de Nível Superior—Brasil (CAPES)—
Finance Code 001. The authors thank the anonymous reviewers for
their constructive coments and suggestions.

References
Amorim ACB, Chaves RR, Santos e Silva CM (2014) Influence of the
tropical Atlantic ocean’s sea surface temperature in the eastern
Northeast Brazil precipitation. Atmos Clim Sci 4:874–883
ANA (2017) Conjuntura dos recursos hídricos no Brasil 2017: relatório
pleno/Agência Nacional de Águas. https:​ //www.snirh.​ gov.br/porta​
l/snirh​/centr​ais-de-conte​udos/conju​ntura​-dos-recur​sos-hidri​cos/
conj2​017_rel-1.pdf. Accessed 23 June 2018
ANA (2018) Conjuntura dos recursos hídricos no Brasil 2018: informe
anual/Agência Nacional de Águas. https​://arqui​vos.ana.gov.br/
porta​l/publi​cacao​/Conju​ntura​2018.pdf. Accessed 09 Jun 2020
Andreoli RV, Kayano MT (2004) Multi-scale variability of the sea surface temperature in the Tropical Atlantic. J Geophys Res 109:1–12
Aragão LEOC, Anderson LO, Fonseca MG, Rosa TM, Vedovato LB,
Wagner FH, Silva CVJ, Silva Junior CHL, Arai E, Aguiar AP,
Barlow J, Berenguer E, Deeter MN, Domingues LG, Gatti L,
Gloor M, Malhi Y, Marengo JA, Miller JB, Phillips OL, Saatchi
S (2018) 21st century drought-related fires counteract the decline
of Amazon deforestation carbonon emissions. Nat Commun. https​
://doi.org/10.1038/s4146​7-017-02771​-y
Azevedo SC, Cardim GP, Puga F, Singh RP, Silva EA (2018) Analysis of the 2012–2016 drought in the northeast Brazil and its
impacts on the Sobradinho water reservoir. Remote Sens Lett
9(5):439–447
Barbosa HA, Kumar TL, Paredes F, Elliott S, Ayuga JG (2019) Assesment of Caatinga response to drought using Meteosat-SEVIRI
normalized difference vegetation index (2008–2016). ISPRS
J Photogramm Remote Sens. https​: //doi.org/10.1016/j.isprs​
jprs.2018.12.014
Barichivich J, Gloor E, Peylin P, Brienen RJW, Schongart J, Espizona
JC, Pattnayak KC (2018) Recent intensification of Amazon flooding extremes driven by strengthened Walker circulation. Sci Adv
4:1–7
Baudoin MA, Vogel C, Nortje K, Naik M (2017) Living with drought
in South Africa: lessons learnt from the recent El Niño drought
period. Int J Disaster Risk Reduct 23:128–137
Betts AK, Dugan FJ (1973) Empirical formula for saturation pseudoadiabats and saturation equivalent potential temperature. J Appl
Meteor 12:731–732
Bezerra BG, Silva LL, Santos e Silva CM, Carvalho GG (2018)
Changes of precipitation extremes indices in São Francisco river
basin, Brazil from 1947 to 2012. Theor Appl Climatol. https:​ //doi.
org/10.1007/s0070​4-018-2396-6
Bolton D (1980) The computation of equivalent potential temperature.
Mon Weather Rev 108:1046–1053
Brasil (2017) Resolução No. 107/2017, de 27 de Julho de 2017.
Delimitação do semiárido. Recife, PE. https​://suden​e.gov.br/

2339

image​s/2017/arqui​vos/Resol​ucao-107-2017.pdf. Accesssed 11
Sept 2018
Brito SSB, Cunha APMA, Cunningham CC, Alvalá RC, Marengo
JA, Carvalho MA (2017) Frequency, duration and severity of
drought in the Semiarid Northeast Brazil region. Int J Climatol.
https​://doi.org/10.1002/joc.5225
Cabral Júnior JB, Santos e Silva CM, Almeida HA, Bezerra BG,
Spyrides MHC (2019) Detecting linear trend of reference evapotranspiration in irrigated farming areas in Brazil’s semiarid
region. Theor Appl Climatol. https​: //doi.org/10.1007/s0070​
4-019-02816​-w
Cai W, McPhaden MJ, Grimm AM, Rodrigues RR, Taschetto AS,
Garreaud R et al (2020) Climate impacts of the El Niño-Southern Oscillation on South America. Nat Rev Earth Environ
1:215–231. https​://doi.org/10.1038/s4301​7-020-0040-3
Chaves RR, Cavalcanti IFA (2001) Atmospheric circulation features
associated with rainfall variability over Southern Northeast Brazil. Mon Weather Rev 129:2614–2626
Coelho CAS, Gan MA, Conforte JC (2004) Estudo da variabilidade
da posição e da nebulosidade associada à ZCIT do atlântico,
durante a estação chuvosa de 1998 e 1999 no Nordeste do Brasil. Rev Bras Meteorol 19:23–34
Coelho CAS, Cardoso DHF, Firpo MAF (2015) Precipitation diagnostics of an exceptionally dry event in São Paulo, Brazil. Theor
Appl Climatol. https​://doi.org/10.1007/s0070​4-015-1540-9
Coelho CAS, Oliveira CP, Ambrizzi T, Reboita MS, Carpenedo CB,
Campos JLP, Tomaziello ACN, Pampuch LA, Custódio MS,
Dutra LMM, Rocha RP, Rehbein A (2016) The 2014 southeast
Brazil austral summer drought: regional scale mechanisms and
teleconnections. Clim Dyn 46:3737–3752
Condel (2017) Conselho Deliberativo da SUDENE. Resolução No.
115/2017, de 23 de Novembro de 2017. https​://www.suden​e.gov.
br/compo​nent/conte​nt/artic​le?id=834. Accesssed 08 June 2020
Cook BL, Mankin JS, Anchukaitis KJ (2018) Climate change and
drought: from past to future. Curr Clim Change Rep 4:164–179
Cunha APMA, Alvalá RC, Nobre CA, Carvalho MA (2015) Monitoring vegetative drought dynamics in the Brazilian semiarid
region. Agric and Forest Meteorol. https​://doi.org/10.1016/j.
agrfo​r met.2015.09010​
Cunha APMA, Tomasella J, Ribeiro-Neto GG, Brown M, Garcia SR,
Brito SB, Carvalho MA (2018) Changes in the spatial-temporal
patterns of droughts in the Brazilian Northeast. Atmos Sci Lett.
https​://doi.org/10.1002/as1.855
Cunha APMA, Zeri M, Leal KD, Costa L, Cuartas LA, Marengo
JA, Tomasella J, Vieira RM, Barbosa AU, Cunningham C, Garcia JVC, Broedel E, Alvalá R, Ribeiro-Neto G (2019) Extreme
drought events over Brazil from 2011 to 2019. Atmosphere.
https​://doi.org/10.3390/atmos​10110​642
Curtin S, Hastenrath S (1995) Forcing of anomalous sea surface temperature evolution in the tropical Atlantic during Pacific warm
events. J Geophys Res. https​://doi.org/10.1029/95JC0​1502
De Souza EB, Ambrizzi T (2002) ENSO impacts on the South
America rainfall during 1980s: Hadley and Walker circulation.
Atmósfera 15(2):105–120
De Souza EB, Kayano MT, Ambrizzi T (2005) Intraseasonal and
submonthly variability over the Eastern Amazon and Northeast
Brazil during the autumn rainy season. Theor Appl Climatol.
https​://doi.org/10.1007/s0070​4-004-0081-4
Dee DP, Uppala SM, Simmons AJ, Berrisford P, Poli P et al
(2011) The ERA-Interim reanalysis: configuration and perfomance of the data assimilation system. Q J R Meteorol Soc
137(656):553–597
Drumond A, Nieto R, Trigo R, Ambrizzi T, Souza E, Gimeno L (2010)
A lagrangian identification of the main sources of moisture affecting Northeastern Brazil during its pre-rainy and rainy seasons.
PLoS ONE 5(6):e11205

13

2340
Espinoza JC, Ronchail J, Marengo JA, Segura H (2018) Contrasting
North-South changes in Amazon wet-day and dry-day frequency
and related atmospheric features (1981–2017). Clim Dyn. https​
://doi.org/10.1007/s0038​2-018-4462-2
Ferraz JS (1950) Iminencia duma grande seca no Nordeste. Rev Bras
Geogr 12:3–15
Foltz GR, Brandt P, Richter I, Rodríguez-Fonseca B, Hernandez F et al
(2019) The tropical Atlantic observing system. Front Mar Sci.
https​://doi.org/10.3389/fmars​.2019.00206​
Freitas ACV, Aímola L, Ambrizzi T, Oliveira CP (2017) Changes
in intensity of the regional Hadley cell in Indian Ocean and its
impacts on surrounding regions. Meteorol Atmos Phys 129:229–
246. https​://doi.org/10.1007/s0070​3-016-0477-6
Funk C, Harrison L, Shukla S, Hoell A, Korecha D, Magadzire T,
Husak G, Galu G (2016) Assesssing the contributions of local
and east Pacific warming to the 2015 droughts in Ethiopia and
southern Africa. Bull Am Meteorol Soc. https​://doi.org/10.1775/
BAMS-D-16-0167.1
Gadelha AN, Coelho VHR, Xavier AC, Barbosa LR, Melo DCD, Xuan
Y, Huffman GJ, Petersen WA, Almeida CN (2019) Grid box-level
evaluation of IMERG over Brazil at various space and time scales.
Atmos Res 218:231–244
Gandu AW, Silva Dias PL (1998) Impact of tropical heat sources on the
South American tropospheric upper circulation and subsidence. J
Geophys Res 103:6001–6015
Getirana A (2016) Extreme water deficit in Brazil detected from space.
J Hydrol. https​://doi.org/10.1175/JHM-D-15-0096.1
Gloor M, Brienen RJW, Galbraith D, Feldpaush TR, Schöngart J,
Guyot JL, Espinoza JC, Lloyd J, Phillips OL (2013) Intensification of the Amazon hydrological cycle over the last two decades.
Geophys Res Lett. https​://doi.org/10.1002/grl.50277​
Griffin D, Anchukaitis KJ (2014) How unusual is the 2012–2014 California drought? Geophys Res Lett. https:​ //doi.org/10.1002/2014G​
L0624​33
Grimm AM (2003) The El Niño impact on the summer monsoon
in Brazil: regional processes versus remote influences. J Clim
16:263–280
Grimm AM (2004) How do La Niña events disturb the summer monsoon system in Brazil? Clim Dyn 22:123–138
Hastenrath S (2006) Circulation and teleconnection mechanisms
of Northeast Brazil droughts. Prog Oceanogr. https ​ : //doi.
org/10.1016/j.pocea​n.2005.07.004
Hastenrath S (2012) Exploring the climate problems of Brazil’s Nordeste: a review. Clim Change 112:243–251
Hastenrath S, Heller L (1977) Dynamics of climate hazards in Northeast Brazil. Q J R Meteorol Soc 103:77–92
Jiménez-Muñoz JC, Mattar C, Barichivich J, Santamaría-Artigas A,
Takahashi K, Malhi Y, Sobrinho JA, Schrier GV (2016) Recordbreaking warming and extreme drought in the Amazon rainforest
during the course of El Niño 2015–2016. Sci Rep. https​://doi.
org/10.1038/srep3​3130
Jiménez-Muñoz JC, Marengo JA, Alves LM, Sulca JC, Takahashi K,
Ferret S, Collins M (2019) The role of ENSO flavors and TNA
on recent droughts over Amazon forests and the Northeast Brazil
region. Int J Climatol. https​://doi.org/10.1002/joc.6453
Kane RP (1997) Prediction of droughts in North-East Brazil: role of
ENSO and use of periodicities. Int J Climatol 17:655–665
Kayano MT, Andreoli RV, Souza RAF (2013) Relations between
ENSO and the South Atlantic SST modes and their effects on the
South America Rainfall. Int J Climatol. https​://doi.org/10.1002/
joc.3569
Kayano MT, Andreoli RV, Garcia SR, Souza RAF (2018) How the two
nodes of the tropical Atlantic sea surface temperature dipole relate
the climate of the surrounding regions during austral autumn. Int
J Climatol 38:3927–3941
Kendall MG (1975) Rank correlation methods. Grifin, London

13

F. J. de Medeiros et al.
Lapola DM, Braga DR, Di Giulio GM, Torres RR, Vasconcellos MP
(2019) Heat stress vulnerability and risk at the (super) local
scale Brazilian capitals. Clim Change. https​://doi.org/10.1007/
s1058​4-019-02459​-w
Liebmann B, Smith CA (1996) Description of a complete (interpolated) outgoing longwave radiation dataset. Bull Am Meteorol
Soc 77:1275–1277
Lin Q, Wu Z, Singh VP, Sadeghi SHR, He H, Lu G (2017) Correlation between hydrological drought, climatic factors, reservoir operation, and vegetation cover in the Xijang Basin, South
China. J Hydrol. https​://doi.org/10.1016/j.jhydr​ol.2017.04.020
Lucena DB, Servain J, Gomes Filho MF (2011) Rainfall response
in Northeast Brazil from ocean climate variability during the
second half of the twentieth century. J Clim 24:6174–6184
Manatasa D, Mukwada G (2017) A connection from stratospheric
ozone to El Niño-Southern Oscillation. Sci Rep. https​: //doi.
org/10.1038/s4159​8-017-05111​-8
Mann HB (1945) Nonparametric tests against trend. Econometrica
13:245–259
Marengo JA, Torres RR, Alves LM (2016) Drought in Northeast
Brazil—past, present, and future. Theor Appl Climatol. https​://
doi.org/10.1007/s0070​4-016-1840-8
Marengo JA, Alves LM, Alvalá RCS, Cunha AP, Brito S, Moraes
OLL (2017) Climatic characteristics of the 2010–2016 drought
in the semiarid Northeast Brazil region. Anais da Academia
Brasileira de Ciências. https​: //doi.org/10.1590/0001-37652​
01702​06
Marengo JA, Cunha APC, Soares WR, Torres RR, Alves LM, Brito
SSB, Cuartas LA, Leal K, Neto GR, Alvalá RCS, Magalhaes AR
(2019) Increase risk of drought in the semiarid lands of Northeast
Brazil due to regional warming above 4 °C. ClimChange Risks
Braz. https​://doi.org/10.1007/978-3-319-92881​-4_7
Martins ESPR, Coelho CAS, Haarsma R, Otto FEL, King A, Oldenborgh GJV, Kew S, Philip S, Vasconcelos FC Jr, Cullen H (2017)
A multimethod attribution analysis of the prolonged Northeast
Brazil hydrometeorological drought (2012–2016). Bull Am Meteorol Soc 99:65–69
Medeiros FJ, Lima KC, Caetano DA, Silva FJO (2018) Impacto da
variabilidade interanual da precipitação nos reservatórios do semiárido do Nordeste do Brasil. Anuário do Instituto de Geociências
41:731–741
Medeiros FJ, Oliveira CP, Gomes RS, Silva ML, Cabral Júnior JB
(2020a) Hydrometeorological conditions in the semiarid and east
coast regions of Northeast Brazil in the 2012–2017 period. Ann
Braz Acad Sci
Medeiros FJ, Oliveira CP, Santos e Silva CM, Araújo JM (2020b)
Numerical simulation of the circulation and tropical teleconnection mechanisms of a severe drought event (2012–2016) in Northeastern Brazil. Clim Dyn 54:4043–4057. https​://doi.org/10.1007/
s0038​2-020-05213​-6
Mo KC, Lettenmaier DP (2018) Drought variability and trends over the
central United States in the instrumental record. J Hydrol. https​://
doi.org/10.1175/JHM-D-17-0225.1
Molion LCB, Bernardo SO (2002) Uma revisão da dinâmica das chuvas
no nordeste brasileiro. Rev Bras Meteorol 17:1–10
Mossman RC (1919) The rainfall of Fortaleza, Ceara, Brazil. Q J R
Meteorol Soc 45:69–79
Moura AD, Shukla J (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
Nobre P, Shukla J (1996) Variations of sea surface temperature, wind
stress, and rainfall over the tropical Atlantic and South America.
J Clim 9:2464–2479
Oliveira PT, Santos e Silva CM, Lima KC (2014) Linear trend of occurrence and intensity of heavy rainfall events on Northeast Brazil.
Atmos Sci Lett. https​://doi.org/10.1002/as12.484

Climatic aspects and vertical structure circulation associated with the severe drought in…
Oliveira CP, Ambrizzi T, Aimola L (2016) Influence of intraseasonal
variability on precipitation in northern South America during the
winter season. Int J Climatol. https​://doi.org/10.1002/joc.4845
Oliveira PT, Santos e Silva CM, Lima KC (2017) Climatology and
trend analysis of extreme precipitation in subregions of Northeast Brazil. Theor Appl Climatol. https​://doi.org/10.1007/s0070​
4-016-1865-z
Oliveira CP, Aímola L, Ambrizzi T, Freitas ACV (2018) The influence
of the regional Hadley and Walker circulations on precipitation
patterns over Africa in El Niño, La Niña and Neutral years. Pure
Appl Geophys. https​://doi.org/10.1007/s0002​4-018-1782-4
Ovando A, Tomasella J, Rodriguez DA, Martinez JM, Siqueira-Junior
JL, Pinto GLN, Passy P, Vauchel P, Noriega L, von Randow C
(2016) Extreme flood events in the Bolivian Amazon wetlands. J
Hydrol Reg Stud 5:293–308
Pereira MPS, Justino F, Malhado ACM, Barbosa H, Marengo J
(2014) The influence of oceanic basins on drought and ecosystem dynamics in Northeast Brazil. Environ Res Lett. https​://doi.
org/10.1088/1748-9326/9/12/12401​3
Rao VB, Hada K, Herdies DL (1995) On the severe drought of 1993 in
North-east Brazil. Int J Climatol 15:697–704
Rodrigues RR, McPhaden MJ (2014) Why did the 2011–2012 La Niña
cause a severe drought in the Brazilian Northeast? Geophys Res
Lett. https​://doi.org/10.1002/2013G​L0587​03
Rodrigues RR, Haarsma RJ, Campos EJD, Ambrizzi T (2011) The
impacts of inter-El Niño variability on the tropical Atlantic and
Northeast Brazil climate. J Clim 24:3402–3422
Rodrigues RR, Taschetto AS, Gupta AS, Foltz GR (2019) Common
cause for severe droughts in South America and marine heatwaves
in the South Atlantic. Nat Geosci. https​://doi.org/10.1038/s4156​
1-019-0393-8
Ropelewski CF, Halpert MS (1987) Global and regional scale precipitation patterns associated with the El-Nino Southern Oscillation.
Mon Weather Rev 115:1606–1626
Servain J (1991) Simple climatic indices for the tropical Atlantic Ocean and some applications. J Geophys Res. https​://doi.
org/10.1029/91JC0​1046
Silva PE, Santose e Silva CM, Spyrides MHC, Andrade LMB (2018)
Precipitation and air temperature extremes in the Amazon and
Northeast Brazil. Int J Climatol. https​://doi.org/10.1002/joc.5829
Smith TM, Reynolds RW, Peterson TC, Lawrimore J (2008) Improvements to NOAA’s historical merged land-ocean surface temperature analysis (1880–2006). J Clim 21:2283–2296

2341

Stojanovic M, Drumond A, Nieto R, Gimeno L (2018) Variations in
Moisture supply from the Mediterranean Sea during meteorological drought episodes over central Europe. Atmosphere. https:​ //doi.
org/10.3390/atmos​90702​78
Tedeschi RG, Grimm AM, Cavalcanti IFA (2015) Influence of central and east ENSO on extreme events of precipitation in South
America during austral spring and summer. Int J Climatol. https​
://doi.org/10.1002/joc.4106
Torres RR, Lapola DM, Marengo JA, Lombardo MA (2012) Socioclimatic hotspots in Brazil. Clim Change. https:​ //doi.org/10.1007/
s1058​4-012-0461-1
Utida G, Cruz FW, Etourneau J, Bouloubassi I, Schefuß E, Vuille M,
Novello VF, Prado LF, Sifeddine A, Klein V, Zular A, Viana JCC,
Turcq B (2019) Tropical South Atlantic influence on Northeastern
Brazil precipitation and ITCZ displacement during the past 2300
years. Sci Rep. https​://doi.org/10.1038/s4198​-018-38003​-6
Valadão CEA, Carvalho LMV, Lucio PS, Chaves RR (2017) Impacts of
the Madden-Julian Oscillation on intraseasonal precipitation over
Northeast Brazil. Int J Climatol 37:1859–1884
Waliser DE, Gautier C (1993) A satellite-derived climatology of the
ITCZ. J Clim 9:2162–2174
Wilhite DA, Sivakumar MVK, Pulwarty R (2014) Managing
drought risk in a chaning climate: The role of national drought
policy. Weather Clim Extreme. https ​ : //doi.org/10.1016/j.
wace.2014.01.002
Xavier AC, King CW, Scanlon BR (2016) Daily gridded meteorological variables in Brazil (1980–2013). Int J Climatol. https​://doi.
org/10.1002/joc.4518
Ye X, Li X, Xu C, Zhang Q (2016) Similarity, difference and correlation of meteorological and hydrological drought indices in
a humid climate region - the Poyang lake catchment in China.
Hydrol Res 47(6):1211–1223
Zilli MT, Carvalho LMV, Lintner BR (2018) The poleward shift of
South Atlantic Convergence Zone in recent decades. Clim Dyn.
https​://doi.org/10.1007/s0038​2-018-4277-1
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

13