Photojournalist Framing in the Ecological Crisis: The DANA Flood Coverage

Abstract

This research focuses on current photojournalism practices in the media representation of the climate crisis, exploring how images published by Spain’s leading digital newspapers contribute to shaping public perception of the phenomenon. Through a mixed-methods approach, this study examines eco-visual reporting in 2023—a year characterized by unprecedented global temperatures—alongside an examination of the initial month of coverage of the 2024 DANA floods. The main objective of this study is to evaluate whether the visual depiction of climate change in mainstream media aligns with or deviates from the established visual patterns associated with stereotyped iconography and alarmist narratives. This research provides a taxonomy of the predominant thematic elements and their frequency, addressing key issues such as the persistence of worn-out visual tropes, the reliance on stock images, and the episodic nature of its news treatment. The inclusion of the most catastrophic climate disaster of the 21st century in Spain allows for an exploration of how the media pressure inherent to crisis situations intensifies the risk of errors in image attribution and contextualization. Finally, we offer practical recommendations to guide visual coverage toward comprehensive, accurate, and human-centered approaches, aiming to foster public trust and maximize the potential of photojournalism as a key tool in collective mobilization against the climate emergency.

1. Introduction

Climate change (CC) has established itself as one of the most pressing challenges of our time, with impacts that go beyond the environmental sphere and affect global economic, social, and political stability. According to the World Economic Forum’s Global Risks Report 2025, extreme weather events, biodiversity loss, and natural resource scarcity top the ranking of the most urgent threats for the coming decade—surpassing even concerns such as risks associated with artificial intelligence, cyberattacks, or social polarization (WEF, 2025).

In a context where it is increasingly evident that climate policy in democratic societies will not progress without public support and participation, the previous literature has stressed the importance of implementing communication strategies capable of promoting mobilization and collective action (Ockwell et al., 2009). While social media platforms have emerged as key communication channels, traditional media still play a crucial role as intermediaries between science and the public, acting “as a non-specialist source of information that citizens can use to make their political and lifestyle choices” (Nurmis, 2017, p. 2). Consequently, they remain fundamental in drawing public attention to CC and shaping audience perceptions of the phenomenon (Uyheng et al., 2021; Xu et al., 2023). Nonetheless, the impersonal transmission of scientific information has been proven insufficient in prompting substantial behavioral modification, perhaps resulting in indifference and dismissal (Deryugina & Shurchkov, 2016). As Lorenzoni et al. (2007) argue, simply understanding environmental issues is not enough to elicit a positive response to mitigation and adaptation measures; rather, it is essential to establish an emotional connection with the public that makes these facts relevant to their daily lives. In this regard, photojournalism is considered capable of fostering such a connection, as “visual images come to the forefront as ways of communicating that can involve all three dimensions of engagement: thinking, feeling, and doing” (Nurmis, 2017, p. 31; see also Baberini et al., 2015; Zelizer, 2004). However, news images also operate from a symbolic sphere, as markers for venues, social collectives, and organizations, and even more, to provide familiar visual motifs of staged pseudo-events—speeches, press conferences, government or international meetings, etc., influenced by the presence of news reporters (Griffin, 2012)—instead of promoting ecomedia literacy and civic engagement. If an image fails to reflect significant issues, it may only signify an isolated incident of marginal importance (Griffin, 2010).

The visual representation of CC faces significant challenges that compromise its effectiveness and call for a reassessment of the current state of the field. First, the visual representation of the phenomenon has thus far been dominated by a limited set of archetypal images—receding glaciers, smoking chimneys, polar bears, political figures (S. J. O’Neill, 2017)—that, while initially impactful, have led to audience fatigue and desensitization over time (Chapman et al., 2016; Corner et al., 2015). The increasing reliance on stock photographs, which emphasize esthetic appeal rather than informational content (Aiello et al., 2022), has led to a visual homogenization characterized by the repetition and convergence of similar images, thereby diminishing the quality and diversity of visual information regarding the phenomenon (Wang et al., 2018).

Moreover, the dynamics of representation become even more complex in crisis situations, as the demand for immediacy, the difficulties on the ground, and the inherent uncertainty of these events disrupt journalistic routines (Pellegrini et al., 2015). Improvisation in the early hours, imprecise graphic treatment, the repetitive use of images focused on devastation and human suffering (Brusi et al., 2008; Meneses-Fernández & Torvisco, 2024), and the growing adoption of User-Generated Content (UGC) (Fernández-Castrillo, 2014) due to limited access to official sources increase the risks of misinformation and the spread of “fast news” (Fernández-Castrillo & Ramos, 2023; Sacco & Bossio, 2015). Recent examples, such as the dissemination of decontextualized images during the 2019 Queensland floods, illustrate how these dynamics further complicate the public’s ability to distinguish fact from fiction, underscoring the need to ensure accuracy and transparency in visual coverage (Thomson et al., 2020).

In light of these challenges and given that much of the research on the visual dimension of the climate crisis remains poorly synthesized—particularly in the context of digital media (Wang et al., 2018)—this study examines the Spanish case by analyzing images published in the digital editions of the country’s leading national newspapers—El País, El Mundo, and La Vanguardia. The timeframe of the research is focused on two critical periods: (1) the general coverage of CC throughout 2023—which the World Meteorological Organization (WMO) declared the hottest year on record (WMO, 2023), overcome by 2024, when the temperatures rise about 1.55° above pre-industrial levels (WMO, 2025)—and (2) the first month of coverage of the flooding caused by DANA (Spanish acronym for isolated high-level depression) in Valencia in 2024, considered the most severe disaster in Spain so far this century. This study offers a unique contribution by examining the ethical and narrative challenges of climate photojournalism in both general and emergency contexts, highlighting important implications for the visual treatment of CC in international media. To this end, we analyze how predominant visual frames either reinforce or question traditional patterns and how the inherent challenges of crisis situations affect the quality, accuracy, and credibility of coverage, with particular attention to the use of UGC and the risks associated with visual misinformation, such as the decontextualization of photographs.

2. Fostering Civic Engagement Through Visual Information

Given the complexity of constructing a compelling narrative on CC, the current research suggests that the use of photojournalism as a medium capable of fostering both affective and behavioral connections to the issue may be essential to mobilize the citizenship, due to the crucial role of affective components in risk perception (Muñoz-Pico & León Anguiano, 2024; Nurmis, 2021; S. O’Neill, 2020). This power of emotional evocation is closely linked to the ability of images to activate the experiential behavioral system in the human brain, a cognitive mechanism that, unlike analytical processing—which is more logical and deliberative—operates by “relating current situations to memories of one’s own or others’ experience” (Marx et al., 2007, p. 48). Shome et al. (2009) emphasize that appealing to this system is key when reporting on CC, as it is the most effective option in motivating action. It is well established that the experiential system is activated more rapidly by visual stimuli than by textual information. This is because visual information is processed almost instantaneously, allowing images to influence behavior before the rational part of the brain has time to intervene (Williams, 2005). Consequently, photographs can bypass the mechanisms of the analytical system and affect audiences first on an emotional level (Todd & Anderson, 2013).

Nonetheless, the effectiveness of images in prompting action depends largely on the visual framing adopted, which is, at its core, a highly ideological structuring process that shapes specific interpretations by emphasizing certain elements of reality while omitting others (Entman, 1993; Guenther et al., 2024). This narrative construction, as Sánchez Montalbán (2018) points out, is based on the very codes of photography, where each image, beyond being a mere specular reflection of reality, functions as an iconological proposal, loaded with meaning and subjectivity. This process implies that the viewers establish a relationship with the image based on how the visual message has been constructed, and this perception can differ significantly from what they would have experienced by directly witnessing the photographed reality. As O’Neill states, CC imagery and civic engagement are shaped by the “cultural politics of climate change” in such a way that “(…) If particular frames become dominant and unchallenged, they can limit the participation of particular actors, affect the amount of media coverage, influence public opinion, and ultimately even impact regulatory options” (S. O’Neill, 2020, pp. 11–12).

3. Effective Visual Communication on Climate Change

Although recent studies point to a progressive shift in visual framing towards more human and local perspectives, both in mainstream media (García Herrero & Vicente Torrico, 2021; Lopes & Azevedo, 2023) and social networks (Ardèvol et al., 2021; S. J. O’Neill & Graham, 2016), the overall trend continues to rely on a hackneyed iconography that reinforces a distant and abstract perception of the problem. The overuse of stereotyped images has contributed to a “climate fatigue” (Wang et al., 2018) that desensitizes audiences and reinforces emotional and intellectual disconnection from the issue (Nurmis, 2017; Chapman et al., 2016). Moreover, social media has also sustained the circulation of these “stalled images” through the relentless processes of reposting and sharing (Rusgrove & Catanzaro, 2020).

There is substantial evidence that fear-based visual appeals and images that emphasize negative aspects are generally ineffective in fostering personal engagement or encouraging reflective thinking on individual contributions (Muñoz-Pico & León Anguiano, 2024; Metag et al., 2016; S. O’Neill & Nicholson-Cole, 2009). Moeller (2006) cautions that capturing attention is merely the first step in visual communication and that, without a strategy to channel it into action, the impact of such images dissipates rapidly. What happens in CC coverage is that often only this first step is fulfilled; in short, what we find is a redundant and stalled iconography that has failed to evolve alongside the broader discourse on climate crisis and has gradually lost its ability to mobilize the public with the same intensity as before. As a result, there has been a growing detachment between audiences and CC imagery, which is often perceived as too distant, overly abstract, or emotionally disconnected from everyday life (Manzo, 2010).

The increasing use of stock photography, especially in digital media, is one of the main elements exacerbating this visual homogenization. News organizations currently use pre-packaged, generic, and inexpensive images more frequently to illustrate their articles thanks to platforms like Getty Images and iStockphoto (Gürsel, 2016). Although useful in terms of accessibility, these photos lack the documentary and contextual depth needed to properly inform viewers about the climate crisis, making photography only a mere decorative accompaniment to text (Aiello et al., 2022). In this context, Vobič and Tomanić Trivundža (2015) introduced the concept of the “tyranny of the empty frame” to illustrate how, due to the demands of rapid news production, journalists increasingly utilize symbolic photographs that lack a direct relationship to the events they cover. This leads to a transition from indexical–iconic visual communication to a more metaphorical visual language, hence diminishing the informational depth and documentary importance of photojournalism. Furthermore, in some cases, these images are repurposed to accompany unrelated news stories (S. J. O’Neill & Smith, 2013), further contributing to the decontextualization of information. Hansen and Machin (2008) identified this trend in the “Getty Green Images” collection, which tended to depict CC through abstract and disconnected representations.

Leiserowitz (2006) concluded that beyond factual knowledge, what lacks in the public’s mind are “vivid, concrete, and personally relevant affective images of climate change” (p. 55). Similarly, Doherty (2014) and Doherty and Webler (2016) found that “descriptive social norms” and “perceptions of self-efficacy” are predictors of engagement. This finding suggests that seeing others taking action increases an individual’s willingness to get involved, and that the stronger the belief that their actions can make a difference, the more likely they are to participate. Along these lines, several studies have shown that frames that emphasize solutions, as well as images depicting “real” and identifiable people in everyday contexts—rather than the usual politicians, activists, or celebrities—generate higher levels of motivation and behavioral intentions (Dahmen et al., 2021; Eskjær, 2017; Hart & Feldman, 2016). The Climate Visuals project, led by Climate Outreach (n.d.; Corner et al., 2015), incorporates these aspects into its recommendations for effective visual communication on CC. Its core principles include (1) depicting “real people” rather than staged photo-ops; (2) telling new stories through thought-provoking and unfamiliar imagery; (3) illustrating climate causes at scale rather than at the individual level; (4) harnessing the emotional power of climate impacts while pairing them with actionable solutions; (5) balancing local but serious climate impacts to ensure relevance without trivialization; (6) being cautious with protest imagery as it resonates primarily with activists; and (7) tailoring visual communication to the audience, recognizing that responses to climate imagery vary based on concern, skepticism, and political affiliation.

The declining role of photojournalism as a narrative tool and its relegation to a merely illustrative function represents a missed opportunity to strengthen climate communication. Following Midberry and Dahmen (2019), we argue that only when photojournalism is recognized as an autonomous source of information—rather than a mere visual supplement to written text—will the media be able to fully harness the affective and connotative power of images in the fight against CC.

4. Photojournalist Framing in Ecological Emergencies

In times of crisis, photojournalism assumes a critical role as a medium capable of shaping public perception and influencing social responses to collective trauma, particularly from a local perspective (Buoncompagni, 2025). During such events—which range from natural disasters to conflicts or humanitarian crises—the social function of journalism reaches its highest level of significance, “as the public’s need for timely, truthful, and contextualized information is heightened” (Pellegrini et al., 2015, p. 250). According to Barbie Zelizer and Allan (2002/2011), in these circumstances, journalists do more than just report the news; they become active agents in helping communities transition from trauma to recovery. Here, images—with their unique capacity to “bring the issue home” (Anne DiFrancesco & Young, 2011) and fix collective memory (Zelizer, 2004)—are especially useful, as these are moments in which people need to see what is happening (Mortensen et al., 2017; Vevea et al., 2011).

However, the visual coverage of crises has traditionally prioritized spectacle and dramatization over solution-oriented perspectives (Monahan & Ettinger, 2018). This is a tendency that Holt and Carnahan (2020) link to an audience-building strategy that capitalizes on the media profitability of “the negative”. Cottle (2009) warns that there is a risk of the public becoming mere spectators of the imminent catastrophe if communicators dwell on spectacular, even sublime, images of what we are going to lose or what has already been lost. This type of visual representation may serve as an educational function by providing indirect experiences that help audiences imagine risk scenarios and possible responses (Andersen, 2015), but they can also evoke feelings of helplessness, fear, and anxiety among both viewers and affected communities (S. O’Neill & Nicholson-Cole, 2009; S. J. O’Neill, 2013). In this regard, Meneses-Fernández and Torvisco (2024) found a negative correlation between the overuse of a “dramatic voyeurism” during the media coverage of La Palma volcanic eruption (Canary Islands, Spain). Along with these findings, the Official College of Psychology of Santa Cruz de Tenerife publicly urged the media to limit the dissemination of images and videos of collapsing homes due to their emotional impact (COPTFE, 2021).

The tension between emotional impact and journalistic rigor is further amplified in a media environment where visual narratives are increasingly vulnerable to (dis)misinformation. In the early moments of an emergency, the high demand for information and scarcity of clear data create the perfect conditions for the rapid spread of falsehoods and misleading visual content that complicate journalists’ tasks and sow confusion among audiences (Social Media Working Group, 2018; Tran et al., 2020). Whether deliberately or by mistake, the viralization of visual information through social media without verifying its authenticity or context generates and intensifies this problem (Thomson et al., 2020). Additionally, the increasingly common trend of visual content in breaking news coverage coming from non-professionals at the scene of events/on the ground (Andén-Papadopoulos, 2014), adds another layer of complexity. Even though this content can provide alternative narratives, it also raises critical concerns regarding “anonymity, reliability, and intelligibility” (Mast & Hanegreefs, 2015, p. 597). The pressure to publish rapidly, noted by Fernández-Castrillo and Ramos (2024), combined with the difficulty of verifying these materials, sometimes leads media outlets to prioritize speed over accuracy, thereby amplifying the dissemination of misleading and decontextualized images (Thomson et al., 2020).

To answer these challenges, Midberry and Dahmen (2019, p. 6) propose that visual coverage should strive to be “comprehensive, precise, and humanizing”. This means including images that depict both problems and solutions while avoiding alarmist or reductionist narratives (comprehensive), ensuring that photographs are original and contextually relevant to prevent the spread of misinformation (precise), and portraying people respectfully, emphasizing their dignity and fostering empathy rather than mere spectacle (humanizing).

Given that the visual framing adopted in news coverage directly influences how people think and respond to events (Messaris & Abraham, 2001), it is crucial to assess how the media performs under high-complexity scenarios that challenge its social function. The visual coverage of the DANA floods serves as a paradigmatic case study for analyzing these dynamics, allowing us to evaluate the extent to which photojournalistic practices met the need for accurate and contextualized information or, conversely, contributed to the distortion of public discourse at a critical moment instead of providing specific oriented solutions.

5. Materials and Methods

5.1. Objectives and Research Questions

This study analyzes the photographic representation of CC and the 2024 DANA floods in Spain’s leading general-interest newspapers—El País, El Mundo and La Vanguardia. Grounded in research on the media construction of events, this research draws on the framing perspective (Entman, 1993), which posits that the media influence public perception by selecting, emphasizing, or omitting certain elements of reality (Scheufele & Tewksbury, 2007). Consequently, this study operates on the premise that the visual frames adopted in the coverage of these events can shape public understanding and influence their impact on public opinion by either promoting or limiting certain interpretations or responses.

The main objective is to examine whether visual coverage reinforces established visual patterns associated with CC and climate-related disasters—typically characterized by hackneyed climate iconography and predominantly alarmist narratives—or whether it adopts approaches more aligned with the principles of solutions photojournalism. To this end, this study is structured in three phases: (1) content analysis of published images to identify their sources, dominant themes, and framing choices; (2) alignment assessment of the visual coverage with the principles of solutions photojournalism, focusing on how images highlight solutions and community responses versus alarmism and catastrophism; and (3) identification of ethical issues in visual coverage, including image decontextualization, misattribution, and other factors that may undermine the credibility of coverage in these contexts.

The research questions are as follows:

RQ1.

Do newspapers rely more on wire service and stock photos or on original images taken by staff photographers and regular contributors?

RQ2.

What ethical issues, such as lack of attribution or misattribution, arise in visual coverage, and how do they relate to the use of stock images, UGC, and professional photography?

RQ3.

What themes dominate the visual dimension of CC and DANA, and how do these images challenge or reinforce established visual patterns?

RQ4.

To what extent do the photographs in the selected articles reflect the problem, the solution, or both?

Finally, this study offers practical recommendations aimed at steering visual coverage toward comprehensive, accurate, and human-centered approaches to foster public trust and maximize the potential of photojournalism as a key tool for collective mobilization in the face of the climate crisis.

5.2. Sample and Design

The sample corpus consists of a total of 11,583 articles with images related to CC and the DANA floods of 2024, published by El País, El Mundo, and La Vanguardia. The units of analysis are divided into two periods: (1) 8021 articles on CC published between 1 January and 31 December 2023 and (2) 3562 articles on the DANA published between 29 October and 29 November 2024.

To select these outlets, we relied on data from the Marco General de Medios en España (AIMC, 2025) and the Digital News Report 2024 (Newman et al., 2024), which rank them as the most widely read general-interest newspapers in Spain to date. Furthermore, their inclusion provides a comprehensive perspective on the Spanish media landscape concerning the events under analysis, covering different ideological tendencies, with more progressive (El País), liberal (El Mundo), and centrist (La Vanguardia) outlets.

We adopted content analysis as our primary methodology because it allows us to identify publication patterns, narrative trends, and emerging journalistic practices (Rose, 2012). To achieve the research objectives, we designed a sequential two-phase study: (1) a content analysis focused on the denotative meaning of images to provide a quantitative overview of image types, themes, and frequencies; and (2) a framing and connotative meaning analysis to explore the broader qualitative implications of visual coverage.

5.3. Data Acquisition and Processing

We used the MyNews database to compile the sample, collecting all articles published by El País, El Mundo, and La Vanguardia during the defined periods that contained at least one image related to CC or the DANA floods. A search strategy was applied using key words combined with the Boolean operator “OR”, including for CC the terms “climate change”, “global warming”, “climate crisis”, “climate emergency”, “greenhouse gases”; and for DANA the terms “DANA” and “cold drop”. As a result, we obtained 12,360 articles with images (8613 from the CC and 3747 from the DANA) that contained at least one of the aforementioned terms in the headline, caption, or body of the article. After applying a filtering process to remove duplicates and invalid URLs and ensure thematic relevance, the final corpus was reduced to 11,583 articles (

Table 1. Media subscribers, ownership, and sample corpus.

Media Outlet	Subscribers	Ownership	Articles CC 2023	Articles DANA 2024
El País	350,000	Grupo Prisa	3947	928
El Mundo	115,000	Unidad Editorial	1239	920
La Vanguardia	77,000	Grupo Godó	2835	1714

Source: Author’s edit.

). Finally, in line with studies applying probabilistic sampling methods for content analysis in the media (Kim & Cooke, 2018), two manageable representative samples—with a confidence level of 95% and a margin of error of 5%—were selected: 606 articles for the 2023 coverage and 363 articles for the DANA event. The selection was conducted through simple random sampling using Random.org.

5.3.1. Content Analysis and Denotative Meaning

This phase of the analysis focuses on the denotative level of meaning, which allows for a clearer understanding or “background map” (Bell, 2004) of the images that receive the most attention and those that are absent from the media. A content analysis was carried out on the photographs associated with the news about the CC of 2023 and the DANA floods published by the three newspapers. The variables used include the name of the media outlet, the subject of the image, the date it was taken, and its authorship. In addition, reverse image searches were conducted using tools such as TinEye and Google Lens to verify provenance and assess potential misattribution and decontextualization.

The visual representation of CC in the media has been studied through various frames, such as impacts, causes, solutions, adaptation, nature, people, and protests, among others (Rebich-Hespanha & Rice, 2016). However, Culloty et al. (2018) point out that many of these frames often overlap within the same analysis and are not mutually exclusive, making it challenging to classify and interpret coherent results. To address this overlap, we adopted their thematic coding model, which assigns categories to each image based on the visual dominance of the depicted object in terms of camera focus and positioning. For example, if an image shows political figures at a climate summit and the focus is a close-up that emphasizes the individuals over the action, it is classified under the category “people” with the subcategory “political figures”. On the other hand, if it is a general shot that emphasizes the action over the people, it is classified as “topical events” with the subcategory “summits-political meetings”.

The categories used were mutually exclusive; that is, each image was assigned a single category based on its dominant theme. In cases of ambiguity, the immediate textual context—such as the caption and headline—was analyzed to determine the most appropriate category. Finally, descriptive and inferential statistical techniques applied in IBM SPSS Statistics (SPSS sin AMOS-V.26) were used to identify patterns and significant differences among the studied elements. A non-parametric analysis was carried out using contingency tables with chi-square tests. The coding scheme of the main significant variables used in this phase of the study is as follows:

• Media outlet: El País, El Mundo, La Vanguardia.
• Date: dd/mm/yyyy
• Image source: staff photographer; freelance, wire service; stock photo agency; official source—e.g., institutional communication services; corporate entities and ONGs—e.g., Iberdrola, Greenpeace; UCG; uncredited.
• Theme: topical events; agriculture and food; iconography; landscapes; people; animal kingdom; technology/energy; and others.
• Location: national, international.
• Attribution: complete—the distributor and the individual author are identified; partial—only the distributing source is identified; unattributed.

5.3.2. Framing Analysis and Connotative Meaning

To determine visual framing, the images were coded along with their captions and the text of the news article, as their meaning is partially derived from the accompanying textual information (Caple, 2013). The analysis followed the conceptual framework of solutions photojournalism proposed by (Midberry & Dahmen, 2019; Midberry et al., 2024) by assessing whether the images highlighted the problem, the solution, or a combination of both. Under the “problem-only” condition, we coded photographs depicting the harmful effects of DANA or CC without showing active responses—e.g., images of destruction, environmental loss, human suffering, or political conflict. Images depicting responses to solve the problem were classified as “solution-only” images—e.g., emergency and relief services, educational initiatives, community cooperation. The “problem–solution” frame was used for those that combined both elements within the same visual composition—e.g., images of volunteers removing debris from scenes of destruction. In addition, the category “other” was added for those images that could not be clearly classified into any of the options—e.g., images of institutional buildings, money, etc. Finally, we enhanced the analysis by assessing the coherence between the images and the article’s text, examining whether the photographs reinforced, contradicted, or had no relationship to the narrative of the text.

6. Results

6.1. Content Analysis

The first research question examined whether news articles use more original photographs taken by staff photographers than images from news services and stock sites. Wire service photographs were widely used in both contexts, accounting for 31.5% in CC 2023 and 52.3% in DANA, suggesting a greater reliance on rapidly distributed external sources, particularly during crisis situations. Images taken by staff photographers were less frequent, comprising 18.8% and 17.1%, respectively. Consistent with previous research (León & Erviti, 2015), there was a recurrent use of the same agencies—EFE led with 198 appearances, followed by Europa Press (45), Reuters (37), and Agence France Presse (37)—as well as the same photographs. In the case of DANA, 17.4% (63 out of 363) of the images were repeated across the analyzed media outlets.

Images from corporate and stock sources, mainly associated with promotional content in CC 2023 coverage (14.7% and 13%, respectively), decreased in response to the immediate informational needs of DANA (6.1% and 4.7%, respectively;

Table 2. Image source distribution in CC 2023 and DANA floods.

Source Type	CC 2023		DANA Floods
	n	%	n	%
UGC	30	5	21	5.8
Corporate/ONG	89	14.7	22	6.1
Freelance	59	9.7	8	2.2
Official source	23	3.8	20	5.5
Staff photographer	114	18.8	62	17.1
Stock/Photo agency	79	13	17	4.7
Wire service	191	31.5	190	52.3
N/A	21	3.5	23	6.3
Total	606	100	363	100

Note: There is a statistically significant difference in the distribution of image sources between CC 2023 and DANA floods (χ²(7) = 80.70, p < 0.001) (Table A1). To determine the source of the images, reverse searches were performed on the entire sample so that those classified as N/A correspond to unattributed images whose origin could not be verified. Source: Authors’ edit.

). Meanwhile, the presence of UGC was similar in both contexts, at 4.9% in CC 2023 and 5.8% in DANA.

Regarding the second research question, which explored ethical issues related to attribution depending on the image source, this study found no statistically significant difference in the quality of image attribution between the two events (χ²(3) = 4.76, p = 0.19). A total of 9.4% of DANA images and 13.5% of CC 2023 images lacked proper credit, while 12.1% and 13.7%, respectively, only mentioned the distributing source without citing the individual photographer. Cases of misattribution were minimal, with 1.1% in DANA and 1.2% in CC 2023, though incomplete attribution accounted for 22.6% and 28.4% of the images, respectively (

Table A2. Image attribution in CC 2023 and DANA floods.

		CC 2023		DANA Floods
		n	%	n	%
Attribution	Complete	434	71.6	281	77.4
	Partial	83	13.7	44	12.1
	Misattributed	7	1.2	4	1.1
	Unattributed	82	13.5	34	9.4
	Total	606	100	363	100

Note: There is no statistically significant difference in the distribution of image attribution between CC 2023 and DANA floods (χ²(3) = 4.76, p = 0.19, Cramér’s V = 0.07).

).

Additionally, 53.4% of the analyzed captions in CC 2023 and 64.2% in DANA did not include the date of capture. Reverse image searches to determine their temporality revealed that 30.2% (n = 183) of those of CC 2023 were archival, indicating a partial disconnection from the events depicted. In the case of DANA, this percentage was 12.9% (n = 47), consisting primarily of images of political figures taken before the floods.

The analysis of the relationship between incomplete attribution and image source type showed statistically significant differences (χ²(6) = 139.81, p < 0.001). Stock photos had the highest percentage of incomplete attributions (64.6%), followed by corporate sources (34.2%) and UGC (27.5%), while original staff-captured photos (8%) and freelancers (9%) had the lowest rates, suggesting that visual content further removed from journalistic standards is more likely to be used without proper attribution (

Table A3. Image attribution by source type for CC 2023 and DANA floods.

		Attribution
		Incomplete	Complete	Total	% Incomplete
Source type	Stock/Photo agency	62	34	96	64.6
	Corporate/ONG	38	73	111	34.2
	UGC	14	37	51	27.5
	Official source	9	34	43	20.9
	Wire service	67	314	381	17.6
	Freelance	6	61	67	9.0
	Staff photographer	14	162	176	8.0
	Total	210	715	925 1

Note: The “incomplete” category includes misattributed, unattributed, or partially attributed images (where only the distributing agency or organization is cited). ¹ The N/A category (44 images) was excluded as their origin could not be verified. The results of the Chi² test show a significant association between attribution and source type (χ²(6) = 139.81, p < 0.001, Cramér’s V = 0.39).

).

Finally, this study evaluated whether the distribution of images within each article served a more illustrative or narrative function, i.e., whether they appeared as standalone images or were accompanied by other visuals that constructed a visual narrative. In CC 2023, 34.7% of the analyzed articles featured only one image, compared to 70.5% in DANA, suggesting that the pressure to publish during high-demand situations limits the use of composed visual narratives.

6.2. Thematic Hierarchy

The third research question asked if the visual coverage surpasses or reinforces traditional visual patterns—centered on stagnant climate iconography in CC and predominantly catastrophic narratives in crisis situations.

In the case of CC 2023, images were classified into eight main categories (Figure 1). The most frequent was “people” (38.4%,

Table A4. Thematic hierarchy of the images in CC 2023.

		Frequency	%
Category	People	233	38.4
	Topical events	125	20.6
	Technology/energy	101	16.7
	Landscapes	45	7.4
	Agriculture and food	35	5.8
	Animal kingdom	29	4.8
	Iconography	19	3.1
	Other	19	3.1
	Total	606	100

Source: Author’s edit.

). Within this group, the most common subcategory was “ordinary people” (36.1%, 84 out of 233), portrayed either as victims of CC (53) or as role models in mitigation, adaptation, and advocacy for solutions (24) (Figure 2A). They were followed by images of “politicians” (29.2%, 68), “scientists” (16.3%, 38), “business leaders” (9.9%, 23), and “celebrities” (8.6%, 20).

The second thematic category was “topical events” (20.6%), with “unusual weather” (49.6%, 62 out of 125) as the most prominent subcategory, illustrating the impacts of wildfires, floods, and droughts (Figure 2B). Of these images, 58.1% (36 out of 62) were set in an international context, emphasizing the global dimension of the climate crisis, while the remaining 41.9% (26 out of 62) depicted national disasters, with the drought in Doñana and wildfires in Spain as notable examples. Notably, 59.7% (37 out of 62) of these images did not feature people. Other subcategories included “protests” (21.6%, 27 out of 125) (Figure 2C), “political summits” (15.2%, 19), and “scientific congresses” (13.6%, 17).

Third was the category “technology/energy” (16.7%), split between images of “renewables and efficiency” (50.5%, 51 out of 101), “fossil fuels and pollution” (27.7%, 28), and “other infrastructures” (21.8%, 22). These images were mostly stock (54.5%, 55 out of 101) and showed iconic representations of climate solutions and causes: wind turbines and solar panels projecting a “clean future” esthetic, and images of industrial chimneys and factory complexes, both typically shown in uninhabited environments (Figure 2D). The remaining categories included “landscapes” (7.4%), “agriculture and food” (5.8%), “animal kingdom” (4.8%), “iconography” (3.1%), and “other” (3.1%). Unlike previous climate coverage, classic representations of Arctic landscapes and iconic species such as polar bears were minimal, with only 10 and 4 instances, respectively.

In the visual coverage of DANA’s first month, media outlets primarily focused on “politicians” (23.1%), followed by “destruction” (21.5%), depicting the material damage caused by the floods (Figure 3). The third most represented group was “rescue services and volunteers” (19%), while “victims” accounted for 13.5%. Images of “influential people”—including celebrities, community leaders, and business leaders—represented 9.1%. Meanwhile, photographs of “topical events” (5.8%)—including protests (9), tributes (6), political meetings (4), and looting (2)— had a marginal presence, along with images of the “animal kingdom” (1.1%) and “scientists” (1.1%). The “other” category (5.8%) included images that did not fit into any of the previous ones (Figure 3,

Table A5. Visual frames by article and image theme distribution in DANA coverage.

	Article Frame
	Problem-Only		Problem-Solution		Solution-Only		Neutral		Total
	n	%	n	%	n	%	n	%	n	%
Destruction	71	19.5	7	1.9	0	0	0	0	78	21.5
Politicians	43	11.8	6	1.7	31	8.5	4	1.1	84	23.1
Topical events	12	3.3	8	2.2	1	0.3	0	0	21	5.8
Influential people	13	3.6	2	0.5	7	1.9	11	3.0	33	9.1
Scientists	3	0.8	0	0	1	0.3	0	0	4	1.1
Other	5	1.4	5	1.4	5	1.4	6	1.7	21	5.8
Victims	27	7.4	16	4.4	3	0.8	3	0.8	49	13.5
Rescue services/volunteers	6	1.6	46	12.7	17	4.7	0	0	69	19
Animal kingdom	2	0.5	2	0.5	0	0	0	0	4	1.1
Total	182	50.1	92	25.3	65	17.9	24	6.6	363	100

Note: “Rescue services/volunteers” (n = 69; volunteers: 29, rescue teams: 36), “Victims” (n = 49; deceased: 5, affected people: 44), “Topical events” (n = 21; protests: 9, tributes: 6, political meetings: 4, looting: 2).

).

6.3. Visual Framing: Problems vs. Solutions

This section addresses the fourth research question, which explores whether the analyzed images depict the problem, the solution, or both. The case study of DANA was selected for this analysis because it allows for a more detailed examination of visual narratives within a specific crisis context. Images were analyzed both individually and within the context of their respective articles. At the individual level, out of the 363 coded images, 53.2% (193) highlighted the problem, 20.1% (73) focused on the solution, and 19.8% (72) included both perspectives. A total of 6.9% (25) of the images were classified as neutral, as they did not clearly depict either the problem or the solution. In considering the articles that included images as a whole, 50.1% (182) focused exclusively on problems, 25.3% (92) combined problem- and solution-oriented images, 17.9% (65) centered solely on solutions, and 6.6% (24) were classified as neutral (Table A5).

6.3.1. Problem- and Solution-Only Articles

The most common photos in “problem-only” articles emphasized the destruction caused by the floods, featuring images of cars piled up and swept away by the water, submerged streets, and damaged infrastructure. The absence of human presence in many of these photographs reinforced the scale of the disaster, aligning with the esthetic of the “apocalyptic sublime” (Figure 4a). News focusing on victims depicted them mainly in vulnerable situations. Notable examples include images of individuals passively observing the damage or retrieving belongings from the debris (Figure 4b).

Problem-only articles featuring political figures framed their representation within the political conflict surrounding accountability for crisis management, often captured during public statements or press conferences. Carlos Mazón, President of the Generalitat Valenciana, appeared in 42.8% (36 out of 84) of the “politicians” images, frequently depicted with a serious or downcast expression.

Conversely, solution-only articles highlighted the work of rescue services removing debris and conducting rescue operations. Volunteers were shown organizing supplies and clothing for those affected, as well as assisting in street clean-up efforts. Political figures featured within this solution-only condition were portrayed in a more active role, either announcing financial aid measures or participating in meetings aimed at coordinating recovery efforts.

6.3.2. Combined Framing: Problems and Solutions

Although most articles adopted a one-sided perspective, offering images solely of problems or solutions, a subset incorporated both perspectives within the same visual narrative. These articles aligned most closely with the principles of solutions photojournalism, as they provided the public with a more comprehensive visual account of the DANA floods, illustrating both the severity of the disaster and the collective efforts toward recovery. By including both types of images, the media leveraged the emotional resonance of problem-centered visuals while simultaneously fostering a sense of agency and resilience through solution-oriented representations.

This combined framing was particularly evident in articles that juxtaposed images of destruction with scenes of community solidarity and volunteerism. For example, a report by El País paired an image of a woman crying amid the aftermath of the DANA with photographs of a large group of volunteers mobilizing in the affected areas and military personnel working to clear the mud (Figure 5a–d). Likewise, another article depicted a group of volunteers removing debris from the streets, observed from a balcony by an older woman, highlighting the intergenerational nature of the community-led recovery (Figure 5e).

The portrayal of the deceased, with only five instances, was handled respectfully through indirect or distant scenes, such as a photograph taken by Alberto Saiz published in El País, showing emergency workers transporting a covered body to a funeral van (Figure 5f). In a similar vein, El Mundo juxtaposed an image of emergency teams removing a body with another depicting members of Madrid’s Emergency and Immediate Response Unit (ERICAM) distributing supplies (Figure 5g,h). These kinds of approaches helped humanize the disaster, emphasizing that while the damage was extensive, the response demonstrated the power of collective action in reconstruction efforts.

7. Conclusions

The primary aim of this study was to assess whether the visual representation of CC and related events, such as the 2024 floods in Spain, in mainstream media conforms to or diverges from the conventional visual patterns linked to the stagnant iconography discussed in the theoretical review and predominantly alarmist narratives. The quantitative analysis facilitated the identification of dominant themes and their frequency of occurrence, offering an overview of the visual organization of both coverages. Certain images, especially those depicting individuals and the effects of the crisis, predominated the visual narrative in both examined situations. Although the coverage of CC 2023 diverges from depictions of icy landscapes and endangered species, favoring more sociocentric narratives that prioritize individual experiences—consistent with the global trend of personifying the phenomenon (García Herrero, 2024; S. O’Neill, 2020)—both case studies showed a preference for imagery that emphasizes the scale of the disaster, particularly through the “unusual weather” category in CC 2023 and the emphasis on “destruction” in DANA.

The portrayal of the most catastrophic climate disaster of the 21st century in Spain somewhat aligns with the concepts articulated by Midberry and Dahmen (2019), which underscore thorough, precise, and humanizing visual storytelling. While problem-oriented photos were predominant, the inclusion of mixed and solution-focused frames contributed to a more balanced visual narrative compared to traditional disaster coverage. The depiction of rescue services and volunteers effectively communicated the community’s collaborative activity and resilience. Nonetheless, the prospects for enhancing the humanization of replies, such as emphasizing personal narratives, were limited. Furthermore, the presence of politicians frequently lacked explicit visual connections to concrete solutions and tended to be tied to conflict and blame, constraining their capacity as seeming agents of change. This observation is particularly relevant considering that images classified as “solution-oriented” featuring politicians often derived their framing from textual context rather than visual cues, indicating that the potential of images to convey political agency remains underexplored.

Attaining visual equilibrium between depicting the severity of the issue and emphasizing the remedies remains a central challenge. As noted by Midberry and Dahmen (2019), an overemphasis of solutions may convey the illusion of problem resolution, while an exclusive focus on the issues can lead to compassion fatigue among audiences. The incorporation of imagery depicting desolate landscapes or a single, reflective human figure amidst scenes of devastation enhances visual spectacle and amplifies the drama; however, it may also foster emotional detachment. The esthetic of the “apocalyptic sublime” (Toadvine, 2020), although powerful visually, perpetuates the notion that CC is inevitable rather than a challenge that can be mitigated by collective action, thus dissuading public engagement. Therefore, the ideal approach involves presenting both perspectives within the same article, offering a holistic visual narrative that acknowledges the gravity of the crisis while showcasing effective responses.

To achieve this visually balanced narrative, multiple photographs are often required to effectively convey both aspects of a tale. In the DANA case, articles featuring mixed framing generally incorporated a higher number of images, facilitating a more comprehensive and nuanced depiction of the incident. The juxtaposition of poignant emotional imagery, such as a woman watching the cleanup from her balcony, against glimpses of concrete progress, such as workmen clearing vehicles or reopening highways, generated a visual story that was both engaging and affirmative. Nonetheless, our findings suggest that these cases were more exceptions than the norm, as most articles took a one-sided visual approach, and photographs were predominantly used as illustrative supplements to text rather than as complementary visual narratives. The widespread reliance on isolated images in both case studies indicates that visual storytelling was not a priority and points to a resurgence of what Gürsel (2016) terms “fast photos”, a consequence of modern exigencies in journalism, wherein immediacy and resource scarcity compel journalists and editors to depict news with a single image, often lacking a direct correlation to the reported event. The latter was particularly noticeable in the 2023 CC coverage, probably due to its extensive use of stock and archive photographs, many of which were employed to illustrate solutions. The monotonous and frequently impersonal employment of clichéd imagery, such as chimneys and solar panels in uninhabited locations, sustains a constrained depiction of the issue and may foster a sense of detachment among observers. Corner et al. (2015) caution that these visuals are often ineffective in mobilizing the audience due to their deficiency in fostering the emotional connection required to encourage particular climate action. Moreover, the use of stock, corporate, and UGC images was also associated with lower precision in photographic content and captions. Accuracy in both visual content and captions is essential to ensure that crises are not misrepresented or misattributed to unrelated events. The findings suggest that original photography is preferable whenever possible, as it enhances narrative depth and also mitigates risks related to decontextualization and misleading associations. In cases where archive images are unavoidable, greater transparency is required regarding their selection and use—e.g., captions should consistently indicate the date and clearly state the image’s relationship to the event.

In summary, we can conclude that, while certain parts of the coverage were in line with recommended strategies, such as highlighting “real” people over dramatizing and politicizing the phenomenon—e.g., in the case of CC 2023, these outnumber images of political figures—others fell short. For example, while in DANA we see a narrative that shows us the problem and its immediate solution, in the general coverage of 2023, the representation of the causes of CC and its possible solutions were considered underrepresented and reductionist. Limiting an article that talks about solutions to the climate crisis to a close-up of plastic bottles on a beach to evoke the need for recycling, which is also a stock image, in no way follows a narrative that can foster engagement with the issue, thus wasting the potential of photography in this task and contributing to a predictable and unmemorable representation of events.

These findings underscore a broader editorial challenge in visual journalism, which is the tension between immediacy and precision, between urgency and depth. The predominance of quick photos and decontextualized visuals in climate crisis reporting reflects an over-reliance on image selection based on efficiency rather than documentary and constructive visual storytelling. However, this approach can have unintended consequences for public engagement. If the constant repetition of similar visual motifs desensitizes the public and contributes to the normalization of climate disasters, photojournalistic innovation in climate reporting may be essential to counteract public fatigue and maintain long-term commitment. The prevailing patterns of representation, although deeply rooted, are not immutable, but achieving a more balanced and engaging visual representation requires a shift in editorial priorities toward treating images as autonomous narrative components and toward more deliberate and transparent visual strategies that emphasize long-term recovery efforts by showing the before and after, clearly establish connections between images and text, as well as narrative bridges to other related issues, and elevate documentary photography above generic and institutional material, which could significantly improve the effectiveness of the visual coverage of environmental issues.