Portraying Animals in Disasters: news media representations of the Portuguese wildfires of 2017

1 Policarpo, V., Soares, F., Zhang, X., & Pereira da Costa, L. (2025). Portraying Animals in Disasters: News Media Representations of the Portuguese Wildfires of 2017. Environmental Communication, 1–17. https://doi.org/10.1080/17524032.2025.2576721 'The Version of Record of this manuscript has been published and is available in https://www.tandfonline.com/doi/full/10.1080/17524032.2025.2576721 2 TITLE: Portraying Animals in Disasters: news media representations of the Portuguese wildfires of 2017 Authors: Verónica Policarpo, Instituto de Ciências Sociais, Universidade de Lisboa (corresponding author) ORCID: https://orcid.org/0000-0002-9245-1057 Filipa Soares, Instituto de Ciências Sociais, Universidade de Lisboa Xinyue Zhang, School of Research, Sciences Po Paris Leonor Pereira da Costa, Lusófona University, HEI‐Lab: Digital Human‐Environment Interaction Labs. ABSTRACT: This article explores how news media report and construct the situation of animals in disasters, drawing on a specific case study: the tragic wildfires that hit Portugal in 2017. It explores to what extent animals are present in the narration of the wildfires, and how they are included in these narratives, indirectly participating in the communication of disasters. Drawing on thematic content analysis of animal-related content published in Portuguese newspapers and primetime television news, we examined, through multivariate Multiple Component Analysis, followed by Cluster analysis, whether the resulting thematic macro-categories clustered around narrative topoi. We propose four portrays of media narratives about animals in wildfires, which highlight their entangled lives with humans: risk, emergency and rescue; emotional relationship and vulnerability; economic relationship and loss; and political and economic (lack of) support. We explore whether these four discursive spaces reinforce or challenge a human-centred approach to disasters’ representation. We argue that insofar as the news value attributed to animals is linked to human experience, as economic or emotional loss, it concurs to erase animals as subjects with interests and lives of their own from the media agenda. We further argue that such type of visibility of animals paradoxically relates to the invisibility of the structural conditions that ultimately contribute to the vulnerability of both humans and nonhumans to disasters. The article ends with reflections and recommendations on how to develop compassionate forms of journalism and media communication towards other species. KEYWORDS: wildfires; disasters; animals; media; media coverage; Portugal. 1. Introduction In the summer of 2017, Portugal underwent its deadliest wildfire season, claiming the lives of 116 civilians (San-Miguel-Ayanz et al., 2020) and of countless non-human animals. Over half a million hectares burnt, with significant economic, social, and environmental impacts. It was the most catastrophic wildfire season on record in Portugal and one of the worst in Europe (CTI, 2018; San-Miguel-Ayanz et al., 2020). 3 The two most catastrophic situations occurred in mid-June and mid-October in central Portugal, both outside the official wildfire season. The first, known as the ‘Pedrógão Grande wildfires’ (hereafter ‘June wildfires’), started on June 17th, and lasted one week, burning over 45,000 hectares (CTI, 2017). It affected mostly the municipality of Pedrógão Grande and neighbouring municipalities. These fires caught national and international attention in many ways, but mostly due to the death of 65 civilians (CTI, 2017; Viegas et al., 2017), which made it the most tragic wildfire in Portuguese recorded history. Just as the country was recovering from the shock, another set of wildfires blasted the north and centre of Portugal on October 15th (hereafter ‘October wildfires’), notably in areas close to Pedrógão Grande. Multiple simultaneous fires over two days resulted in a very large burnt area (over 210,000 hectares) and 51 human casualties (CTI, 2018; San-Miguel-Ayanz et al., 2020; Viegas et al., 2019). The ‘unprecedented’ and tragic conditions of both wildfires would, expectedly, called the attention of the media. But were nonhuman animals (hereafter, animals i ) part of the public sphere of concern? This interrogation arises from a striking invisibility of animals in analyses of media coverage of disasters in general—and wildfires in particular. For instance, Pantti et al.’s (2012) groundbreaking and comprehensive work on media and disasters has no special section dedicated to the topic of animals, nor does the word ‘animal’ appear in the onomastic index. In their study about conflicting discourses of risk and local media representations of fires in Amazonia, Silva et al. (2019) mention animals only in relation to animal husbandry. The blurred categories ‘wildlife’ and ‘biodiversity’ seem to get more attention, although not without conflicts, such as the overexposure of charismatic species, with studies calling for a deeper and broader engagement between media practitioners and scientists in the field (Gessa et al., 2024; Silva et al., 2019; Smit et al., 2022). In their study about media depictions of fires in South African National Parks, Smit et al. (2022) mention animals only once, as “animals’ deaths or welfare”. The latter occur roughly below 10% and are higher among pieces relying on scientists’ input than non-scientists’. On the other hand, while media narratives may foster public opinions, media and the public’s representations do not always align. Jacobson et al. (2001) describe the mismatch between a media focus on the spread of fires as a detrimental effect of prescribed burning, as opposed to the public, who elected the harm to wild animals. Drawing upon the June and October 2017 wildfires in Portugal, this article explores how news media communicate and represent the situation of animals in disasters. More specifically, it asks: were animals widely covered by the Portuguese mass media while reporting the 2017 wildfires? If so, how are the media talking about them? Which is the kind of discourse produced, and with what implications to the social, political and ethical dimensions of humananimal relationships? Analysing the media discourses and representations of animals during a disaster is crucial for a more thorough understanding of the multifaceted impacts of disasters and the interdependencies between humans, animals, and the environment. We argue that including animals in media narratives and communication about disasters is critical to reduce risk. By inclusion we mean not only rendering them visible but also acknowledging them as subjects with a life of their own. We align with Pezzullo’s (2024) proposal to re-imagine environmental communication as a care discipline, with the duty not only to prevent harm, but also to honour those with which we share our world, humans and nonhumans alike. This article is structured as follows. In the next section, we briefly describe the societal context in which the severe wildfires of 2017 occurred, in Portugal. We then address the theoretical background in media studies that may help us to understand how animals are being made 4 visible and invisible, followed by the methods section. We proceed to present our main results, a topology of media coverage about animals caught in the fires. In the end, we summarise our main conclusions, as well as future steps and some recommendations for media practitioners. 2. Background and context: Portugal’s wildfires of 2017 In recent decades, wildfires have become more frequent, larger, and more destructive in Mediterranean Europe—the most affected area in Europe. Herein, Portugal stands out. Between 1980 and 2021, when nearly one third of the total area of the country burned, it had the highest number of fires and the highest percentage of burnt area recorded (San-MiguelAyanz et al., 2022). The 2017 wildfires, as mentioned before, were particularly tragic. They marked a turning point in public policy and opinion regarding wildfires (Madeira et al., 2023). In their aftermath, and due to the unprecedented dimensions, severity, and deadly impacts, the Portuguese government requested technical reports by a team of experts from a research centre (Viegas et al., 2017, 2019). In turn, the Parliament created an independent technical commission to provide detailed reports (CTI, 2017, 2018). The objectives were the same: to analyse both events and to suggest policy changes. None of the documents produced mentions the impacts of the wildfires on animals, nor the loss of nonhuman lives. When animals are mentioned, it is only briefly and indirectly—and only those that hold economic and emotional values (companion and domestic animals). For instance, one of the reports about the June wildfires mentions complaints about a lack of support in terms of food for people and their surviving animals (Viegas et al., 2017). Regarding the October wildfires, animals are mentioned as an indirect cause of human fatalities and injuries. It is reported that half of those that died were trying to save companion animals, livestock, and beekeeping equipment, which are included under the label “assets” (CTI, 2018, p. 138). Another report mentions cases of people that died or were seriously injured when trying to save their domestic animals (Viegas et al., 2019). For the authors, “no matter how much emotional or economic value one or more animals may have, they do not justify sacrificing the physical integrity or life of a person” (p. 233). The actual number of animals that were directly or indirectly affected by the wildfires is unknown. The only official records available pertain to livestock. For the June wildfires, there are estimates of the death of “several dozens” of sheep, goats, and poultry (MAFDR, 2018). For the October wildfires, over 500 thousand animals died: mostly poultry (around 500,000), followed by sheep and goats (5398), pigs (1091), and cattle (881) (MAFDR, 2017). In what concerns other types of animals, there are no official records. Likewise, animals are seemingly absent in the various public policies enacted in response to the 2017 wildfires, and institutional (and scientific) discourses in relation to these wildfires seem to render them invisible. The media agenda being closely intertwined with the political agenda (McCombs & Shaw, 1972McCombs & Shaw, 1972McCombs & Shaw, 1972McCombs & Shaw, 1972), was this political invisibility of animals reflected in the media coverage of the 2017 events? If not, how visible were they? How were they represented and framed, in terms of discourse? In the following section, we draw on the framework of Critical Media Studies to attempt to address these questions. 5 3. Representing animals in wildfires: a critical approach to media coverage Critical media studies have a long tradition of emphasising that media do not merely reflect or mirror reality in neutral ways, nor do they simply convey information and facts. Rather, social constructionist approaches have highlighted the professional practices and processes of news making, such as newsworthiness, or the professional practices inside the editorial room (Gans, 1979; Golding & Elliott, 1979). In such practices and languages tailored to capture the public’s attention, emotions play a key role, with the stories called of ‘human interest’ being asserted high value in the production of news, especially in the television and audiovisual formats. Such professional (and thus social) practices, among others, show how the media are constructors of a second order reality, with great power and influence in the political, economic and public agendas (McCombs & Shaw, 1972; Shaw, 1979). The agenda-setting function of the mass media highlights the influence that media exerts on public affairs by directing the public’s attention towards the topics that news media cover with the most salience. Approaches to the media as economic and political structures and agents have also stressed the underlying relationships between the media and the economic systems within which they operate, such as capitalism (Golding & Murdoch, 1997). Regarding the audience and media effects, media scholars have widely unveiled the complex processes of meaning making that challenged the direct effects hypothesis. In their seminal work, Lazarsfeld et al. (1944) elaborated on the importance of personal relationships as mediators of media messages aiming at influencing public opinion. Since then, social factors such as gender, social class, ethnic identity, family dynamics, and other forms of social interactions and belongings, were explored as adding complexity to the processes of meaning making. Audiences are considered active, as they use their own repertoires when decoding media messages. This means that decoding processes may also be empowering, countering the power structures and relations through which the media and cultural industries often silence the “voiceless” and the more vulnerable (Hall, 1973; Hoggarth, 1998 [1957]). Critical Media Studies, on their side, examine the ideologies and power relations ingrained within the media. They explore the role of media in further reinforcing the domination, inequality, and biases in contemporary capitalist societies. Their convergence with Critical Animal Studies gave rise to Critical Animal and Media Studies (Almiron et al., 2016 and 2018), whose purpose is to reveal the media’s long tradition of neglecting the exploitation and oppression of other animals by humans, rendering certain species invisible while glossing over the suffering of some others (Almiron et al., 2016 and 2018; Molloy, 2011). In this bourgeoning field, scholars examine various types of media to see how it perpetuates speciesism and anthropocentrism. Through specific public events, they investigate how media narrates animals, and their influence on the human-animal relationship. For example, Dare and Fletcher (2019) examine the public discourse of the armed occupation of the Malheur National Wildlife Refuge and its subsequent debates. They criticise how media discourse reinforces human exceptionalism, neglecting the intricate entanglement between humans, animals, and the land. Other scholars focus on media portraits of certain types of animals in comparison with others to shed light on the intricate relationship between media narratives, exploitations of non-human animals and broader discourse on speciesism. For instance, Freeman (2009) conducted a textual 6 analysis of over one hundred stories within national print and broadcast news on farmed animals. He highlights how news media frequently objectify animals through practices such as commodification, omission of their emotional experiences, and failure to portray them as inherently valuable individuals. Despite the diverse news genres and types of animals, one common yet significant thread weaving these works is the belief that the speciesist portrays of non-human animals and their exclusion from media visibility can lead to real-life consequences to the lives of animals, their welfare and survival. This is evident, for instance, in the case of farmed animals, whose (in)visibility reinforces and legitimises their breeding and killing for human utilitarian purposes. We thus join the discussion by looking at one specific scenario in which animals are deeply involved but nevertheless overlooked: disasters, in particular wildfires. As disaster studies scholars have been stating for long, disasters are not ‘natural’(Kelman, 2020). Rather, they are naturalcultural (after Haraway, 2008) processes. They are profoundly rooted in and intersect with existing social inequalities, shaping how different social groups experience and are affected by such catastrophes (Gill & Aldrich, 2014; Hoffman & OliverSmith, 2012; Kelman, 2024; Tierney, 2019; Wachinger et al., 2013). Media coverage of disasters often mirrors and reinforces these social inequalities. For instance, prior research has analysed how race and gender interact with media reporting of disasters, exacerbating the existing social inequalities by focusing on certain groups while sidelining or stigmatizing others (e.g., Kempton, 2019; Walker et al., 2020). Moreover, emotions play a key role in conveying messages about disasters. As Pantti et al. (2012) argue, emotions are “integral to processes of assigning meaning” (p. 61), and they shape how we understand, and then respond to, specific disasters. On the other hand, through the media and the representations they convey, we can process our individual experiences and emotions, transmuting them into political and collective ones (ibidem). Some studies have even demonstrated a positive role of negative emotions, such as fear and anxiety, in promoting adaptative behaviours, during a crisis (Seo, 2021). In their study about the media coverage of the wildfires that affected the coastal public-managed forest areas located in the Centre Region of Portugal, in October 2017, Figueiredo et al. (2024) found a difference between the national and local coverage of the news, according to the publication time, with local newspapers giving more attention to issues related to the recovery phase, months after the incident. The authors also found that institutional actors were given more relevance than local peoples, whose emotions were also less a focus whenever economic and material losses were at stake. This leads them to point to the heterogeneity of the media coverage of wildfires. As the social dimensions of disasters are uncovered, much of the scholarly focus remains centred on human experiences. This is also true in the Portuguese case, where academic studies about fires have not addressed the experiences of animals, with farmed animals only appearing as proxies of forms of land management (or lack thereof) (Oliveira et al., 2017). In other contexts, animals have surfaced in media coverage in different ways. Focusing on koalas and the impacts of the 2019-20 wildfires in Australia, Stalenberg et al. (2024) highlight the danger of relying on numbers as a discursive element when reporting on animals affected by fires. They approach the topic from a conservation biology point of view and call for a deeper 7 involvement of journalists with experts in the field, in order to provide more accurate reports. Focusing on the same context, Atchison & Pilkinton (2022) analysed 82 news pieces from private and public sources. They point to how Australian media built on the notion of “invasive species” to draw a narrative that presents them as “strongly negative or unworthy of moral status”, threatening native forms of life (p. 4), and that calls for action with lethal consequences for such species. In their article on media representations of ‘Sam the Koala’, a koala that became famous through a video footage during the 2009 Victorian wildfires in Australia (‘Black Saturday’), Due et al. (2014) discussed how Sam was pictured in 93 mainstream media pieces, in ways that contradict traditional disaster imagery. Sam gained human characteristics and carried a symbolism that turned her in an “identifiable victim”, triggering identification and empathy. In Brazil, Baptistella & Nobre (2024) focused on the wildfires of 2020 in Pantanal and report on the “petification” of wild animals in the media, with charismatic species such as the jaguar or the giant anteater getting much of the public attention, while bovines only appear as commodities and “economic losses” (p.18). However, as the authors stress, the depiction of animals as victims of human action does not obliterate species hierarchies, with flagship species getting more attention and personalization. In the Canadian context, and referring to the severe wildfires in Nova Scotia, in 2023, Ru et al. (2025) focus on the analysis of 57 pieces of grey literature news articles, Twitter posts, and official debriefings from government websites to discuss the role of media coverage as a protective factor for the populations. They conclude that the analysed media channels “played a key role in disseminating information that could minimize the risks faced by affected residents and animals during the wildfires” (p. 7). However, they focus more on the institutional use of media channels to convey such information, rather than on how the newsmakers report on the disasters, and animals therein. In this article, we address this gap from a critical point of view that assumes that media are not neutral producers of information. Rather they are complex political and economic institutions, with an active role in the construction of social reality. Notwithstanding this emergent presence of animals, the fact that they are equally vulnerable to disasters due to structural factors has still not been under media attention, another expression of the anthropocentric bias that cuts across disaster management and other social structures. During disasters, animals are affected in many direct and indirect ways, including sharing the same affected dwellings with humans. Yet their vulnerabilities tend to be overshadowed by those of humans’, reinforcing the predominant human-centric focus of the media (Irvine, 2009). 4. Materials and methods 4.1. Data collection and corpus This study explores the mainstream media coverage of the Portuguese wildfires of June and October 2017. Drawing on previous research experience of the first author (REFERENCE 2002, 2015, 2019), and on the vast experience in data processing and analysis of the first and fourth authors, we followed an ad-hoc methodology, specifically addressing our research objectives. To ensure a rich data corpus, we focused on Portuguese national press newspapers and television primetime news broadcasts. The corpus comprised three mainstream printed 8 newspapers: Público, a daily quality newspaper known for its in-depth reporting and comprehensive coverage of various topics, with an average audience of 4,6% in 2017; Correio de Manhã, one of the most widely read daily newspapers in Portugal, with an average audience of 11,6% in 2017, recognised for its tabloid-style format; and Expresso, a weekly quality newspaper with a broad national readership, with an average audience of 5,4% in 2017 (OberCom - Observatório da Comunicação, 2024). This selection enables news of different time frequencies and reporting formats. The television news segments were collected from two major Portuguese primetime programs: Telejornal, from the public broadcasting service RTP - Rádio e Televisão de Portugal; and Jornal Nacional, from TVI - Televisão Independente, a private channel. These two programs represent the largest public and private television channels in Portugal, with an average audience of 2,7% and 4,3% in 2017, respectively (OberCom - Observatório da Comunicação, 2024). From these media outlets, we screened and collected news related to our topic of interest, wildfires, focusing on the June and October 2017 wildfire events. For a comprehensive coverage of the entire news lifecycle, the selected time window spanned from the outset of each wildfire to one and a half months later (from June 17th to July 31st; and from October 15th to November 30th) until the fire had largely disappeared from the media landscape. All the news pieces that had any mention to the fires were selected and registered in a database built for this purpose and were codified according to several variables, such as date, media name, disaster (June or October), or location. One dichotomic variable registered whether the news made any kind of reference to animals or not, either textual or visual. To identify articles featuring animals, we searched for animal-related keywords or images (photos and televised content), such as ‘animals’, ‘cows’, ‘sheep’, ‘dogs’, ‘biodiversity’, etc. For the news that made references to animals, we further codified them according to a wider range of variables, which included the sources, keywords, actors and/or stakeholders, the locations, and the main thematic codes. In total, 2409 news pieces about wildfires were collected, 1790 of which pertained to the disaster occurred in June, and 609 to the wildfires that occurred in October (Table 1). Most of these pieces were published by the newspaper Correio da Manhã (n=1008). This high number is due to its tabloid format, in which the news values are massively associated with the dramatic aspects of the disaster (deaths, losses), and to the tabloid format itself. It was followed at a large distance by the private television broadcaster TVI (n=506), the reference newspaper Público (n=391), and the public television broadcaster RTP (n=348). Expectedly, the weekly newspaper Expresso, with fewer editions (one per week), showed the smallest number of occurrences (n=156). Table 1 – Description of the corpus: Number and percentage of pieces about fire, by name of Media and Disaster Disaster Total Name of Media June 2017 October 2017 Other Printe d newsp apers Público N 269 122 0 391 9 % of Total 11,2% 5,1% 0,0% 16,2% Expresso N 98 58 0 156 % of Total 4,1% 2,4% 0,0% 6,5% Correio da Manhã N 904 99 5 1008 % of Total 37,5% 4,1% 0,2% 41,8% Televisions RTP N 218 130 0 348 % of Total 9,0% 5,4% 0,0% 14,4% TVI N 302 204 0 506 % of Total 12,5% 8,5% 0,0% 21,0% Total N 1791 613 5 2409 % of Total 74,3% 25,4% 0,2% 100,0% 4.2. Data Analysis To examine how animals were portrayed within the narratives of the two wildfires, we conducted a thematic analysis of the news that referred to animals. Content analysis is frequently used in media studies for identifying, analysing, and reporting patterns and themes within data (Braun & Clarke, 2006). It has the potential to investigate how media professionals, in this case, journalists, conceptualise social constructs, and therefore propose a certain interpretation of events (Krippendorf, 2004; Willig, 2003), addressing both the manifest, or explicit, textual meaning, and the latent — implied or inferred — meaning surrounding an issue (Joffe, 2011). This will be key to the selection of the saliences that will constitute the public agenda and, therefore, to the construction of public opinion. The thematic analysis followed a multi-step coding process. Firstly, one coder classified a subcorpus, in an in-depth approach which allowed the inductive construction of categories suggested by the materials themselves. Secondly, the resulting set of categories was put to test to the different media sources, verifying their descriptive power across different media affordances and platforms. Thirdly, this set of categories was systematically used to classify the wider set of news, across the several media sources. To identify potential recurring themes in the data, the same coding process was repeated until data saturation had been achieved (i.e., no new themes emerged from the data). The coding was performed in different moments in time, and by four different coders. The first author coordinated the coding process, defined the 16 chain prevented the burning. However, this presents a bleak twist: it mystifies the structural conditions at the very root of the vulnerability of animals. Topos 3: Economic relationship A third discursive space focuses on the economic consequences of wildfires for the most vulnerable human populations, particularly the elderly. It draws mainly on emotions to produce a discourse around grief and loss, for families, victims and communities (Pantti et al., 2012). Animals add to this, appearing as economic losses. This was already the case in the previous topos, where the emotional undertone was always present, suggesting bond and connection as an underlying sub-narrative. The difference is that here the focus is mainly on the economic loss, the forms of support (mainly financial) that the State, the private sector, or civil society organisations provide to the human victims. Like Baptistella & Nobre (2024) found in their analysis of Brazilian news about Pantanal, cows, sheep or goats are represented as “livestock”, commodities now subsumed to economic losses. For instance: Many of these people lived from livestock breeding and saw the fire steal away their sustenance. (…) ‘Were it not for the major disgrace, I wouldn’t be able to relativise the fact that I lost my breadwinning’. (Tragédia em Pedrógão, 2017) Some of these pieces also focus on the types, amounts, and timings of the monetary support provided by public authorities – or the lack thereof. One recurrent topic is the fact that, unlike houses, it is still uncertain whether the recovery funds cover domestic animals and the facilities where they are kept: “The [charity] Santa Casa [da Misericórdia] asked us if we wanted chickens. The problem is that we lack the facilities to put them.” (Faria & Garrido, 2017). This is the discursive space where more deaths of animals are acknowledged, as they stand for the loss of human livelihoods: “The country mourns once more the loss of dozens of human lives and hundreds of animals, who are the livelihood of those who make their living from the land and pastoralism.” (País a arder deixa rasto de destruição, 2017). These accounts may escalate in the visual messages that sometimes complement the news. For example, one photograph of the tabloid newspaper shows how “inside of the warehouse there are still 108 carbonised sheep” (É um perigo para a nossa saúde, 2017). Another sub-narrative runs alongside this image: the critique of public authorities and their slow response to the needs of the populations in the aftermath of the disaster. In this case, the focus is on public health issues: “I don’t know what one needs to do for them to come and take away the [carbonised] sheep. It is an intense smell, a danger for our health.” (É um perigo para a nossa saúde, 2017). Mentions to the high numbers of animal lives’ losses are also linked to the impact on related economic activities, such as cheese and honey production: “Serra da Estrela [a famous special cheese produced in the region] at risk, with 5 thousand sheep killed.” (Rodrigues & Pinto, 2017). Overall, and again, animals and their horrific deaths appear as proxies for human economic losses and problems, rather than as subjects of suffering themselves. The visual language of the news reiterates this erasure: unlike humans’ deaths, the deaths and corpses of animals are openly displayed, used denotatively as referents of burnt goods. 17 Topos 4: Political and economic (lack of) support The fourth and final discursive space addresses the economic and social consequences of the fires, with an emphasis on the political and economic measures to support the local populations. This is the second most prevalent topos (33%). Once again, it draws heavily on the representation of animals as economic property—therefore economic loss. Recurring topics include the sharp decline of production of cheese or honey. Like the previous topos, the risk/safety or death of sheep and goats appear as a major concern for the affected populations. However, the difference is that, in this topos, the overarching tone conveys a tension between a demand for support from authorities, and a critique of their actual performance, both during and after the emergency phase. One dimension of such critique is the failure of support systems. As the president of a local governance institution and coordinator of the veterinarian team working in the affected areas states: “there is not a contingency plan to help in these situations (…) we have asked for support from SIRCA [system of collection of carcasses from food explorations] but it has failed.” (Valente & Batista, 2017). This is thus a discursive space where emotions are mobilised to produce anger, blame and responsibility (Pantti et al., 2012), namely of the central authorities. This concurs to produce a discourse in which the affected populations are depicted as being on their own, relying only on themselves and in streams of solidarity by civil society, as the following examples show: “We are alone, we need help, we lost 40 people.” (Pedida ração, 2017); “It was of isolation and lack of communication that many people complained about, while surrounded by the flames” (Villalobos, 2017); “We do not have firemen, nor Air Force, we have nothing. It was based on our own arms force.” (RTP, 2017c); “Everyone sighs for help. ‘Any form of help that comes is welcome.’” (RTP, 2017b). Support comes in several forms and by different actors. As opposed to the perceived abandonment by the authorities, praise emerges for veterinarians, associations, NGOs, enterprises, fellow municipalities and farmers from other regions, and other civil society actors. Local and volunteering veterinarians and clinics are praised for their help in treating injured animals, collecting the corpses of dead animals, and providing food and medicines for companion animals. Other actors (NGOs, associations, corporate enterprises) are praised for providing economic help, mostly in genera – food for the animals, but also construction materials to rebuild houses, barns and animals’ facilities: Several animal protection associations – and others – are raising food and pharmaceutical products for the treatment of the affected animals. (Apoio a animais, 2017) The Order of the Veterinarians created a crisis group and provided food and medicines for the affected animals. It opened an account and is collecting donations. (Recolha de ajuda, 2017) On the other hand, public support comes through public (economic) measures to feed animals: “Tons of sugar to feed honeybees. Government supports beekeepers affected by fires.” (Toneladas de açúcar para alimentar abelhas, 2017); “we are going to make payments to compensate for losses in vegetable gardens and animals up to 1053 euros.” (Vamos fazer pagamentos para compensar perdas em hortas e animais até 1053 euros, 2017). 18 6. Conclusions In this article, we explored whether and how animals are visible in media mainstream news about wildfires. We took as a case study two extreme wildfires that took place in Portugal, in June and October 2017. Due to their uniqueness—both in the Portuguese and European contexts—in terms of scale, magnitude, extension and number of fatalities, they represented a turning point in public policy and opinion regarding wildfires, including animals therein. Across the two different timelines that framed both incidents, was there an increasing visibility of animals in the flow of published and broadcasted news? Our analysis of the news covering both wildfires in three national newspapers and two television primetime news programs, along the news cycle, showed otherwise: only 8% (n=195) of the news pieces mentioned animals. This clearly points to the low visibility of animals in media narratives of disasters. Due to the lack of data from previous years to provide a comparison, we cannot assert whether 8% represents an actual increase or not in visibility. More data series are necessary to produce more consistent statements about the actual ‘increase’ or ‘decrease’ of prevalence and visibility, and what it might mean. Moreover, other reasons may concur to this, such as the specific political agendas for each year/season, or the characteristics of the fires and socio-ecologies of the affected regions. We are currently following this thread of enquiry. When animals do appear in the news, they are seldom considered important enough to become the main characters of the story. When they are, their heroic actions are mostly associated to humans – their rescue, or their demise. In this manner, the media perform the invisibility or erasure of animals in multiple layers. Despite the limited coverage of animals in the media, our analysis allowed us to identify four main discursive topoi in which the re-presentation of animals takes significantly different tones: emergency and rescue; emotional relationships and victims; economic relationships; and post-fire consequences and support. The most prevalent are those anchored in a highly emotionally charged discourse, thus reproducing the emotional language of news media, particularly when portraying disasters (Pantti et al., 2012). This is in line with what Figueiredo et al. (2024) found in their study about the media coverage of the wildfires that affected the coastal public-managed forest areas located in the Centre Region of Portugal, in October 2017, their sample showing how the topics of suffering, grief, despair, and sadness were used to mobilise and convey emotions. While this displaying of suffering may offer an entrance point for humans to empathise with other-than-human forms of pain and grief, our results suggest that, paradoxically, it renders animals invisible once again. Contrary to what Due et al. (2014) found about ‘Sam’, the koala, in this case the emotional upheaval around the affected animals talk, mostly, about the humans. This is either because human loss and tragedy are still the core of the concern or, more importantly, because the structural conditions that made both humans and non-humans vulnerable to the tragedy are erased. A striking example is the little dog celebrated for having survived the fire precisely due to the chain to which she was held, which apparently cleared the area around her doghouse from flammable shrubs. The question to be asked is whether such emphasis on (human) emotions risks emptying the political potential of acknowledging animals in themselves, their interests and lives, rather than as merely part of the humans’ sphere of interests, such as property. There are other important dimensions of the re-presentations of animals in the context of disasters that could be explored in future studies, namely how overlapping power relations 19 intersect within these discursive spaces, ensuring the perpetuation of the invisibility of animals. For instance, issues of gender and age of the animals are seldom present in these representations — an erasure on top of another erasure. Furthermore, in this article we did not address how the media agenda in this respect is also built, and repercussed, via social media. This is a highly relevant topic, worth exploring in a dedicated piece, given its very specific features (affordances, uses, and role within the mediascape). Moreover, it was not the focus of this study to address in depth the relationships between the media and the public’s agendas, i.e., the effective impact that such media representations may have in the construction of specific worldviews about animals. This can be done through mixed methods approaches combining thematic analyses of the media and methods to assess possible thematic saliences in the public’s agenda(s) (e.g., focus groups, surveys), to identify possible links. More interestingly, qualitative methodologies researching the processes of meaning-making on the part of the audiences may bring insight into this topic. Finally, our results offer some recommendations for media practitioners, including journalists, as well as NGOs and other civil society associations that may engage with, and make use of, the powers of media communication. Firstly, increase the focus on the structural factors that produce vulnerability, which pre-exist any disaster. This includes the economic relations and tissue that underpin human-animal relationships, and how they concur to turn animals invisible as beings with rights and lives of their own. Secondly, we suggest a responsible use of emotional tones and languages, overcoming the dichotomy emotionality-rationality. Rather than a “a humanistic journalism that combines reason and emotion” (Ward, 2010), we advocate for a more-than-human form of journalism. It may bring empathy and compassion towards all living beings implicated in the tragedy, acknowledging their entangled lives. This, of course, includes humans, but is not exclusive to them. To that purpose, media practitioners should engage in a constant reflexive practice of interrogating who will benefit from their work. This reflexivity will increase the potential of emotional discourses to produce a politically engaged audience, who can consistently bear witness, and encourage action towards animal protection. Finally, we suggest that media practitioners train themselves in asking continuously: where, if at all, are the real animals in my story of the events? What may their perspectives and interests be? How are they attended to, or not? In so doing, media narratives and environmental communication may contribute to untangle the knots of speciesism that cuts across disaster management and communication. REFERENCES Almiron, N., Cole, M., & Freeman, C. P. (2016). Critical animal and media studies: communication for nonhuman animal advocacy. Routledge. Almiron, N., Cole, M., & Freeman, C. P. (2018). Critical animal and media studies: Expanding the understanding of oppression in communication research. European Journal of Communication, 33(4), 367–380. Atchison, J., & Pilkinton, M. (2022). Abject life and disaster: Opportunity and invasive species governance following the 2019–2020 Australian bushfires. Political Geography, 99. https://doi.org/10.1016/j.polgeo.2022.102774 20 Baptistella, E., & Nobre, C. (2024). Sociozoologic Chronicles: Pantanal Creatures’ Narrative Shift from “Demons” to Adored “Pets” Through Media’s Lens. Global Journal of Animal Law, 12(1), 6–19. https://www.scielo.br/scielo.php?script=sci_arttext&pid=S0104Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. CTI (2017). Análise e apuramento dos factos relativos aos incêndios que ocorreram em Pedrógão Grande, Castanheira de Pera, Ansião, Alvaiázare, Figueiró dos Vinhos, Arganil, Góis, Penela, Pampilhosa da Serra, Oleiros e Sertã, entre 17 e 24 de junho de 2017. Comissão Técnica Independente, Assembleia da República. CTI (2018). Avaliação dos incêndios ocorridos entre 14 e 16 de outubro de 2017 em Portugal Continental. Relatório Final. Comissão Técnica Independente, Assembleia da República. Dare, A. M., & Fletcher, C. V. (2019). A Bird’s Eye View of the Malheur Wildlife Refuge Occupation: Nonhuman Agency and Entangled Species. Environmental Communication, 13(3), 412–423. Due, C., Thomson, K., & Every, D. (2014). Representations of Sam the Koala in the Australian Mainstream News Media. Media International Australia, 51(1). https://doi.org/10.1177/1329878X1415100107 Fishman, M., & Cavender, G. (1998). Entertaining Crime: Television Reality Programs. Adline de Grutyter. Freeman, C. P. (2009). This Little Piggy Went to Press: The American News Media’s Construction of Animals in Agriculture. The Communication Review, 12(1), 78–103. Gans, H. (1979). Deciding What's News: A Study of CBS Evening News, NBC Nightly News, News, Newsweek and Time. Pantheon Books. Gill, T., & Aldrich, D. P. (2014). Building Resilience: Social Capital in Post-Disaster Recovery. Social Science Japan Journal, 17(1), 118–122. Golding, P., & Elliot, P. (1979). Making the news. Longman. Golding, P., & Murdock, G. (1997). The political economy of the media. Edward Elgar Publishing. Gessa, S. J., Tayeebwa, W., Muwanika, V., & Rothman, J. M. (2024). Conflict or Harmony: Framing of Wildlife News in a Biodiversity Hotspot. Journalism and Media, 5(1), 1– 13. Hall, S. (1973). Encoding and Decoding in Television Discourse, CCCS selected working papers. Vol. 2 / edited by Ann Gray et al (Abingdon, 2007) pp. 386-398, available at http://epapers.bham.ac.uk/2962. Haraway, D. (2008). When Species Meet. Minnesota Press. 21 Irvine, L. (n.d.). Why Study Animals in Disasters? https://doi.org/10.1515/jhsem-2016 Hoffman, S., & Oliver-Smith, A. (2012). Catastrophe and Culture: The Anthropology of Disaster. School for Advanced Research Press. Hoggarth, R. (1998 [1957]). The Uses of Literacy. Routledge. Irvine, L. (2009). Filling the Ark: Animal Welfare in Disasters. Temple University Press. Jacobson, S. K., Monroe, M. C., & Marynowski, S. (2001). Fire at the Wildland Interface: The Influence of Experience and Mass Media on Public Knowledge, Attitudes. Source: Wildlife Society Bulletin, 29(3), 929–937. Joffe, H. (2011). Thematic analysis. In D. Harper, & A. R. Thompson (Eds.), Qualitative research methods in mental health and psychotherapy: A guide for students and practitioners (pp. 209-223). John Wiley & Sons, Ltd. Kelman, I. (2020). Disaster by Choice. Oxford University Press. https://global.oup.com/academic/product/disaster-by-choice9780198841340?cc=pt&lang=en& Kelman, I. (2024). Choosing or avoiding disaster. Geography, 109(1), 36–39. Kempton, S. D. (2019). Racialized Reporting: Newspaper Coverage of Hurricane Harvey vs. Hurricane Maria. Environmental Communication, 14(3), 403–415. Krippendorff, K. (2004). Content Analysis: An Introduction to Its Methodology. Sage Publications, Inc. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people's choice. Duell, Sloan & Pearce. Madeira, P. M., Mourato, J. & Soares, F. (2023). ‘Florestas’. In L. Schmidt (Ed.), 50 anos de Políticas Ambientais em Portugal: da Conferência de Estocolmo à actualidade (pp. 237–264). Edições Afrontamento. MAFDR (2017). Resposta à pergunta n.º 147/XIII/3ª de 23-10-2017, “Ausência de apoio aos animais vítimas dos trágicos incêndios”. Available at https://www.parlamento.pt/ActividadeParlamentar/Paginas/DetalhePerguntaRequerim ento.aspx?BID=102959 MAFDR (2018). Resposta à pergunta n.º 556/XIII/3ª de 18-12-2017, “Ausência de apoio aos animais vítimas dos trágicos incêndios”. https://www.parlamento.pt/ActividadeParlamentar/Paginas/DetalhePerguntaRequerim ento.aspx?BID=103402 Mccombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176–187. Molloy, C. (2011). Popular media and animals. Springer. OberCom-Observatório da Comunicação. (2024). Anuário da Comunicação 2023. www.obercom.pt 22 Oliveira, S., Zêzere, J. L., Queirós, M., & Pereira, J. M. (2017). Assessing the social context of wildfire-affected areas. The case of mainland Portugal. Applied Geography, 88, 104– 117. https://doi.org/https://doi.org/10.1016/j.apgeog.2017.09.004 Pantti, M., Wahl-Jorgensen, K., & Cottle, S. (2012). Disasters and the Media. Peter Lang. https://www.peterlang.com/view/title/21344 Pezzullo, P. C. (2024). On Environmental Communication as a Care Discipline. Environmental Communication, 18(1–2): 1–7. Ru, S., Parzniewski, S., Breen, K., Heyland, L., Farhat, L., & Wu, H. (2025). Voice for the Voiceless: Amplifying Animal Issues in Disaster Management and Media. Natural Hazards Review, 26(1). https://doi.org/10.1061/NHREFO.NHENG-2203 San-Miguel-Ayanz, J., Durrant, T., Boca, R., Maianti, P., Libertá, G., Artés-Vivancos, T., Oom, D., Branco, A., de Rigo, D., Ferrari, D., Pfeiffer, H., Grecchi, R., Onida, M., & Löffler, P. (2022). Forest Fires in Europe, Middle East and North Africa 2021. Publications Office of the European Union. San-Miguel-Ayanz, J., Oom, D., Artès, T., Viegas, D.X., Fernandes, P., Faivre, N., Freire, S., Moore, P., Rego, F., & Castellnou, M. (2020). “Forest fires in Portugal in 2017”. In: A. Casajus Valles, M. Marin Ferrer, K. Poljanšek, & I. Clark (Eds.), Science for Disaster Risk Management 2020: Acting today, protecting tomorrow (pp. 414-430). Publications Office of the European Union. Seo, M. (2021). Amplifying Panic and Facilitating Prevention: Multifaceted Effects of Traditional and Social Media Use During the 2015 MERS Crisis in South Korea. Journalism and Mass Communication Quarterly, 98(1), 221–240. https://doi.org/10.1177/1077699019857693 Shaw, E. (1979). Agenda-Setting and Mass Communication Theory. Gazette (International Journal for Mass Communication Studies), XXV(2), 96-105. Silva, N. T. C. da, Fra.Paleo, U., & Ferreira Neto, J. A. (2019). Conflicting Discourses on Wildfire Risk and the Role of Local Media in the Amazonian and Temperate Forests. International Journal of Disaster Risk Science, 10(4), 529–543. Smit, I. P. J., Joubert, M., Smith, K., van Wilgen, N., Strydom, T., Baard, J., & Herbst, M. (2022). Fire as friend or foe: the role of scientists in balancing media coverage of fires in National Parks. African Journal of Range and Forage Science, 39(1), 136–147. Stalenberg, E., Lunney, D., & Moon, C. (2024). ‘It’s dangerous to put a number on them’. Media coverage of koalas during the 2019–2020 ‘Black Summer’ bushfires in Australia. Pacific Conservation Biology, 30(6). https://doi.org/10.1071/PC24019 Tierney, K. (2019). Disasters: A Sociological Approach. John Wiley & Sons. Viegas, D. X., Almeida, M. F., & Ribeiro, L. M. (2017). O complexo de incêndios de Pedrógão Grande e concelhos limítrofes, iniciado a 17 de Junho de 2017. Centro de Estudos sobre Incêndios Florestais, Universidade de Coimbra. Viegas, D. X., Almeida, M. F., & Ribeiro, L. M. (2019). Análise dos incêndios florestais ocorridos a 15 de outubro de 2017. Centro de Estudos sobre Incêndios Florestais, Universidade de Coimbra. 23 Wachinger, G., Renn, O., Begg, C., & Kuhlicke, C. (2013). The risk perception paradox— implications for governance and communication of natural hazards. Risk analysis, 33(6), 1049-1065. Walker, H. M., Reed, M. G., & Fletcher, A. J. (2020). Wildfire in the news media: An intersectional critical frame analysis. Geoforum, 114, 128-137. Willig, C. (2003). Discourse analysis. Qualitative psychology, 159-183. Wu, H., Heyland, L. K., Yung, M., & Schneider, M. (2023). Human–Animal Interactions in Disaster Settings: A Systematic Review. International Journal of Disaster Risk Science, 14(3), 369–381. https://doi.org/10.1007/s13753-023-00496-9 Funding: The research leading to the present article was funded by two main entities, in two different phases. The first phase was funded by the Portuguese Foundation for Science and Technology (FCT), through an individual research contract granted to the first author for the project Liminal Becomings: reframing human-animal relationships in disasters [CEECIND/02719/2017]. The second phase of the research was funded by the European Union through a Consolidator Grant granted to the first author for project ABIDE – Animal Abidings: recovering from disasters in more-than-human communities (ERC, ABIDE, CoG ID 101043231). Views and opinions expressed are those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them. Acknowledgements: The authors wish to thank Sara Silva, junior researcher of project ABIDE, for her diligent help with the final consolidation of the database, processing of descriptive statistics, and reviewing and formatting the bibliography. The authors also wish to express their gratitude to Isabel Férin Cunha, scientific advisor of ABIDE, for her thorough reading and critical review of the first version of this article. Authors’ contributions: Verónica Policarpo: coordination, research design, methodological and theoretical framework, statistical analysis (univariate, bivariate and multivariate), writing, reviewing. Filipa Soares: data collection, thematic coding, writing, reviewing. Xinyue Zhang, data collection, thematic coding, writing, reviewing. Leonor Pereira da Costa: data collection, thematic coding, reviewing. 24 i We choose to use the expression “animals” for the sake of simplicity. However, this does not mean we endorse the human-animal divide that would separate “animals” from “humans”. Rather, we acknowledge the animal condition of humans, just as the human condition of other species, insofar they are beings with interests and lives of their own.