RESEARCH ARTICLE
Reflections on the impact and response to the
Peruvian 2017 Coastal El Niño event: Looking
to the past to prepare for the future
Marisol Yglesias-Gonza´lez1, Armando Valde´s-Vela´squezID1,2*, Stella M. Hartinger1,3,4,
Ken Takahashi5, Guillermo Salvatierra6, Rodrigo Velarde1,2, Alvaro Contreras7,
Hugo Santa Marı´a8, Marina Romanello9, Valerie Paz-Solda´n1,10, Juan Bazo1,11, Andre´s
G. Lescano1,6
1 Clima, Latin American Centre of Excellence for Climate Change and Health, Universidad Peruana
Cayetano Heredia, Lima, Peru, 2 EcoSalud, Laboratorio de Ecosalud y Ecologı´a Urbana, Facultad de
Ciencias e Ingenierı´a, Universidad Peruana Cayetano Heredia, Lima, Peru, 3 Swiss Tropical and Public
Health Institute, Basel, Switzerland, 4 University of Basel, Basel, Switzerland, 5 Geophysical Institute of
Peru, Lima, Peru, 6 Emerge, Emerging Diseases and Climate Change Research Unit, School of Public
Health and Administration, Universidad Peruana Cayetano Heredia, Lima, Peru, 7 Boston University, Boston,
Massachusetts, United States of America, 8 APOYO Consultorı´a, Lima, Peru, 9 University College London,
London, United Kingdom, 10 Health Office of Latin America, School of Public Health and Tropical Medicine,
Tulane University, Lima, Peru, 11 Red Cross Red Crescent Climate Centre, The Hague, Netherlands
* armando.valdes@upch.pe
Abstract
Climate-related phenomena in Peru have been slowly but continuously changing in recent
years beyond historical variability. These include sea surface temperature increases, irregular
precipitation patterns and reduction of glacier-covered areas. In addition, climate scenarios
show amplification in rainfall variability related to the warmer conditions associated with El
Niño events. Extreme weather can affect human health, increase shocks and stresses to the
health systems, and cause large economic losses. In this article, we study the characteristics
of El Niño events in Peru, its health and economic impacts and we discuss government pre-
paredness for this kind of event, identify gaps in response, and provide evidence to inform ade-
quate planning for future events and mitigating impacts on highly vulnerable regions and
populations. This is the first case study to review the impact of a Coastal El Niño event on
Peru’s economy, public health, and governance. The 2017 event was the third strongest El
Niño event according to literature, in terms of precipitation and river flooding and caused impor-
tant economic losses and health impacts. At a national level, these findings expose a need for
careful consideration of the potential limitations of policies linked to disaster prevention and
preparedness when dealing with El Niño events. El Niño-related policies should be based on
local-level risk analysis and efficient preparedness measures in the face of emergencies.
Introduction
El Nino-Southern Oscillation (ENSO) is a global-scale climate variability phenomenon. It is
caused by atmosphere-ocean interactions in which the tropical atmosphere responds to
PLOS ONE
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 1 / 20
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Yglesias-Gonza´lez M, Valde´s-Vela´squez A,
Hartinger SM, Takahashi K, Salvatierra G, Velarde
R, et al. (2023) Reflections on the impact and
response to the Peruvian 2017 Coastal El Niño
event: Looking to the past to prepare for the future.
PLoS ONE 18(9): e0290767. https://doi.org/
10.1371/journal.pone.0290767
Editor: João Miguel Dias, Universidade de Aveiro,
PORTUGAL
Received: June 20, 2022
Accepted: August 16, 2023
Published: September 26, 2023
Copyright: © 2023 Yglesias-Gonza´lez et al. This is
an open access article distributed under the terms
of the Creative Commons Attribution License,
which permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting information
files. Health, climate, and economic information are
available through the government web pages and
institutions (Law of Transparency and Data
Availability). This data has no restrictions and is
readily available (list of sources available in
Table 1). Interview information is not available.
Even though data could be de-identified (to a
certain point) because of the identities and
positions of the people interviewed (a very reduced
fluctuating ocean temperatures in the equatorial Pacific Ocean. ENSO events occur on a time-
scale of 3 to 7 years [1], which affects the global climate through atmospheric teleconnections
[2, 3]. The warm and cool ocean phases are called El Niño and La Niña, respectively, while the
Southern Oscillation is the large-scale atmospheric component of this interaction.
In 1983 and 1998, ocean warming from the coast of South America to the central equatorial
Pacific developed into extreme El Niño events and developed according to well-established
ENSO mechanisms, albeit with enhanced ocean-atmosphere feedbacks associated with west-
erly winds along the equator [4, 5]. These elements allowed for prediction several months in
advance [6]. In contrast, the onset of the 2017 coastal El Niño event was much more abrupt,
with warming of around 7–9˚C at the coast over 1–2 weeks [7, 8]. Coastal El Niño events are
associated with fast ocean-atmosphere feedback mechanisms restricted to the far-eastern
Pacific, involving northerly winds [7–11]. Global climate models could not adequately predict
the event, even with a one-month lead time [8, 10].
In general, ENSO events have wide-ranging impacts on global climate conditions and the
frequency of hydro-meteorological anomalies, such as floods and droughts [12]. Consequently,
ENSOs can be associated with considerable public health impacts, as they can cause increases
in the incidence of vector-borne diseases such as dengue, yellow fever and malaria. Moreover,
ENSO can reduce the continuity of care due to health infrastructure damage and can affect
food insecurity due to adverse effects on agricultural systems and fisheries. Changing rainfall
patterns and increasing temperatures on crop yields and water, impact food systems, transpor-
tation, market access and other deleterious effects on agriculture and artisanal fisheries [13–
16]. Possible damage to public infrastructure or population displacement can also be associ-
ated with ENSOs, leading to additional socioeconomic impacts [14, 17, 18].
In recent years, the scientific community has started focusing on the diversity among
ENSO events, particularly concerning the spatial distribution and strength of the warming or
cooling patterns in the equatorial Pacific [19], as these can determine the resultant climatic
anomalies around the world and also. For instance, El Niño warming in the eastern equatorial
Pacific enhances precipitation along the coast of Peru, while warming in the central equatorial
Pacific tends to reduce precipitation in the Andes and Amazon region [16, 20, 21].
Very intense and recurrent rainfall on Peru’s northern and central coast was observed in
2017. Above-average precipitation led to increases in seasonal river flows, with up to a 50-fold
increase compared to historical averages in the Rimac basin located in Lima, and with seasonal
river flows above 250% of the average in several other basins and resulted in landslides along
the whole length of the Peruvian coast. The event was categorised as a “Coastal” El Niño due to
the anomalies in the SST in Zones 1+2 (quadrant that covers the northern part of the Peruvian
sea) having its effects on the Pacific coast of South America [22].
The 2017 measured precipitation and subsequent increase in river flows had a significant
impact on the health and socioeconomic wellbeing of an estimated 1.9 million Peruvians
(approx. 6% of the population) [22–25]. It caused 169 deaths and 18 missing persons, displaced
194,000 people, and damaged more than three hundred thousand homes while leaving 65,000
homes uninhabitable [26]. Significant economic losses were attributed to this and other public
infrastructure damage, such as the collapse of water distribution systems and associated drink-
ing water shortages [27, 28].
We describe relevant climatic conditions during the Coastal El Niño 2017, differentiating
the development and impacts of that event from those of the well-known extreme global-scale
El Niño events of 1982–1983 and 1997–1998, which featured the strongest warming in the far
eastern Pacific since 1900, far exceeding other events in terms of the warming strength in this
region [29]. On the other hand, Takahashi & Martı´nez (2017) [7], using different indicators of
the hydroclimate of coastal Peru, showed that these two events and the 1925 coastal El Niño
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 2 / 20
group of possible people), there would be
information within the transcriptions or notes that
could be used to identify the interviewees. Due to
the nature of the data and the political value of
some of the information, data will be made
available only upon written request and description
of the use of data to this administrator Zoila Vela
(zoila.vela@upch.pe).
Funding: MYG and MR received funds from an
unrestricted grant from the Wellcome Trust
(209734/Z/17/Z). https://wellcome.org/ The
funders had no role in study design, data collection
and analysis, decision to publish, or preparation of
the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
were the strongest El Niño events between 1910 and 2000, while Peng et al. (2019) [10] showed
that the 2017 coastal El Niño event was the strongest in terms of the far eastern Pacific SST and
precipitation after 1983 and 1998. Note that even though the 2015–16 event is generally con-
sidered to be a strong El Niño, it did not enhance coastal precipitations [6] and can be classi-
fied as moderate in terms of its eastern Pacific signature [30]. Thus, in terms of their
hydroclimatic manifestations, we focus on 1925, 1983, 1998 and 2017 as the strongest El Niño
events along the coast of Peru since the early XX century.
Using publicly available climate, health, and economic data, as well as historical literature
on global ENSO events, we also assessed the national and sub-national economic impacts of
Peru’s 2017 Coastal El Niño, described, and compared the epidemiological data and recorded
rates of mandatory-reporting infectious diseases, and analysed the associated government pre-
paredness and response during the event. Lessons learned from the 2017 Coastal El Niño can
provide a good basis for better preparedness and responses for future events.
The analysis and findings of this study are relevant for broader disaster and risk manage-
ment in Peru. Considering that beyond El Niño events, scenarios of heavy rains and floods
linked to a changing climate are also expected, this case study aims to highlight the urgency of
adopting preparedness and response measures to prevent future impacts on Peru’s public
health and the economy as those experienced during and after the 2017 Coastal event.
Materials and methods
Study area
The 2017 Coastal El Niño strongly affected the coastal regions of Peru, from Tumbes to Ica.
(Fig 1). These regions have a large altitude range, from 0 to 3,979 metres above sea level. Peru’s
coastal departments (officially called “regions” in Peru) host 60% of the country’s population
and most of the country’s largest cities. Annual mean temperatures and precipitation also vary
across coastal departments, ranging from 3˚C to 32.9˚C and 0 to 1,308 mm [31–33] respec-
tively; with limited rainfall in non-El Niño years. Access to water and sanitation is almost uni-
versal in this region; however, households not having access to 24-hour water availability on
the coast range from 3% to 34.8%, affecting primarily the departments of Ica and Tumbes [34].
Study design
We assessed the impact of the 2017 Coastal El Niño on four components: climate, economy,
health, and policy. For the climatic characterisation, we searched historical climate, atmo-
spheric, and oceanographical records in Peru and literature reviewing the distinctions between
ENSO phases and circumstances to define the Coastal El Niño event. For the economic analy-
sis, we used a vector autoregressive (VAR) methodology to determine the economic impact of
the 2017 Coastal El Niño event and presented the economic valuation of direct losses by the
economic sector measured by the Government.
For the health component of this study, we compiled publicly available national data on
cases of mandatory-reporting infectious diseases previously associated with severe weather
events. We compared reported infection rates during the 2017 Coastal El Niño with those
observed during equivalent periods in the previous five years. For the policy dimension of this
case study, to identify potential bottlenecks, limitations and to understand and assess the
response of the Peruvian government to the emergency, we conducted in-depth interviews
with individuals who led response efforts to the event and reviewed national legislation and
policies for disaster emergency response and preparedness at the national, sub-national
(regions) and local geopolitical levels (provinces).
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 3 / 20
Fig 1. Weather warning for intense rainfall from March 19th to 25th, 2017 (self-elaboration 2021, based on [35] using
an open-source natural earth shapefile base).
https://doi.org/10.1371/journal.pone.0290767.g001
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 4 / 20
Data collection
Climatic characterisation. We used the following observational data: (i) monthly Sea
Surface Temperature (SST) data for Puerto Chicama, located in La Libertad department on the
northern coast of Peru (7.7˚S, 79.4˚W) for the 1925–2017 period, from the University of
Washington’s Joint Institute for the Study of the Atmosphere and Ocean (JISAO) and the
Peruvian Ocean Institute (Instituto del Mar del Peru´; IMARPE); and (ii) the monthly Piura
river discharge measured at Sanchez Cerro bridge in the city of Piura 1925–2020 from the
National Meteorological and Hydrological Service (SENAMHI), JISAO, the National Water
Authority (Autoridad Nacional del Agua, ANA), the Chira-Piura Special Project (Proyecto
Especial Chira-Piura, PECHP) and the study of Takahashi and Martı´nez [7]. The monthly
Niño 1+2 SST time series is from NOAA OISST v2. The global gridded monthly SST and pre-
cipitation data are from NOAA ERSST v5 [36] and GPCP v2.3 [37], respectively.
Economy. To determine the economic impact of the 2017 Coastal El Niño event and the
economic valuation of direct losses by the economic sector, we used economic data from three
national agencies: (i) regional Gross Domestic Product (GDP) data for 1998 from the Institute
of National Statistics (INEI), (ii) macroeconomic data related to aggregated GPD (private
Investment, private consumption, mining GDP), terms of trade, 1982–1983 total GDP and
agriculture and livestock production GDP from the Peruvian Central Bank (BCR) and (iii)
ENSO 2017 economic valuation of direct losses of public infrastructure by sector in US$ mil-
lion from the Special Authority for Reconstruction (RCC) [38–40].
Health. To describe the epidemiological context before, during and after the El Niño
2017, we used surveillance data on weekly cases of mandatory-reporting infectious diseases
(dengue, chikungunya, zika, leptospirosis, diarrheal disease, respiratory illness, and pneumo-
nia) from 2012 to 2017 from the Peruvian Ministry of Health’s Center for Epidemiology, Dis-
ease Control and Prevention (CDC-MINSA).
Policy. To collect the data, we conducted in-depth semi-structured interviews with key
stakeholders who were involved in the 2017 response. The interviewees were related to the
Ministry of Health (MINSA), the Presidency of the Council of Ministers (PCM) and the pri-
vate industry and provided context and further understanding of the institutional response to
the event. These in-depth interviews were tailored to each stakeholder focused on the issues
within their jurisdiction. The questions were framed under the main measures applied in the
emergency. All interviews were taped-recorded with prior participant consent and took
detailed notes. Interviews were transcribed verbatim in Spanish and compared with the notes
supporting the transcription process. We completed the assessment by searching and review-
ing technical legislation and policy documents regarding disaster response and preparedness
from 2010 to 2017.
Details and sources used for this manuscript can be seen in Table 1.
Data analysis
Climate characterisation of El Niño events. The climatological mean (1981–2010) of the
monthly time series of different climate variables was calculated separately. A monthly series
descriptive analysis of the SST of Puerto Chicama, the El Niño 1+2 index, and the Piura River
discharge for the El Niño years 1925, 1983, 1998, and 2017 were described with their respective
climatological values. We also contrast the El Niño 1+2 index with the reference threshold of
25ºC as an indicator of the onset of heavy rainfall and strong river discharge on the northern
coast of Peru. Finally, we compared the spatial patterns of SST and precipitation anomalies in
the Tropical Pacific for two El Niño events, 1998 and 2017.
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 5 / 20
Economy. The economic impact of the 2017 Coastal El Niño event was estimated using a
Vector Autoregression (VAR) methodology (See S1 Appendix for a detailed explanation). This
econometric analysis used macroeconomic and climate variables to isolate the effects of weather
anomalies and produce a referential forecast with information as of the 4Q2016 (before the
event). Specifically, variables SST and Piura River flow were considered for the economic VAR
analysis since they are usually associated with the occurrence of El Niño disasters in the past, and
therefore serve two purposes: detection of temperature changes and measure of flooding proba-
bility for a representative city (Piura). By comparing forecasted values to the actual performance
of the variables mentioned above during 2017, we quantified the “total” impact of the event on
GDP growth (both direct and indirect), reflected in macroeconomic data. The economic valua-
tion of direct losses in each economic sector due to El Niño was gathered from official statistics
and surveys conducted after the event by the Institute of National Statistics and the RCC.
Table 1. Summary of data sources.
Source Data collected
National Centre for Epidemiology Disease Prevention
and Control (CDC-Peru), of the Peruvian Ministry of
Health (MINSA) / https://www.dge.gob.pe/portalnuevo/
Number of weekly cases of mandatory-reporting
conditions (dengue, chikungunya, zika, leptospirosis,
diarrheal disease, respiratory illness, and pneumonia).
National Oceanic and Atmospheric Administration
(NOAA) / https://www.noaa.gov/
ENSO information
Peruvian Ocean Institute (Instituto del Mar del Peru´;
IMARPE) / https://www.gob.pe/imarpe
Monthly SST data for Puerto Chicama on the northern
coast of Peru (7.7˚S, 79.4˚W) for the 2003–2017 period.
Proyecto Especial Chira-Piura (PECHP) provided by the
Autoridad Nacional del Agua (ANA) / https://www.gob.
pe/pechp / https://www.gob.pe/ana
Monthly SST data for 2016–2017 was calculated from the
daily values obtained from the website of the PECHP
(http://www.chirapiura.gob.pe/). The climatological
average was made for each month separately for the
period 1981–2010.
Joint Institute for the Study of the Atmosphere and
Ocean (JISAO) / http://research.jisao.washington.edu/
data_sets/
Monthly SST for Puerto Chicama on the northern coast
of Peru (7.7˚S, 79.4˚W) for the 1925–2002 period Piura
river discharge measured in the city of Piura for 1926–
1998 (data for 1925 reconstructed by Takahashi)
Special Authority for Reconstruction (RCC) / https://
www.rcc.gob.pe/2020/
Economic valuation of direct losses of public
infrastructure by sector
Peruvian Central Bank (BCR) / https://www.bcrp.gob.
pe/
Macroeconomic data related to aggregated GPD (private
Investment, private consumption, mining GDP) Terms
of trade 1982–1983 total GDP Agriculture and livestock
production GDP from the Peruvian Central Bank
Institute of National Statistics (INEI) / https://www.inei.
gob.pe/ / https://www.gob.pe/inei
Regional Gross Domestic Product (GDP) data for 1998
Former Council of Ministries (PCM) Peruvian Ministry
of Health (MINSA), Private Industry
In depth interviews with key stakeholders from PCM (2
interviewees), MINSA (2 interviewees) and private
industry (2 interviewees)
National Disaster Risk Management System
(SINAGERD) (Law 29664) Regulatory Framework for
Disaster Risk in Peru
State policy "Disaster Risk Management"—national
agreement Law 29664—Law that creates the National
Disaster Risk Management System. Supreme Decree 048-
2011-PCM approving the regulation of Law 29664.
Supreme Decree 111–2012 PCM that provides for the
approval of the National Policy as a mandatory policy for
the entities of the National Government. Supreme
Decree 034–2014 PCM which provides the approval of
the National Plan for Disaster Risk Management—
PLANAGERD 2014–2021. Law Nº 29230—Works for
Taxes
Source: Self-elaboration 2021
https://doi.org/10.1371/journal.pone.0290767.t001
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 6 / 20
Health. First, we evaluated if, in the Coastal El Niño year (2017), there was a country-wide
increase in key weather-sensitive infectious diseases cases under mandatory reporting surveil-
lance (dengue, zika, leptospirosis and diarrheal disease, respiratory illness and pneumonia
among children<5 years old), compared to the average of previous, recent and most compara-
ble years (2012–2016). Then, we evaluated whether observed cases of dengue and diarrheal dis-
ease, respiratory illness, and pneumonia among children <5 years old increased during and
after the onset of Coastal El Niño, both for all of Peru and each of the four northern coastal
regions most affected (La Libertad, Lambayeque, Piura, and Tumbes). We compared weekly
2017 cases grouped in five-week periods versus the mean of the equivalent weeks in 2015–
2016 and used analysis of variance to determine whether cases increased above expected levels
of previous years and, if so, when upon the El Niño onset. Comparisons may slightly underes-
timate the impact of the 2017 Coastal El Niño, as there was a mild El Niño event between 2015
and 2016, potentially increasing incidence rates in that period.
Policy. We followed a semi-structured guide for the interviews to assess the appropriate-
ness of the 2017 Coastal El Niño response at different governmental levels and the effect of the
governance structure on this response. By using the international guidance documents on risk
management, “Disaster risk management systems analysis: A guide book” from the Food and
Agriculture Organization, and the “Information management and communication in emer-
gencies and disasters: Manual for disaster response teams” from the Pan American Health
Organization, we assessed and adapted three key capabilities: (i) information management, (ii)
rapid adaptation in response to changing conditions, and (iii) coordination between govern-
mental agencies [41, 42]. In addition, we wanted to understand if the existing emergency
response protocols, fund allocations and preparedness were appropriate before the event
occurred. We read, edited, and coded the information for the analysis based on the three key
capabilities. We complemented the analysis with additional information provided by the inter-
viewees (quoted in the results).
Ethics
The ethical review board of Universidad Peruana Cayetano Heredia reviewed and approved
the study (REF: 297-14-19, SIDISI: 103898). All participants who agreed to participate in the
study provided verbal consent before being interviewed.
Results and discussion
Climate characterisation of El Niño events
SST data from Puerto Chicama (Fig 2a) showed that the coastal ocean warming during the
extreme El Niño events of 1982–1983 and 1997–1998 was strong but gradual, albeit with some-
what different timing. In the 1982–1983 event, monthly SST was 2.2˚C above normal in Octo-
ber and peaked in May 1983 at 9.6˚C above normal. In the 1997–1998 event, SST was 3.9˚C
above normal in October and peaked in January 1998 (8.7˚C anomaly). A similar trend was
seen in the Niño 1+2 SST Index (Fig 2b), which represented an average over a larger area, so
the anomalies have a smaller amplitude, peaking in June 1983 (4.6˚C) and December 1998
(4.1˚C).
It is important to note that, rather than SST anomalies, it is the absolute SST that matters
most for precipitation. The reference threshold of around 25˚C can be taken roughly as the
minimum SST required to trigger precipitation in northern Peru [43], although, for the Piura
River, it is more accurate to use the relationship with the SST gradient [7]. Fig 2b shows that
the period when the SST of Niño 1+2 index exceeds 25˚C coincides well with the period when
the Piura River flow starts increasing, which in the climatological mean corresponds to
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 7 / 20
February-April. For the extreme Niño of 1982–1983 and 1997–1998, the condition of
SST> 25˚C was met for December-July and November-June, respectively (Fig 2b). However,
the highest discharge of the Piura River was observed in January-June and January-May,
respectively, with the peaks of the monthly river flow also coinciding with the SST peaks of the
Niño 1+2 index.
In contrast, the onset of the 1925 and 2017 Coastal El Niño was abrupt and of shorter dura-
tion according to the Puerto Chicama time series SST (Fig 2a), with the peaks (both in absolute
SST and anomaly) in March 1925 (7.7˚C above normal) and February 2017 (5.2˚C). The SST
of the El Niño 1+2 index for 2017 showed an excess of 25˚C between January and April, start-
ing only one month earlier than the climatological mean. Consistent with this, the timing of
the monthly Piura River flow for both 1925 and 2017 is similar to climatology, with the peaks
both in March, with values 3 and 6 times the climatological mean, respectively (Fig 2c).
Despite similarities in the impacts on the northern coast of Peru in terms of high SST and
heavy rainfall (Fig 3a and 3b), the 1998 global ENSO events and the extreme coastal El Niño
events of 2017 had very different signatures in the tropical Pacific Ocean during the February-
March high rainfall season. Firstly, the 1998 event was not only more intense along the coast,
but its warm and wet anomalies extended along the equator to the central Pacific (Fig 3a). On
the other hand, during the 2017 event, the warmest SST anomalies were confined to the east-
ern end of the Pacific (Fig 3b), with precipitation anomalies of opposite signs north and south
of the equator, consistent with the strengthening of the southern branch of the Intertropical
Fig 2. Monthly mean SST (˚C) a) at Puerto Chicama, b) in the Niño 1+2 region (reference level of 25˚C indicated
with a dashed black line), and c) Piura River discharge (m3/s) for the 1981–2010 climatology (black dotted) and
during the El Niño events of 1925 (green dashed and crosses), 1983 (red short-long dashed and circles), 1997 (blue
dot-dashed and filled circles) and 2017 (purple solid and squares). (Self-elaboration 2021 based on NOAA, JISAO,
IMARPE, Takahashi and Martı´nez 2017, ANA, PECHP, SENAMHI).
https://doi.org/10.1371/journal.pone.0290767.g002
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 8 / 20
Convergence Zone (ITCZ) at the expense of the northern branch. As we will see in the next
sections, the intensity and abruptness of this event, despite its short duration, caused impor-
tant economic losses and public health impacts.
Economic impacts of coastal El Niño
Economic activity. El Niño events of high intensity (such as those registered in 1982–
1983 and 1997–1998) are among the climate phenomena events that have recently caused the
greatest negative impact on economic activity and living conditions in Peru [44]. In both the
1982–1983 and 1997–1998 events, agriculture was the economic activity that suffered the most
in terms of destroyed capital stock and goods produced, affecting mainly small-scale, low-
income farmers. Commerce and transportation were other economic sectors with consider-
able income losses due to the blockages in main highways caused by flooding and mudslides,
which isolated several communities for extended periods [23].
In 1983, Peru’s GDP fell by 10%, and estimates suggest that nearly half of the contraction in
economic activity was explained by weather conditions related to the 1982–1983 El Niño [45].
The Impacts of the 1998 El Niño were lower: the country’s GDP fell by 0.4%, and agriculture
and livestock production stagnated. However, the most affected region was Peru’s northern
Fig 3. February-April mean SST (shading; ˚C) and precipitation (contours, solid positive; mm/day) variation
across the tropical Pacific basin for two El Niño events: a) extreme ENSO (1998) and b) a Coastal El Niño (2017).
There were no warming signals across the central Pacific basin before the Coastal El Niño event (Data from NOAA
ERSST v5, GPCP v2.3).
https://doi.org/10.1371/journal.pone.0290767.g003
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 9 / 20
coast due to mudslides (or "huaicos", “huaycos”) that were observed in Tumbes, Piura, Lam-
bayeque La Libertad, and Lima, which affected the economic activity [44].
The northern coast registered a 3% contraction in its GDP level. One of the most affected
economic sectors, fishing, recorded a GDP fall of -35.6% in 1998 compared to 1997’s output
[46]. According to the Development Bank of Latin America [45], if the El Niño event had not
occurred, the 1998 Peruvian GDP would have grown 3% instead of dropping 0.4%. Moreover,
the global economic crisis worsened the overall economic performance of Peru during extreme
El Niño periods, such as the beginning of the Latin American debt crisis during the 1982–1983
El Niño and the Asian crisis in 1998.
Despite the severity of infrastructure damage, the estimated impact of the 2017 Coastal El
Niño on national economic activity–measured in 2017 GDP percentage points lost–was lower
than the recorded effect of previous severe El Niño events. This could be explained by the sig-
nificant diversification of the Peruvian economy in the past 20 years and important innova-
tions in sectors such as agriculture, driven by the increase in international trade and
investment [47]. The national production of agricultural products that require higher techno-
logical investment and are generally exported was not significantly affected during 2017. These
include mango (1.9% growth), avocado (2.5% growth) and asparagus (1.3% growth). In con-
trast, other products which rely on more traditional agriculture showed negative growth fig-
ures in the same year, such as rice (-1.6%), sugar cane (-4.4%) and lemon (-37.9%). This might
have been related to the financially stronger and more technologically developed sectors hav-
ing made greater investments in preparedness systems and assigned greater resources to par-
tially renew damaged areas and promptly resume commercial activities, counteracting climate
effects.
Furthermore, the 2017 Coastal El Niño was not an event of a national-level impact but
mainly a regional event with an intense, focal impact on the northern coast of Peru. This could
also explain why its effects were lower than expected. This region only accounts for 20% of
Peru’s population, 16% of the GDP, and 18% of household spending. Regional total GDP fell
in the main regions of the northern coast during 1Q2017: Piura (-3.6%), Ancash (-2.6%), La
Libertad (-1.1%) and Lambayeque (-0.2%). Particularly, the agriculture and livestock GDP
component remained affected in some of these regions throughout 2017: Piura fell by -20.2%
and Ancash by -0.2%, while La Libertad and Lambayeque presented small growth figures 0,6%
and 0.7%, respectively.
Table 2 shows the expected growth of private investment and private consumption made by
the VAR model and the modelled influence that each variable shock had on the difference
between the baseline and actual scenarios. For both private investment and private consump-
tion, row (A) shows the expected baseline or variable growth (in percentage change) from
4Q2016. Rows (a), (b), (c) and (d) show the effect attributed to each variable shock on the base-
line prediction (in percentage points). By adding up these rows, it is possible to approximate
the actual observed growth for each variable during the period. The last column shows the
average impact that each variable had during 2017.
Our econometric vector autoregression (VAR) analysis of macroeconomic and climate var-
iables isolated the effects of weather anomalies and used data before the Coastal El Niño
(fourth quarter of 2016) to produce a referential 2017 economic activity forecast (baseline).
The differences between the baseline forecast and the actual observed data can be decomposed
by individual shocks related to the outcome of each variable after the prediction. For example,
a private investment shock shows how much of the aggregate forecast error can be explained
by the performance of this variable, which in our results is negative, probably due to a loss of
business confidence in this period. Similarly, a negative consumption shock can be understood
as a loss in consumer confidence, while a positive term of trade shock can be related to higher-
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 10 / 20
than-expected main export prices. Finally, the sea temperature anomalies and the Piura River
water flow negative shocks represent how much of the prediction error is explained by the
unexpected climatic conditions of the Coastal El Niño.
Our estimates showed that the Coastal El Niño had a negative effect of -4.5 and -1.1 per-
centage points on private investment growth and private consumption growth, respectively
(obtained by addition of rows (a) and (b)). Considering that in 2016 these economic aggregates
represented 18% and 64% of GDP, respectively, we can give an approximate aggregate GDP
impact by multiplying each negative effect by the individual weight of these indicators on
GDP. These results translate into an estimated aggregated negative impact of -1.5 percentage
points on 2017 GDP growth. So, GDP growth in 2017 would have reached 4.0% if the Coastal
El Niño had not occurred, instead of the officially recorded 2.5% growth.
Public infrastructure. Despite the well-known impacts of El Niño-related floods on pub-
lic infrastructure such as roads, bridges, schools, and healthcare facilities, investments to pre-
vent damage in the most vulnerable areas have remained insufficient. It was no surprise that
the intense rainfall and mudslides of the 2017 Coastal El Niño brought about such expected
damage to main roads and agricultural areas and floods in urban and rural areas of the north-
ern coast. In total, 1,159 health facilities were affected, of which 32 collapsed, and 39 were
declared inhabitable in a context where 1,644,879 people were affected and 505 injured [23].
In addition, 3,703 schools were damaged, 141 collapsed, and 315 were categorised as inhab-
itable [23]. Transportation infrastructure was severely affected: 493 bridges collapsed, an addi-
tional 943 were damaged, 6,428 km of highways and rural roads were destroyed, and an
additional 16,1905 km were affected [23]. Infrastructure damage was greater on the northern
coast and lesser along the central coast and the highlands. Major flooding took place in three
of the largest cities in Peru: Lima, Trujillo, and Piura, severely affecting urban infrastructure.
Water and sanitation services collapsed due to ruptures generated by mudflows.
Table 2. Historical decomposition of economic growth shocks for 2017.
Private Investment
Shocks 4Q2016 1Q2017 2Q2017 3Q2017 4Q2017 Average 2017
(A) Baseline scenario 0.7 4.1 5.3 6.5 6.7 5,7
(a) Sea temperature anomaly -0.1 -0.6 -1.5 -1.2 -0.1 -0.9
(b) Piura River water flow 0.5 -3.5 -4.5 -3.5 -2.9 -3.6
(c) Terms of trade shock 0.1 0.8 1.3 1.4 2.3 1.5
(d) Private investment shock -6.1 -5.8 -2.3 2.8 -2.6 -2.0
(e) Private consumption shock 0.1 -0.3 -0.9 -0.5 -0.2 -0.5
Actual Private Investment (A) + (a) + (b) +(c) +(d) +(e) -4.8 -5.3 -2.6 5.5 3.2 0.2
Private Consumption
Shocks 4Q2016 1Q2017 2Q2017 3Q2017 4Q2017 Average 2017
(A) Baseline scenario 5.0 4.5 3.8 4.0 3.7 4.0
(a) Sea temperature anomaly 0.0 0.7 -0.6 -1.0 -0.9 -0.5
(b) Piura River water flow 0.0 -0.2 -0.8 -0.8 -0.7 -0.6
(c) Terms of trade shock 0.1 0.2 0.1 0.3 0.6 0.3
(d) Private investment shock -1.0 -1.8 0.6 -0.1 0.5 -0.2
(e) Private consumption shock -1.1 -1.2 -0.7 0.4 -0.6 -0.5
Actual Private Consumption (A)–(a)–(b)–(c)–(d)–(e) 3.0 2.2 2.5 2.7 2.6 2.5
All figures are expressed in terms of real per cent growth.
Source: Self-elaboration 2021
https://doi.org/10.1371/journal.pone.0290767.t002
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 11 / 20
Almost 49,000 housing units were affected or damaged, and streets and sidewalks were
severely affected. Vulnerable areas, due to their proximity to riverbanks, were worse off.
Repairing all damaged public infrastructure in urban and rural areas and installing prevention
infrastructure would cost approximately US$ 7.7 billion, approximately 3,5% of Peru’s GDP
and 11% of the total GDP of the five most affected regions (Table 3).
Required investments to complete such massive reconstruction could take many years, con-
sidering Peru’s track record of deficient public investment. Between September 2017 and April
2019, only 27% of the total approved budget related to the reconstruction works was used [48].
Since the start of the reconstruction program, there have been major changes in the legal
reconstruction framework. A Government to Government (G2G) agreement with the United
Kingdom was recently implemented to accelerate the project’s pipeline and execute the
required prevention projects [49].
Health assessment. In 2017, the CDC-Peru documented over 68,000 dengue cases, repre-
senting Peru’s largest dengue outbreak in history and a two-fold increase from the maximum
annual cases observed between 2012 and 2016 (Table 4). The four most battered regions by El
Niño, all on the northern coast, accounted for 82% of the cases, despite having only 17% of
Peru’s population, and Piura, the epicentre, had 65% of all the dengue cases that year. Mean
dengue cases in Piura in 2017, before the El Niño emergency status was declared, were 12%
below the mean of the corresponding 2015 and 2016 weeks. In the following five weeks of the
event, the mean of cases increased marginally significantly to 60% above the 2015–2016 mean
(p = 0.070). Then, in the next five weeks, it rose to 404% above the 2015–2016 mean
(p<0.001), remaining increased for 15 additional weeks (For more details, see Table 1 in S2
Appendix).
Table 3. Estimated costs of reconstruction of public infrastructure damaged by the 2017 Coastal El Niño. Used an average exchange rate of S/3.26 per US$ to trans-
form it from S/ to US$ [40].
Type of infrastructure Description Total cost (US$ Million) % of total
Transportation Highways, roads, and bridges at the national and local level 2,994 38.8
Education Public schools’ renewal or reconstruction 819 10.6
Water and sanitation Renewal of water supply and sewage facilities 626 8.1
Urban roads Reconstruction of affected roads and sidewalks 455 5.9
Agriculture Reconstruction of irrigation channels 413 5.4
Health Renewal or reconstruction of hospitals and small-scale health facilities 413 5.4
Housing Renewal or relocation and reconstruction of affected houses 342 4.4
Prevention River defences, reservoirs, and sewage systems 1,650 21.4
Total 7,712 100.0
https://doi.org/10.1371/journal.pone.0290767.t003
Table 4. Number of annual, countrywide cases of climate-sensitive infectious diseases under mandatory surveillance [50].
Disease 2012 2013 2014 2015 2016 2017 2012–2016 mean ± standard
deviation
2017/2016-2012 excess
(%)
Dengue 28,505 13,092 17,234 35,817 25,160 68,290 23,962 ± 9,028 185.0%
Zika - - - - 1,572 6,099 1,572 288.0%
Leptospirosis 1,966 1,889 2,413 2,376 2,062 3,319 2,141 ± 240 55.0%
Diarrhoeal disease, <5 years old 525,811 511,228 487,601 513,975 532,553 499,567 514,234 ± 17,234 -2.9%
Respiratory infections, <5 years
old
2,801,593 2,912,228 2,644,474 2,653,659 2,793,147 2,595,350 2,761,0220 ± 112,529 -6.0%
Pneumonia, <5 years old 30,793 30,065 25,895 25,169 26,405 26,112 27,665 ± 2,574 -5.6%
https://doi.org/10.1371/journal.pone.0290767.t004
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 12 / 20
More than 5,000 weekly dengue cases were reported countrywide for five weeks between
April and May, with a peak of 7,072 cases in the second week of May (week 18). Weekly cases
increased substantially both countrywide and in Piura following the flooding in the city of
Piura on March 25th (week 12). While the timing of the peak in cases followed dengue’s typical
seasonal patterns, cases in 2017 exceeded by 4.9 standard deviations the 2012–2016 annual
mean (Table 4), a highly unlikely event given the annual dengue incidence in those years
(probability < 0.00005%). In addition, in 2017, there was a record of 89 deaths due to dengue,
61 of them in Piura. Also, Zika cases increased nearly four-fold in 2017 versus 2016, and most
of the increase took place in the coastal regions affected by the Coastal El Niño event.
Reported countrywide leptospirosis cases in 2017 were also 4.9 standard deviations above
the 2012–2016 annual mean (probability < 0.00005%), mainly due to a sizable increase
between January and June 2017 in Piura and Tumbes [51]. In contrast, the 2017 total cases
countrywide of other El Niño-sensitive diseases with shorter incubation periods, such as diar-
rheal disease, acute respiratory infections, and pneumonia, were lower than the 2012–2016
mean. However, the national total of episodes of diarrheal disease and pneumonia increased in
the first five weeks after the El Niño emergency was declared compared to the same weeks in
2015–2016 (p = 0.017 and p<0.001, respectively).
Mean weekly cases of diarrheal disease, acute respiratory infections and pneumonia
increased significantly after the El Niño onset in Piura (p = 0.006, p>0.001 and p = 0.004,
respectively), in addition to other specific increases in specific regions (See Table 2 in S2
Appendix). The significant increases in diarrheal and respiratory infections observed in the
northern regions account for 55% and 94% of the excess cases observed countrywide during
the 2017 Coastal El Niño. At the same time, although small, the increase in pneumonia cases
represented an early and off-season transmission wave.
Policy assessment
Peru’s disaster risk management system has a complex structure comprising several actors
(Box 1). The 2017 Coastal El Niño event made the gaps in the country’s system preparedness
and response structure visible. The public sector stakeholders mentioned that the most rele-
vant evidence gaps were: (i) the inadequate management and dissemination of information to
national and sub-national levels and general audiences; (ii) the slow adaptation and allocation
of funds in response to the fast-changing conditions of the event and (iii) the managerial
changes which lead to slower coordination and response. During the interviews, the need to
reduce population vulnerability to environmental hazards by improving territorial and envi-
ronmental management and preparedness, specifically in high-risk areas, was also mentioned.
Civil society´s engagement and adequate information management are crucial steps in
disaster risk reduction. Language barriers and inadequate or untimely information communi-
cation may lead to disinformation and confusion during an event. According to all the inter-
viewees, situations where the citizens did not understand the forecast and early warning
systems due to language limitations and inappropriate dissemination strategies hindered the
appropriate response.
"Despite the warnings from the early warning system, many districts failed to carry out pre-
paredness works and communicate effectively to their residents."
(PCM representative)
Moreover, local governments were not prepared for the event despite available annual pre-
paredness funds. Subnational governments (region and province-level) do not have the
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 13 / 20
capabilities to create an integrated disaster risk plan. Furthermore, the interviewee pointed out
many authorities lack the political will to apply for and manage funds, mainly due to the rigidity
in spending and a lack of human capacity. Usually, this context leads local governments to
prioritise initiatives with short-term immediate results and higher visibility and political impact
rather than integrated approaches with more considerable, albeit long-term, public gains.
“Very few mayors and regional presidents used the budget allocated in the previous year to
prepare for the phenomenon of El Niño”
(PCM representative)
In addition, by restricting the funding schemes to one year, the central government has sub-
stantially limited the possibility of developing integrated approaches needed for this kind of
climate event. A PCM representative stated that despite knowing the possibility of an El Niño
event, the magnitude of the event surpassed the prevention efforts put in place by the local
Box 1. Peru’s disaster risk management system.
Since 2011, the National Disaster Risk Management System (SINAGERD) conformed in
the Law 29664 [53] is responsible for identifying and reducing risks, preparing for and
responding to disasters-lays down in the national disaster response system, which is
comprised of several governmental agencies from all five governmental levels, National
(or Central Government), Regional (26 regions), provincial municipalities (196 prov-
inces), district municipalities (1,869 districts) and community governments. The entity
responsible for SINAGERD is the Presidency of the Cabinet of Ministries (PCM), which
can convene the National Council for Disaster Risk Management, a public organisation
responsible for political decision-making and strategic coordination. At the national
level, Peru has the National Civil Defence Institute (INDECI), responsible for prepared-
ness and response in the event of a disaster and the National Center for Estimating Pre-
vention and Reduction of Disaster Risk (CENEPRED), responsible for prevention and
reconstruction. INDECI and CENEPRED were ascribed to PCM, but this was changed
to the Ministry of Defence (MINDEF) in 2016 and 2017, respectively. Furthermore, dur-
ing the 2017 Coastal El Niño, INDECI assumed the functions of the PCM office associ-
ated with disaster risk management and was charged with the technical secretariat of the
National Council. Also, considering its relatively large budget, INDECI was effectively
the responsible entity in the national response to the 2017 El Niño event [54]. At the
regional and municipal levels, the corresponding governments are responsible for all the
processes in disaster risk management. In the case of El Niño, the Regional/Local Emer-
gency Operation Centres (COER/COEL) are responsible for monitoring, analysing, and
disseminating the information from the National Meteorological and Hydrological Ser-
vice (SENAMHI), the Regional Health Offices (DIRESA) and the Commission for the
National Study of El Niño (ENFEN), as well as from the National Emergency Operation
Centre (COEN), led by INDECI. Local governments constitute the first level of response
and surveillance in case of disasters through the National Plan for Disaster Risk Manage-
ment (PLANAGERD), the municipal civil defence groups and the Emergency Centre of
Operation (COE). Local governments execute annual budgets in the Reduction of Vul-
nerability and Attention to Disaster Emergencies Budget Programme PP068 (PRE-
VAED) framework to manage these actions.
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 14 / 20
governments. Once the event was on course, the central government declared a state of emer-
gency to access the emergency funds. Thirteen of the 24 departments of Peru (Piura, Tumbes,
Ancash, Junin, Huancavelica, Loreto, among others), corresponding to 109 provinces and 879
districts, were declared to be in a state of emergency.
“States of emergency allow local governments to make use of different budget items and
overcome control barriers established by the comptroller to speed up the spending required
by the emergency.”
(PCM representative)
A PCM representative mentioned that changes in the government structure in December
2016 led to slower bilateral coordination and created bottlenecks in the initial emergency
response. This led to a centralised response, where the central government coordinated
directly with the National Emergency Operation Centre (COEN), while the ministers of the
different portfolios were sent to the regions to coordinate directly with the Regional Emer-
gency Operation Centres (COER). Efficient and effective coordination and communication
among all sectors were slow, but sending national representatives allowed faster decision-
making and coordination between the municipalities, armed forces, police, firefighters,
regional health, agriculture, transportation and communications directorates and private
companies.
“There were cases of critical patients, which require immediate transfer, who were mobi-
lised thanks to coordination with the armed forces”
(MINSA representative)
The disaster risk response drove local stakeholders (private companies, NGOs, and the gen-
eral population) to try and bridge some of the governmental gaps. They donated “goods” during
the emergency and were involved in the early reconstruction effort through the mechanism of
“works for taxes”, led by the “Pro Inversion” programme of the central government [52].
“An alliance was made between other companies and a conversation with the regional gov-
ernment was made and they presented a list of priority works, and after a brief discussion,
we decided to collaborate in the reconstruction of the bridge”
(Private industry)
A clear example of the existing gaps could be seen in managing supplies (water and food)
and shelter for the displaced people. The response to re-building the lost infrastructure was
slow and inexistent in some areas. Health services could not cope with the increased flow of
patients from other flooded areas. Health staff initially turned away patients because of insuffi-
cient personnel to treat them or lack of identification cards and public insurance (a require-
ment for attendance in health establishments). Many health workers lost their homes and were
themselves displaced.
“Many health establishments denied care to people because they did not have an ID. In the
same way, there were attention problems with social security because in the absence of the
identity document they could not be registered”
(MINSA representative).
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 15 / 20
In addition, the existing health system did not allow for access to health care treatment in
camps for displaced populations. In these camps, people had to deal with additional difficulties
concerning violence, mental health, reproductive health, and access to care for underlying
conditions.
“Many patients with cancer, tuberculosis or HIV lost their medicines and had no access to
new ones.”
(MINSA representative)
The last topic that surfaced during the interviews was reducing risks in illegal housing,
pointing toward the need for local-level risk evaluations, particularly in vulnerable popula-
tions. In recent years, urban expansion has continued to grow disproportionately and often
informally, with the most disadvantaged populations occupying the land in dangerous and
marginal areas of the cities that have an increased risk of landslides and flooding.
“There are many people who are exposed to risks generated by the environment. On the
central coast many people occupy unsuitable spaces, but with the approval of the
municipality. . . (. . .) The municipalities do not always carry out prevention work in high-
risk areas, many wait for an emergency to occur before acting.”
(MINSA representative)
Conclusions
The 2017 Coastal El Niño event was intense and abrupt, with a rapid onset. It did not corre-
spond to a warm ENSO phase but coincided with neutral to cool ENSO conditions and was
confined to the far-eastern Pacific. Because of this intensity and abruptness, climate models
used for projections could present biases, particularly in the far-eastern Pacific, which should
be considered with care. Despite the short duration of this event, restricted to February-April,
a ten-fold increase in heavy rainfall and river flooding did incur important economic losses
and health impacts, similar to those of the 1983 and 1998 severe warm ENSO (El Niño) events.
The econometric analysis based on macroeconomic variables and sea temperature anoma-
lies showed that the Coastal El Niño event impacted more than just infrastructure, harming
private consumption and private investment, with an aggregate negative impact on GDP
growth of -1.5 percentage points in 2017. Higher event frequency than the normal 3–7 year
cycle would have a dire effect on Peruvian economics and the country’s ability to mitigate the
potential impacts, prepare accordingly and bounce back.
On the other hand, public health impacts with increases in infectious and vector-borne dis-
eases in disaster settings are not unexpected during El Niño events. The health assessment
shows that CDC-Peru recorded the largest dengue outbreak ever, with 76,093 confirmed cases,
68,630 (90% of reported cases) in the 2017 Coastal El Niño-affected areas. The analysis also dis-
plays an increase in reported leptospirosis cases in Piura, one of the regions most affected by
extreme weather. Other immediate impacts on health outcomes, particularly on infectious dis-
eases, may have been difficult to assess with surveillance data due to the interruption of health-
care services and disease reporting during the most severe weeks of the event.
The 2017 Coastal El Niño event also highlighted the challenges in Peru’s disaster response
system. It showcased the gaps and potential failures in the health system’s structure and func-
tion and the information management and dissemination system. Furthermore, it portrayed
how the state responded under the pressure of a disaster situation to protect disenfranchised
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 16 / 20
and vulnerable populations. Preparedness is a key topic, particularly for the different climate
change scenarios and their effect on extreme weather events like ENSO or Coastal El Niño
events.
Based on the findings, we conclude that the necessary improvements to Peru’s disaster pre-
paredness and response under all climate change scenarios should span a broad set of issues at
different geopolitical scales and impacts tailored to the diversity of vulnerabilities. From terri-
torial zoning, strategic health facility placement in the Regions, and versatile information and
communication strategies to governance structures and financial mechanisms. Moreover,
since the frequency of ENSO events will likely increase and intensify with unabated green-
house gas emissions [12, 16], there is a need to improve territorial and environmental manage-
ment and preparedness. This can be addressed through local-level climate change risk
assessments (i.e., city level, community level) with a framework of auditable indicators and
ensure the assignment of a focal point person at the local level that manages adaptation and
mitigation plans, response and preparedness, communication, strategies, and effective
budget allocation. Cooperation mechanisms across local, regional, central government and
private sectors should be active in the long term and not only during the emergency response.
These issues will be critical to the country’s ability to anticipate and mitigate future disasters.
Supporting information
S1 Appendix. Economic activity impact estimation methodology.
(DOCX)
S2 Appendix. Selected health cases in Peru (2015–2017).
(DOCX)
Acknowledgments
We would like to thank the staff within the governmental institutions for their support in shar-
ing data, opinions and perspectives that were key to the different analyses of the study. We are
grateful for Arthur Wyns’ support in language editing.
Author Contributions
Conceptualization: Marisol Yglesias-Gonza´lez, Armando Valde´s-Vela´squez, Stella M. Hartin-
ger, Guillermo Salvatierra.
Data curation: Stella M. Hartinger, Ken Takahashi, Guillermo Salvatierra, Rodrigo Velarde,
Alvaro Contreras.
Formal analysis: Stella M. Hartinger, Ken Takahashi, Guillermo Salvatierra, Rodrigo Velarde,
Alvaro Contreras.
Funding acquisition: Marisol Yglesias-Gonza´lez.
Investigation: Stella M. Hartinger, Ken Takahashi, Guillermo Salvatierra, Rodrigo Velarde,
Alvaro Contreras, Hugo Santa Marı´a.
Methodology: Armando Valde´s-Vela´squez, Stella M. Hartinger, Ken Takahashi, Guillermo
Salvatierra, Rodrigo Velarde, Hugo Santa Marı´a, Andre´s G. Lescano.
Project administration: Stella M. Hartinger.
Resources: Stella M. Hartinger, Ken Takahashi, Guillermo Salvatierra, Rodrigo Velarde,
Alvaro Contreras.
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 17 / 20
Supervision: Marisol Yglesias-Gonza´lez, Stella M. Hartinger.
Validation: Stella M. Hartinger, Hugo Santa Marı´a, Andre´s G. Lescano.
Visualization: Marisol Yglesias-Gonza´lez, Ken Takahashi, Andre´s G. Lescano.
Writing – original draft: Armando Valde´s-Vela´squez, Stella M. Hartinger, Ken Takahashi,
Guillermo Salvatierra, Rodrigo Velarde, Alvaro Contreras, Hugo Santa Marı´a, Andre´s G.
Lescano.
Writing – review & editing: Marisol Yglesias-Gonza´lez, Armando Valde´s-Vela´squez, Marina
Romanello, Valerie Paz-Solda´n, Juan Bazo, Andre´s G. Lescano.
References
1. Timmermann A, An S-I, Kug J-S, Jin F-F, Cai W, Capotondi A, et al. El Niño–Southern Oscillation
complexity. Nature. 2018; 559: 535–545. https://doi.org/10.1038/s41586-018-0252-6 PMID:
30046070
2. Yeh S-W, Cai W, Seung-Ki Min, McPhaden MJ, Dommenget D, Dewitte B, et al. ENSO Atmospheric
Teleconnections and Their Response to Greenhouse Gas Forcing. 2018; 56: 185–206. Available:
https://doi.org/10.1002/2017RG000568
3. Taschetto AS, Ummenhofer CC, Stuecker MF, Dommenget D, Ashok K, Rodrigues RR, et al. ENSO
Atmospheric Teleconnections. Washington (DC): American Geophysical Union; 2020. https://
agupubs.onlinelibrary.wiley.com/doi/epub/10.1002/9781119548164
4. Takahashi K, Dewitte B. Strong and moderate nonlinear El Niño regimes. Repositorio institucional—
IGP. 2016 [cited 21 Aug 2018]. http://repositorio.igp.gob.pe/handle/IGP/2362
5. Takahashi K, Karamperidou C, Dewitte B. A theoretical model of strong and moderate El Niño
regimes. Repositorio institucional—IGP. 2018 [cited 21 Aug 2018]. https://doi.org/10.1007/s00382-
018-4100-z
6. L’Heureux ML, Takahashi K, Watkins AB, Barnston AG, Becker EJ, Di Liberto TE, et al. Observing and
Predicting the 2015/16 El Niño. Bulletin of the American Meteorological Society. 2017; 98: 1363–1382.
https://doi.org/10.1175/BAMS-D-16-0009.1
7. Takahashi K, Martı´nez AG. The very strong coastal El Niño in 1925 in the far-eastern Pacific. Climate
Dynamics. 2017 [cited 27 Aug 2018]. https://doi.org/10.1007/s00382-017-3702-1
8. ENFEN. El Niño costero 2017. Informe Te´cnico Extraordinario N˚001-2017/ENFEN. 2017. http://www.
imarpe.pe/imarpe/archivos/informes/imarpe_inftco_informe__tecnico_extraordinario_001_2017.pdf
9. Schott Gerhard. Der Peru-Strom und seine nordlichen Nachbargebiete in normaler und anormaler Aus-
bildung. Ann Hydrogr Marit Meteorol. 1931; 59: 161–196.
10. Peng Q, Xie S-P, Wang D, Zheng X-T, Zhang H. Coupled ocean-atmosphere dynamics of the 2017
extreme coastal El Niño. Nat Commun. 2019; 10: 298. https://doi.org/10.1038/s41467-018-08258-8
PMID: 30655541
11. Echevin V, Colas F, Espinoza-Morriberon D, Vasquez L, Anculle T, Gutierrez D. Forcings and Evolution
of the 2017 Coastal El Niño Off Northern Peru and Ecuador. Frontiers in Marine Science. 2018; 5: 367.
https://doi.org/10.3389/fmars.2018.00367
12. Cai W, Santoso A, Wang G, Yeh S-W, An S-I, Cobb KM, et al. ENSO and greenhouse warming. Nature
Climate Change. 2015; 5: 849–859. https://doi.org/10.1038/nclimate2743
13. Hald T, Aspinall W, Devleesschauwer B, Cooke R, Corrigan T, Havelaar AH, et al. World Health Organi-
zation Estimates of the Relative Contributions of Food to the Burden of Disease Due to Selected Food-
borne Hazards: A Structured Expert Elicitation. PLOS ONE. 2016; 11: e0145839. https://doi.org/10.
1371/journal.pone.0145839 PMID: 26784029
14. Bell JE, Brown CL, Conlon K, Herring S, Kunkel KE, Lawrimore J, et al. Changes in extreme events and
the potential impacts on human health. J Air Waste Manag Assoc. 2018; 68: 265–287. https://doi.org/
10.1080/10962247.2017.1401017 PMID: 29186670
15. Kluger LC, Kochalski S, Aguirre-Velarde A, Vivar I, Wolff M. Coping with abrupt environmental change:
the impact of the coastal El Niño 2017 on artisanal fisheries and mariculture in North Peru. ICES Journal
of Marine Science. 2018; 76: 1122–1130. https://doi.org/10.1093/icesjms/fsy171
16. Cai W, McPhaden MJ, Grimm AM, Rodrigues RR, Taschetto AS, Garreaud RD, et al. Climate impacts
of the El Niño–Southern Oscillation on South America. Nat Rev Earth Environ. 2020; 1: 215–231.
https://doi.org/10.1038/s43017-020-0040-3
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 18 / 20
17. Curtis S, Fair A, Wistow J, Val DV, Oven K. Impact of extreme weather events and climate change for
health and social care systems. Environ Health. 2017; 16: 128. https://doi.org/10.1186/s12940-017-
0324-3 PMID: 29219105
18. Hashim JH, Hashim Z. Climate Change, Extreme Weather Events, and Human Health Implications in
the Asia Pacific Region. Asia Pac J Public Health. 2016; 28: 8S–14S. https://doi.org/10.1177/
1010539515599030 PMID: 26377857
19. Capotondi A, Wittenberg AT, Newman M, Lorenzo ED, Yu J-Y, Braconnot P, et al. Understanding
ENSO Diversity. Bull Amer Meteor Soc. 2014; 96: 921–938. https://doi.org/10.1175/BAMS-D-13-
00117.1
20. Lavado-Casimiro W, Espinoza JC. Impactos de El Niño y La Niña en las lluvias del Peru´ (1965–2007).
Rev bras meteorol. 2014; 29: 171–182. Available: https://www.scielo.br/j/rbmet/a/
8yxvzcDjv6JpT5FFDVj5ZMR/?lang=es&format=pdf
21. Sulca J, Takahashi K, Espinoza J-C, Vuille M, Lavado-Casimiro W. Impacts of different ENSO flavors
and tropical Pacific convection variability (ITCZ, SPCZ) on austral summer rainfall in South America,
with a focus on Peru. International Journal of Climatology. 2018; 38: 420–435. https://doi.org/10.1002/
joc.5185
22. Ministerio del Ambiente Peru. [SENAMHI] Informe Te´cnico El Niño Costero 2017 Condiciones Termo—
Pluviometricas a Nivel Nacional SENAMHI Peru 2017 (Biblioteca SIGRID). 2017 [cited 21 Apr 2021].
https://sigrid.cenepred.gob.pe/sigridv3/documento/3656
23. INDECI. Compendio Estadı´stico del INDECI: Gestio´n Reactiva. Instituto Nacional de Defensa Civil;
2017. https://www.indeci.gob.pe/wp-content/uploads/2019/01/201802271714541.pdf
24. INDECI. RESUMEN EJECUTIVO HISTORICO—TEMPORADA DE LLUVIAS, 2016–2017: (Actuali-
zado al 24 de Enero 2018)—INDECI Tarea de Todos. In: INDECI Tarea de Todos [Internet]. 2018 [cited
21 Apr 2021]. https://www.indeci.gob.pe/emergencias/resumen-ejecutivo-historico-temporada-de-
lluvias-2016-2017-actualizado-al-24-de-enero-2018-2/
25. Takahashi K, Aliaga-Nestares V, Avalos G, Bouchon M, Castro A, Cruzado L, et al. The 2017 coastal El
Niño [in “State of the Climate in 2017”]. 2018 pp. S210–S211. Report No.: 99 (8). Available: https://
www.ametsoc.net/sotc2017/StateoftheClimate2017_lowres.pdf
26. UNICEF. UNICEF Peru Humanitarian Situation Report #18, January—March 2018. 2018. https://
reliefweb.int/report/peru/unicef-peru-humanitarian-situation-report-18-january-march-2018
27. OPS. Lluvias e inundaciones en Peru´. Reporte de Situacio´n N˚ 8. Organizacion Panamericana de la
Salud; 2017.
28. Caritas. Informe final: Llamamiento de Emergencia por Inundaciones y huaicos en el Peru´. Caritas Del
Peru. 2018.
29. Takahashi K, Montecinos A, Goubanova K, Dewitte B. ENSO regimes: Reinterpreting the canonical
and Modoki El Niño. Geophysical Research Letters. 2011; 38. https://doi.org/10.1029/2011GL047364
30. Dewitte B, Takahashi K. Diversity of moderate El Niño events evolution: role of air–sea interactions in
the eastern tropical Pacific. Clim Dyn. 2019; 52: 7455–7476. https://doi.org/10.1007/s00382-017-
4051-9
31. Sistema Estadı´stico Nacional. Tumbes Compendio Estadı´stico 2017. 2017 p. 550. https://www.inei.
gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib1509/libro.pdf
32. Sistema Estadı´stico Nacional. A´ ncash Compendio Estadı´stico 2017. 2017. https://www.inei.gob.pe/
media/MenuRecursivo/publicaciones_digitales/Est/Lib1488/libro.pdf
33. Sistema Estadı´stico Nacional. Compendio estadı´stico 2019 Peru. INEI; 2019. https://www.inei.gob.pe/
media/MenuRecursivo/publicaciones_digitales/Est/Lib1690/COMPENDIO2019.html
34. PERU´ Instituto Nacional de Estadı´stica e Informa´tica. ENAHO Metodologı´a Actualizada, encuesta Con-
diciones de Vida y Pobreza 2017. 2017 [cited 26 Apr 2021]. http://iinei.inei.gob.pe/microdatos/
35. SENAMHI. Avisos meteorolo´gicos. In: Aviso N˚ 038 nivel 4 [Internet]. Mar 2017. https://web2.senamhi.
gob.pe/?p=avisos-detalle&a=2017&b=038&c=022&d=SENA
36. Huang B, Thorne PW, Banzon VF, Boyer T, Chepurin G, Lawrimore JH, et al. Extended Reconstructed
Sea Surface Temperature, Version 5 (ERSSTv5): Upgrades, Validations, and Intercomparisons. Jour-
nal of Climate. 2017; 30: 8179–8205. https://doi.org/10.1175/JCLI-D-16-0836.1
37. Adler RF, Huffman GJ, Chang A, Ferraro R, Xie PP, Janowiak J, et al. The version-2 global precipitation
climatology project (GPCP) monthly precipitation analysis (1979–present). Journal of Hydrometeorol-
ogy. 2003; 4: 1147–1167. https://doi.org/cb5d4s
38. BCRP. Base de datos de estadı´sticas del BCRP. [cited 21 Apr 2021]. https://estadisticas.bcrp.gob.pe/
estadisticas/series/
39. Webb, Richard BF Graciela. Peru en Nu´meros 2000. Lima, Peru: Instituto Cuanto; 2000.
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 19 / 20
40. ARCC. Decreto Supremo N˚ 091-2017-PCM. 11 Sep 2017 [cited 21 Apr 2021]. https://www.rcc.gob.pe/
2020/decreto-supremo-091-2017-pcm/
41. Food and Agriculture Organization. Ana´lisis de Sistemas de Gestio´n del Riesgo de Desastres: Una
Guı´a. 2009. http://www.fao.org/3/i0304s/i0304s.pdf
42. Pan American Health Organization. Gestio´n de la informacio´n y comunicacio´n en emergencias y desas-
tres: GUI´A PARA EQUIPOS DE RESPUESTA. OPS/OMS; 2009. https://www.paho.org/chi/
dmdocuments/GestionDeInformComunica_LowRes%20Dic%2009.pdf
43. Jauregui YR, Takahashi K. Simple physical-empirical model of the precipitation distribution based on a
tropical sea surface temperature threshold and the effects of climate change. Clim Dyn. 2018; 50:
2217–2237. https://doi.org/10.1007/s00382-017-3745-3
44. Galarza E, Ka´miche J. Impactos del Feno´meno El Niño (FEN) en la economı´a regional de Piura, Lam-
bayeque y La Libertad. Cooperacio´n Alemana al Desarrollo.; 2012.
45. Corporacio´n Andina de Fomento. Feno´meno El Niño 1997–1998 Memoria, Retos y Soluciones. 2000.
46. BCRP. Memoria Anual 1998. 1998. https://www.bcrp.gob.pe/publicaciones/memoria-anual/memoria-
1998.html
47. Va´squez, K. Determinantes del crecimiento Agroexportador. Revista Moneda. 2015161: 22–28. https://
www.bcrp.gob.pe/docs/Publicaciones/Revista-Moneda/moneda-161/moneda-161-05.pdf
48. Contralorı´a General de la Repu´blica del Peru´. N˚ 562-2019-CG-GCOC: ESTADO EJECUTO´ 27% DE
INVERSIO´ N APROBADA PARA LA RECONSTRUCCIO´ N CON CAMBIOS. 2019 [cited 28 Sep 2021].
https://www.contraloria.gob.pe/wps/wcm/connect/cgrnew/as_contraloria/prensa/notas_de_prensa/
2019/lima/np_562-2019-cg-gcoc
49. ARCC. Nota de Prensa: Peru´ suscribio´ convenio con Reino Unido para recibir asistencia en obras de
prevencio´n y reconstruccio´n. 2020 [cited 20 May 2021]. https://www.rcc.gob.pe/2020/peru-suscribio-
convenio-con-reino-unido-para-recibir-asistencia-en-obras-de-prevencion-y-reconstruccion
50. Ministerio de Salud del Peru´. Dengue por ditritos a nivel regional. Año 2010–2020. NOTI SERIES. In:
Tableau Software [Internet]. 2020 [cited 28 Sep 2021]. https://public.tableau.com/views/NOTISERIES_
15580211651890/1_TablaCasosSE?:embed=y&:showVizHome=no&:host_url=https%3A%2F%
2Fpublic.tableau.com%2F&:embed_code_version=3&:tabs=no&:toolbar=yes&:animate_transition=
yes&:display_static_image=no&:display_spinner=no&:display_overlay=yes&:display_count=yes&:
loadOrderID=0%20 (accessed%203.26.21)
51. CDC Peru´. Situacion de epidemilo´gica de Dengue. 2022. https://www.dge.gob.pe/salasituacional/sala/
index/SalaRegional/145
52. Congreso de la Repu´blica. Ley N˚ 29230: Ley que impulsa la inversio´n pu´blica regional y local con parti-
cipacio´n del sector privado. 2008. https://www.proinversion.gob.pe/RepositorioAPS/0/0/arc/MOXI_
LEY_29230/Ley-29230.pdf
53. Congreso de la Repu´blica. Ley N˚ 29664: LEY QUE CREA EL SISTEMA NACIONALDE GESTIO´ N
DEL RIESGO DE DESASTRES(SINAGERD). 2011. https://busquedas.elperuano.pe/normaslegales/
ley-que-crea-el-sistema-nacional-de-gestion-del-riesgo-de-de-ley-n-29664-605077-1/
54. Orihuela JC. ¿Reconstruccio´n sin instituciones? Gobierno de los desastres socionaturales en el Peru´
1997–2019. Lima, Peru: Grupo Propuesta Ciudadana; 2019 Oct p. 61. https://propuestaciudadana.org.
pe/wp-content/uploads/2019/10/Reconstrucci%C3%B3n-sin-instituciones-Gobierno-de-los-desastres-
socionaturales-en-el-Per%C3%BA-1997-2019.pdf
PLOS ONE Impact and response to the Peruvian 2017 Coastal El Niño event
PLOS ONE | https://doi.org/10.1371/journal.pone.0290767 September 26, 2023 20 / 20