Climate Study Insights for the Tourism Sector: Analysis of Selected Pilot Regions in Croatia

Abstract

Understanding the impact of climate change on tourism is vital for the economies that rely on it. The tourism sector in Croatia, a country with diverse climatic regions, but also diverse features of tourism, is particularly sensitive to changes in climate variables such as 2 m air temperature and precipitation totals. This study analyzes trends in these two key climate variables from 1961 to 2024 across five representative climatic regions: the-mountainous Lika region (Ličko-senjska County), the Kvarner region on the northern Adriatic coast (Primorsko-goranska County), the Zadar region on the central Adriatic coast (Zadarska Counties), and northern continental Croatia (Varaždinska and Međimurje Counties). Linear trends, 5-year moving averages, and comparisons between two standard climate periods (1961–1990 and 1991–2020) were conducted. Using these data, the monthly self-calibrated Palmer Drought Severity Index (sc-PDSI) and Standardized Precipitation Index (SPI) for seven-time scales were calculated for the period 1961–2024 to assess drought conditions and their implications for tourism across the selected destinations. Frequencies of dry, near normal and wet months, estimated by SPI for a nine-month time scale (SPI-9) and a monthly sc-PDSI, were compared for two subperiods, 1961–1992 and 1993–2024. Meteorological data were contextualized for tourism stakeholders, with a focus on adaptation measures. Semi-structured interviews were conducted with tourism professionals in the study regions, providing qualitative insights into observed changes in climate and tourist behavior, operational challenges, adaptation strategies, level of community engagement, and opportunities envisioned. Objective climatological data were compared with the subjective perceptions of tourism experts using the principle of mixed methods, which allows for triangulation. The climatological data indicated a continuous trend of increasing mean annual air temperatures, as well as anomalies of average precipitation amount. The interviews revealed signals of emerging climate shifts, such as changes in the seasonality of visitors, concerns about water scarcity and heat stress. These findings were interpreted in the context of potential threats and opportunities for the tourism sector, highlighting region-specific adaptation strategies. By combining objective climate data with insights from tourism professionals, this study provides a comprehensive assessment of climate change impacts on tourism and informs for resilient tourism development across Croatia’s diverse regions. This paper presents a methodological framework for developing adaptation recommendations that draw on both empirical climate data and the lived experiences of tourism work practitioners.

1. Introduction

Recent regional assessments by the IPCC [1] for south-eastern European regions highlight the shifts in warming trends as manifested in warmer and increased heat waves and variability in precipitation patterns with drier summers, and more frequent and intense extreme events. These climate shifts carry implications for destination attractiveness, tourist health and infrastructure resilience [2,3]. Namely, the tourism industry is highly sensitive to climate variability and change, particularly in the Mediterranean basin where warming and shifts in precipitation patterns affect seasonality, thermal comfort, and water availability issues [4,5,6].

Local studies [7,8,9] have documented a measurable impact of elevated temperatures in popular Croatian destinations. For instance, a recent biometeorology analysis that applied the modified Physiologically Equivalent Temperature (mPET) index across major eastern-Adriatic tourist cities (including Zadar) found an intensification of thermal risk over recent decades, with implications for visitor comfort and seasonality [10,11]. This line of research suggests increased demand for heat-adapted services like shaded spaces, water amenities, and cooling, as well as rising heat stress in the high-tourism season window. In addition to these adaptation solutions, destinations are increasingly embracing the opportunities that climate change brings. Although the Adriatic coast remains predominantly a summer destination, the increasing thermal comfort in the pre-season and post-season suggests that these periods will become increasingly attractive. It seems that beach conditions during the summer season from May to September change, and that favorable conditions for tourist activities increase significantly in April and November, especially for sightseeing [8,12,13].

Many authors [14,15,16] indicate that climate change is driving pronounced shifts in precipitation patterns across the Adriatic region and in continental parts of Croatia. They provide evidence of redistribution of precipitation during the year, a decrease in the number of days with precipitation and an intensification of precipitation. All these shifts significantly affect the tourist offer and infrastructure in destinations, especially those on the Croatian coast and on the islands. For example, extreme rainfall episodes increasingly trigger flash floods that damage coastal promenades, beaches, marinas, and access roads, causing substantial costs for local governments and interruptions in the tourism season [17]. Heavy rains also accelerate coastal erosion and sediment transport, reducing beach width and compromising the esthetic and recreational value that is central to 3S tourism [18]. Reduced rainfall lowers groundwater recharge and limits freshwater availability, which poses increasing challenges for tourism-dependent water consumption during peak season [19]. In some destinations, prolonged drought conditions accelerate the risk of wildfires, which threaten visitor safety and reduce the attractiveness of forested recreational areas [20].

Amid the increasing droughts period in relation to growing tourism activities, the resilience of water supply systems in Croatia faces rising demand potentially coinciding with diminishing resources. Many studies have highlighted seasonal pressures caused by peak summer tourism, and influx of weekend visitors, rising temperatures and more frequent dry spells, highlighting the necessity of sustainability and climate adaptation measures in the Adriatic region, especially in the island region [21], as well as in continental Croatia [19]. Furthermore, several Croatian studies have applied practical indices to regional contexts to evaluate drought frequency trends and impacts in the Pannonian and Adriatic contexts [22,23,24].

To address these issues and to better engage tourism stakeholders across four climatically and touristically distinct Croatian regions (Figure 1) the mountainous Lika region (Ličko-senjska County), the Kvarner region on the northern Adriatic coast (Primorsko-goranska County), the Zadar region on the central Adriatic coast (Zadarska County), and northern continental Croatia (including Varaždinska and Međimurje Counties), this study analyzes long-term anomalies of mean annual air temperature (°C) at 2 m height and trends of mean annual precipitation totals (mm) including an insight into annual courses of monthly air temperature and monthly precipitation totals for two successive climate periods: 1961–1990 and 1991–2020. In addition, the nine-month time scale Standardized Precipitation Index (SDI-9) and the monthly self-calibrated Palmer Drought Severity Index (sc-PDSI) were calculated for the period 1961–2024 to assess drought risk and to support and motivate climate adaptation actions in tourism.

Additionally, this study incorporates qualitative insights from tourism practitioners in four destinations: Plitvice (Ličko-senjska County), Sveti Martin na Muri (Međimurska County), Krk (Primorsko-goranska County), and Pakoštane (Zadarska County), using semi-structured interviews to examine how climate trends and drought indices align with stakeholder perceptions. Recent multi-sectoral studies similarly show that integrating expert, stakeholder, and climatic knowledge strengthens the identification of vulnerabilities and adaptation needs [3,25].

Overall, this study integrates quantitative climate data with qualitative insights from tourism practitioners across a range of settings, including coastal, inland, island, protected, and rural areas to provide actionable understanding of how climate change is already affecting tourism in selected Croatian destinations. The resulting framework offers a practical tool to guide evidence-based, locally informed adaptation strategies in tourism planning and sustainable development.

To address these issues and provide actionable insights for tourism adaptation, this study combines quantitative analyses of climate variables, including mean annual air temperature, precipitation totals, Standardized Precipitation Index (SPI-9), and self-calibrated Palmer Drought Severity Index (sc-PDSI), with qualitative perspectives from tourism practitioners across four climatically and touristically distinct Croatian destinations (Plitvice, Sveti Martin na Muri, Krk, and Pakoštane).

Based on this integrated approach, the study aims to answer the following research questions:

• How have long-term trends in temperature, precipitation, and drought indices evolved across selected Croatian destinations?
• How do tourism practitioners perceive these climatic changes, and how do these perceptions align with measured climate data?
• How do climate variations, particularly drought and heat stress, influence tourist behavior, operational practices, and adaptation strategies in different destination types?

By explicitly linking climate trends with stakeholder perceptions and adaptation responses, this study provides a framework for understanding the interplay between environmental change and tourism dynamics, supporting evidence-based, locally informed adaptation strategies.

2. Theoretical Framework

The theoretical framework of this study rests on two complementary strands of research: (1) the analysis of the climate parameters of temperature and precipitation, and the drought indices SDI-9 and sc-PDSI, and (2) the examination of perceptions and experiential knowledge of tourism practitioners in shaping climate-related responses within selected destinations. By integrating quantitative climate indicators with qualitative insights from practitioners, the study follows recent recommendations to combine top-down climate data with bottom-up local knowledge to inform effective and context-specific adaptation planning [3,26].

2.1. Climate Data and Indices as Decision-Support Tools for Tourism

This study aims to analyze 2 m air temperature and precipitation totals for the period 1961–2024 for five weather stations across four different climatic regions of Croatia presented in Figure 1. Climate data and indices translate meteorological state into simplified metrics that can be used to assess climate impacts on human activities [4]. Among them, data on air temperature and precipitation shifts are critical for tourism activities, including drought indices that are increasingly important for tourism, particularly in water-dependent coastal and island regions where supply pressures are intensified during the summer season [15,21].

Recently, the Standardized Precipitation Index (SPI) and the self-calibrating Palmer Drought Severity Index (sc-PDSI) have become widely recognized as valuable tools for assessing drought severity, hydrological stress, and long-term soil moisture conditions [27,28,29]. Their application in Croatian climatology [22,24,29] demonstrates their relevance across both continental and Adriatic contexts. The SPI relies solely on precipitation data, offering a simple, flexible tool for monitoring drought across multiple time scales and regions, and is particularly suitable for spatial comparisons [30]. In contrast, the sc-PDSI integrates precipitation, potential evapotranspiration, prior soil moisture, and runoff to simulate a soil water balance, capturing both the severity and the persistence of wet and dry conditions [31,32]. While the SPI is particularly useful for rapid, spatially extensive detection of meteorological drought and short-term water shortages, sc-PDSI captures deeper soil-moisture deficits and long-term hydrological stress [33]. In tourism research, drought indices such as the SPI and the sc-PDSI have been used to examine potential risks to water-intensive tourism activities, destination attractiveness and ecosystem-based tourism resources [30,34].

2.2. Stakeholder Perceptions as a Lens on Climate Impacts

Tourism operators and local or regional government tend to evaluate climate risks through their own experience, knowledge, and sensitivity to environmental changes [35]. Racz et al. [36] consistently show that tourism professionals are aware of climate-related challenges, especially when they manifest tangible disruptions such as shifts in seasonality, more frequent heatwaves, water shortages or wildfire risk. In addition, perceptions shape prioritization of climate risks, preparedness, and willingness to invest in adaptation [37].

For this research, interviews were used to obtain the perceptions of tourism practitioners and to apply a bottom-up approach. The aim was to obtain valuable insight into socially experienced climate impacts, which do not always have to coincide with climatological data. For instance, stakeholders may report increased water stress even when some indicators show only moderate anomalies, reflecting the interaction between natural conditions and growing tourism demand [38]. Thus, they rely on so-called “mental models” supported by the knowledge of the local community [39,40].

Consequently, this approach underscores the critical role of stakeholders’ experiential knowledge in elucidating how climate hazards translate into operational challenges for tourism. While intensifying summer heatwaves and drought threaten water supply, visitor safety, and experience quality, climate-driven improvements in shoulder season comfort [7,12,13] may open adaptation opportunities. Recent research further documents a spectrum of adaptive responses, including offer diversification, infrastructure modifications, and strategic marketing reorientation toward shoulder seasons [40,41].

3. Methodology

The study employed a mixed-methods approach, integrating quantitative climate analyses with qualitative insights derived from stakeholder interviews grounded in observation. This approach, widely recognized in the literature [42,43,44,45], is founded on the principle that climate data and indices alone cannot fully capture the impacts of climate change on tourism, just as stakeholder perceptions by themselves cannot reveal the underlying climatic processes. By integrating both forms of evidence, the study aims to produce a more comprehensive and practice-oriented understanding of climate issues in selected tourist destinations.

3.1. Climate Data Collection and Processing

The starting reference for climate data collection was the climate atlas of Croatia [46], which covers the climate of Croatia for the periods 1961–1990 and 1971–2000. Air surface and precipitation projections for the 21st century were considered with reference to Branković et al. 2013 [47]. Examples of adaptation of climate change in the tourism sector according to the Njoroge study were also considered [48]. The quantitative climate data centers on the use of the sc-PDSI, which offers a long-term and spatially consistent measure of drought severity [36]. Together, these datasets provide an evidence base for identifying the timing, intensity, and geographic patterns of emerging climate pressures.

The sources of climate data used in the present study included 5 pilot meteorological stations: Varaždin (46°16′57.99″ N 16°21′49.99″ E, 167 m amsl), Ličko Lešće (44°48′36.39″ N 15°18′42.09″ E, 454 m amsl), Rijeka (45°20′13.39″ N 14°26′34.19″ E, 120 m amsl), Mali Lošinj (44°31′57.00″ N 14° 28′18.99″ E, 53 m amsl) and Zadar (44°07′48.39″ N 15′12′20.89″ E, 5 m amsl) for the period 1961–2024, i.e., a total of 64 years per meteorological station. Time series of air temperature, precipitation data, and their derived indices have the characteristics of stochastic (statistical) variables and statistical terminology is widely used in climatological studies [49,50]. Linear trends of annual mean air temperature anomalies, in reference to the WMO standard period (1961–1990), were calculated for the 5 pilot locations (Figure A2). Interannual and interdecadal variations were considered using original air temperature mean annual anomalies and 5-year moving averages, respectively. Linear trends significance has been estimated using Standard Normal Homogeneity Test–SNHT (Figure A1).

Air temperature monthly courses during a year for two WMO standard periods P1 and P2 (Appendix A, Figure A2) were calculated and compared using differences (P2-P1) of air temperature courses. This provided more detailed insight into the air temperature anomaly distribution differences for the two successive WMO periods for each month in the year for the 5 pilot locations. Interannual, interdecadal and trends of annual precipitation total anomalies analysis have been performed as for means air temperature for 5 pilot locations (Appendix A, Figure A3). In comparison of precipitation annual courses for two WMO standard periods instead of difference (P2-P1) a P2/P1 quotient is used. Multiplied by 100, the quotient is expressed in percentages (%), which is a more efficient way to compare results.

For calculating the Standardized Precipitation Index (SPI), the work of Mihajlović [51] and Cindrić et al. [52] was considered. They applied SPI to study drought patterns in the Pannonian region of Croatia and observed increasing drought frequency in recent decades. A classification system categorizing SPI values into dry and wet conditions, as proposed by McKee et al. [53], is presented in

Table A1. The SPI values and system of dry/wet categories [25].

SPI Value	Category
2.00 or more	Extremely wet
1.50 to 1.99	Severely wet
1.00 to 1.49	Moderately wet
−0.99 to 0.99	Near normal
−1.00 to −1.49	Moderately dry
−1.49 to −1.99	Severely dry
−2.00 or less	Extremely dry

(Appendix A).

Building on the original PDSI [31], Wells et al. [34] and Wells [54] introduced the sc-PDSI, in which the weighting factors, known as “duration factors,” are computed separately for each location. Van der Schrier et al. [55] applied the sc-PDSI across Europe and later used it to analyze droughts in the alpine region for the period 1800–2003. A classification of sc-PDSI values is provided in

Table A2. The PDSI values and system of dry/wet categories [26,71], using the PDSI classification by Palmer [34].

PDSI Value	PDSI Category
Above 4.00	Extreme wet spell
3.00–3.99	Severe wet spell
2.00–2.99	Moderate wet spell
1.00–1.99	Mild wet spell
0.50–0.99	Incipient wet spell
0.49 to −0.49	Normal
−0.50 to −0.99	Incipient drought
−1.00 to −1.99	Mild drought
−2.00 to −2.99	Moderate drought
−3.00 to −3.99	Severe drought
Below −4.00	Extreme drought

(Appendix A). Consequently, based on the dry/wet categories of the sc-PDSI, the number of dry months is expected to increase in the recent period compared to previous decades.

3.2. Stakeholder Interviews

A semi-structured interview based on a narrative perspective was carried out to conduct the qualitative component of the research. These narratives capture experience, contextual knowledge, and socio-cultural interpretations of climate variability that are not accessible through instrumental records alone [42,56]. A semi-structured interview guide was used to ensure consistency across interviews while allowing flexibility for respondents to elaborate on destination-specific experiences and priorities.

Tourism practitioners, destination management organizations, tourism and protected area managers, local authorities, tour operators, and accommodation providers from the same region were interviewed. Interviewees were selected using a purposive sampling strategy. Prior to data collection, a preliminary mapping of relevant stakeholders was conducted in each destination to identify key actors involved in tourism governance, planning, and service provision, as well as those with direct experience of climate-related impacts. Stakeholders were then approached directly and invited to participate in the study. While the initial research design envisaged conducting between 20 and 25 interviews to ensure broad representation across destinations and stakeholder groups, the final number of completed interviews was lower due to limited stakeholder availability and scheduling constraints. In total, 17 stakeholder interviews were conducted across the 4 study regions: 5 interviews in the Plitvice region (Ličko-senjska County) and 4 interviews in each of the remaining destinations (Sveti Martin na Muri, Krk, and Pakoštane). Despite the reduced sample size, the interviews provided sufficient thematic saturation, with recurring patterns and consistent themes emerging across destinations.

Interviews were conducted between June and November 2025. A small number of interviews were conducted in person, while the majority were carried out online to facilitate scheduling flexibility and participation across regions. Interviews lasted on average between 25 and 40 min. The interviews explore how practitioners perceive climate-related changes, including changes in visitor behaviors, operational challenges arising from heat or water shortages, and the extent to which extreme events influence destination management. The interviews also reveal local knowledge, adaptation priorities and the practical constraints that shape decision-making.

The study adhered to established ethical standards for social science research. Participation was voluntary, and all interviewees were informed about the purpose of the study, the use of the data, and their right to withdraw at any time. Anonymity and confidentiality were ensured by removing identifying information from transcripts and reporting results in aggregated form. No personal or sensitive data were collected beyond professional roles and institutional affiliations.

The qualitative analysis followed a thematic coding approach. An initial coding framework was developed based on the study’s analytical dimensions (climate hazard impacts, community engagement, and adaptation measures). This deductive structure was complemented by inductive open coding to capture emergent themes raised by participants. Codes were subsequently grouped into higher-order categories (e.g., heat stress, water scarcity, infrastructure damage, governance coordination, adaptation maturity), enabling systematic comparison across destinations (

Table A3. Interview coding framework.

Analytical	Code	Description	Example Topics
Dimension
Climate hazard impacts	Heat stress	Perceived effects of high temperatures on tourists, activities, and operations	Activity cancelation, demand for air conditioning, altered daily schedules
	Water scarcity	Impacts of drought and water shortages	Water restrictions, ecosystem stress, destination image
	Extreme events	Effects of storms, floods, wind	Infrastructure damage, safety concerns, service disruption
Tourism system sensitivity	Seasonality shifts	Changes in timing and distribution of tourism demand	Decline in winter tourism, growth of shoulder seasons
	Infrastructure vulnerability	Sensitivity of tourism infrastructure to climate hazards	Trail damage, marina repairs, beach erosion
Community engagement and governance	Institutional coordination	Cooperation among DMOs, municipalities, utilities, park authorities	Integrated planning, governance gaps
	Local participation	Role of local communities and civil society	Associations, awareness, stakeholder involvement
Adaptation measures	Operational adaptation	Practical, short-term responses	Shading, hydration points, flexible scheduling
	Strategic adaptation	Long-term planning and investments	Climate-resilient infrastructure, diversification
Opportunities	Product diversification	New or expanded tourism products linked to climate change	Wellness tourism, eco-tourism, cycling
	Sustainability pathways	Perceived potential for climate-resilient development	Renewable energy, water-saving technologies

, Appendix A). Coding was performed manually by the research team, and the results were iteratively reviewed to ensure internal consistency and analytical transparency. Responses were coded thematically and compared with the objective climatic findings.

This framework combines deductive coding, based on the study’s analytical structure, with inductive coding to capture emergent themes raised by participants. Codes were iteratively refined and grouped into higher-order categories to enable systematic cross-regional comparison.

3.3. Application of Mixed Methods for Descriptive Comparison

While some parameters and indicators can be measured directly, this is more often not the case [57]. Their connection to the phenomenon they aim to represent may vary in strength. For instance, “precipitation amount” serves as a direct indicator of precipitation. In contrast, indirect (proxy) indicators are employed when direct measurements are unavailable or when complex states should be captured like drought. Consequently, vulnerability assessments typically use a combination of direct and proxy indicators [25]. Additionally, within socio-ecological systems affected by climate change, available data often exist in different formats, which requires systematic comparison using accepted methods for processing [58,59].

Integrating climatic data with stakeholders’ knowledge offers a more comprehensive understanding of vulnerabilities and adaptation requirements. The methodological approach for combining quantitative meteorological data and qualitative narratives collected through interviews is elaborated on

Table 1. Rationale for the application and comparison of qualitative and quantitative data.

Reason	Rationale
Climate change impacts are both physical and experiential	Meteorological data captures long-term changes, while stakeholder narratives reveal impacts on tourism operations, and decision-making, which quantitative metrics alone cannot capture [25,42]
Narratives contain early-warning signals	Stakeholders often perceive emerging challenges (e.g., wildfire risk, water scarcity, or tourist behavior changes) before they are detectable in climatological indices, highlighting the value of local experiential knowledge [43,60]
Triangulation strengthens validity	Convergence between qualitative and quantitative data reduces method-specific bias and increases confidence in findings (if various independent data sources produce consistent results), while discrepancies reveal gaps in awareness, mismatches between perception and climate data, or local micro-climatic dynamics [44,61]
Adaptation planning require data and lived context	Narrative insights capture operational constraints, visitor behavior, and socio-cultural meanings, enabling policy recommendations grounded in real-world context [62,63]
International climate agreements support this approach	Mixed-methods studies are widely used in IPCC reports, climate-services research, and socio-ecological systems analysis, aligning this approach with established scientific practices [1,64]

.

Through comparative analysis of temperature trends, precipitation patterns, and SPI and sc-PDSI index results with stakeholders’ narratives by mixed method integration (Figure 2) this study triangulates physical trends, lived experience, and socio-behavioral impacts, generating a more actionable understanding of climate change risks.

To integrate quantitative data and narrative evidence, the following triangulation approach was used:

• Validation (identification of convergence): identifying where stakeholder perceptions align with observed climatic trends.
• Analysis of divergences (focus on differences): detecting mismatches between datasets, e.g., cases where climatic changes are visible in data but not perceived by stakeholders, and where perceived emerging challenges are not yet reflected in long-term climate indices.
• Regional contextualization: linking climate shifts with geographical features and tourism typologies of specific destinations.

4. Results

4.1. By Air Temperature and Precipitation Data Andderived Drought Indices

4.1.1. Trends in Annul Mean Air Temperature and Annual Precipitation Totals

Statistical tools for Microsoft Excel were applied for linear trend calculations using mean annual air temperature time data for five pilot locations. The results are represented for the Varaždin and Rijeka weather station air temperature in Figure 3 and precipitation in Figure 4. These two stations were selected for display in this study, as representative of the mainland and coastal areas of Croatia (other images are available if needed).

The results of the mean annual air temperature and total precipitation shown in Figure 3 and Figure 4 are not robust enough to accurately describe the linear trends shown. Therefore, the theory of homogeneity of climate data [65,66] (Alexandersson and Moberg, 1997, Pandžić and Likso, 2010) was additionally applied within the Standard Normal Homogeneity Test (SNHT) and the results are presented in Figure A1 (Appendix A). For this purpose, the time series should have some characteristics: belonging to a Gaussian distribution, with a constant standard deviation and insignificant autocorrelation. As the first and the last 5 years of T-series data are not confidential, they were removed, as suggested by Alexandersson [67].

Variation coefficients for annual air temperature and precipitation trends for the five pilot destinations (L1–L5) were also calculated. The results presented in

Table 2. Values of statistical parameter “To”. Asterisk (*) indicates equal or higher than critical value of 8-8 for confidence level of p = 95% for 5 pilot locations: Varaždin (L1), LičkoLešće (L2), Rijeka (L3), Mali Lošinj (L4) and Zadar (L5), (see Figure A1).

Period	Variable	L1	L2	L3	L4	L5
1961–2024	(a) Values of To for mean annual air temperature linear trends and location change	37.1 *	31.4 *	40.1 *	40.2 *	41.5 *
1961–2024	(b) Values of To for annual precipitation total’s linear trends and location change	7.6	5.7	3.8	1.9	3.1

show that the mean annual air temperature indicates the presence of a “natural global warming signal” with high confidence, which is not the case when considering precipitation data.

The results presented in Table 2 show that the mean annual air temperature for the pilot regions indicates that a “natural global warming signal” is present, with very high confidence, which is not visible as such in the case of precipitation.

In addition to significant linear trends for air temperature interdecadal variations, represented by 5-year moving averages, could be observed, but still without visible positive trends which occurred just after the beginning of the 1990s at all pilot locations. This significant positive trend is considered to be a consequence of global warming. In contrast to that, a significant positive/negative trend of precipitation was still not observed. However, interdecadal variations are visible, indicated by the 5-year moving averages. The non-significant values of statistical parameter “To” represented in Table 2 support the conclusions stated above.

4.1.2. Annual Courses of Air Temperature and Precipitation for Two Successive WMO Periods

IPCC regional climate projections for Europe [68] revealed that the more precise analysis of climate variable trends requires a detailed study within a year, i.e., on seasonal and monthly scales. A typical example is the trend of precipitation amounts in Mediterranean areas, where colder season precipitation has a positive trend while in the warmer season the trend is negative, especially in its peripheral parts. Air temperature shows the opposite case.

This is partly confirmed by comparing the annual course of air temperature and precipitation amounts for the Varaždin and Mali Lošinj pilot locations, (Figure A2 and Figure A3, Appendix A, respectively). It is evident that the positive difference in air temperature, between two the periods (P2 and P1) were the highest for the summer months for the coastal Mali Lošinj weather station while the difference for the continental Varaždin weather station was not significant. It is important to note that the Mali Lošinj weather station was chosen for this comparison instead of Rijeka because the Mali Lošinj station has the stronger maritime influence [68]. The influence of the sea manifests itself in the difference between spring and autumn air temperature [46]. In Figure A2b (Appendix A) the strong influence of the sea on the Mali Lošinjstation is visible, and autumn was warmer than spring. Corresponding differences in precipitations are visible for the autumn months for both weather stations. From a touristic point of view this insight is important because of the seasonal character of the number of visitors. Figure A3 (Appendix A) clearly shows that P2 recorded higher precipitation compared to P1. This increase may be related to the increase in sea and air temperatures during the summer, leading to higher precipitation in P2 [68].

4.1.3. A Comparison of SPI-9 and sc-PDSI Indices Results

The results of the SPI-9 (calculated with a 1-month time scale) for the Rijeka weather station are presented in Figure A4 (Appendix A) alongside the monthly sc-PDSI values. Although the SPI was computed for seven-time scales, SPI-9 (Appendix A) was selected for detailed analysis because it exhibits the highest average correlation with sc-PDSI, as partially illustrated in Figure A5 (Appendix A). Figure A4 further indicates that the maximum correlation between SPI and sc-PDSI varies with the aridity of the study area that is shown in Obuljen [69], an effect that has been observed at the global scale by Nwayor and Robeson [70]. Despite the limited spatial range of the pilot region, the analysis shows that correlation coefficients between the SPIs and sc-PDSI tended to decrease for more arid pilot stations, suggesting that aridity influences the consistency between these drought related indices.

If the SPI-9 and sc-PDSI numerical values are classified into three categories dry (D), wet (W) and near normal (N), and summed for two subperiods 1961–1992 and 1993–2024, then it can be shown that the monthly sc-PDSI occurred more often for the second period than the first one, which was expected because of the global warming (

Table 3. Comparison of: (a) one-moth time scale and 9-month time scale SPI (SPI-9) frequencies and (b) frequencies of the monthly sc-PDSI for the above, near normal and below normal categories. The comparison refers to subperiods 1961–1992 and 1993–2024, respectively. Additional description is presented in the table.

(a) Standardized Precipitation Index for 9-month scale (SPI-9) wetness/dryness category frequences for pilot locations for monthly Time Scale
Varaždin			Ličko Lešće			Rijeka			Mali Lošinj			Zadar
W	D	N	W	D	N	W	D	N	W	D	N	W	D	N	Period
49	53	282	41	72	271	46	66	272	52	54	278	47	49	288	1961–1992
69	68	247	79	58	247	78	58	248	81	51	252	71	70	243	1993–2024
(b) Monthly Self-calibrated Palmer Drought Severity Index (sc-PDSI) wetness/dryness category frequences for pilot locations
Varaždin			Ličko Lešće			Rijeka			Mali Lošinj			Zadar
W	D	N	W	D	N	W	D	N	W	D	N	W	D	N	Period
79	23	282	102	33	249	97	24	263	71	41	272	130	26	228	1961–1992
45	70	269	105	57	222	87	78	219	86	66	232	56	63	265	1993–2024

(a) Legend for SPI-9: D (dry) ≤ −1, W (wet) ≥ 1, N (normal) > −1 and <1. (b) Legend for sc-PDSI: D (dry) ≤ −2, W (wet) ≥ 2, N (normal) > −2 and <2.

). If global warming continues to rise this effect will become stronger, especially in arid regions around the globe, which was the conclusion Nwayor and Robeson [70]. Thus, whether the SPI or sc-PDSI index is more appropriate for application in practice should be carefully considered, especially in the case of long-term planning.

According to Figure 5, both the SPI-9 and the sc-PDSI describe one of the dryest periods since 1961 until nowadays. Perhaps on a centennial scale, which includes climate projections, a more comprehensive multivariate index such as the sc-PDSI might have an advantage.

4.2. Stakeholder Perceptions

The analysis of climate challenges in selected Croatian destinations reveals differentiated patterns related to three analytical dimensions: the impact of climate hazards, community engagement, and adaptation measures with opportunities that may arise from climate shifts and could improve adaptation strategies (

Table 4. Results of tourism practitioners’ interviews in selected Croatian destinations.

Topic	Ličko-Senjska County (Plitvice Region)	Međimurska County (Sveti Martin na Muri)	Primorsko-Goranska County (Krk)	Zadarska County (Pakoštane)
Seasonality shifts	Fewer winter visitors due to reduced snow; longer spring/autumn season	Higher summer heat reduces daytime outdoor activity; growth in wellness and gastronomy in shoulder seasons	Increased arrivals in pre- and post-season; peak summer less comfortable	Clear shift to early morning/evening activities; spring/autumn promoted for cycling, birdwatching
Heat stress	Afternoon activities are often canceled; forest trails less attractive	Tourists avoid outdoor activities during midday; rise in demand for indoor/cultural programs	Guests demand more shaded and air-conditioned facilities	Strongly impacts tourist comfort; air conditioning now essential
Water scarcity	Concern about reduced water levels in rivers and waterfalls; rationing not yet in place	Drought lowers river flow, affecting Mura/Drava activities	Seasonal shortages on islands; restrictions introduced	Acute issue in peak season; Vrana Lake ecosystem visibly stressed; water restrictions harm destination image
Infrastructure damage	Trails, bridges, and signage frequently damaged by floods or storms	Some damage to bike/hiking paths, managed with local budgets	Coastal promenades, marinas, beaches face frequent storm damage	Storm surges damage ports and beaches; costly repeated repairs
Wind/storms	Strong winds affect Senj/Karlobag safety; impact on outdoor tourism	Stronger winds reduce comfort but rarely cancel activities	Bora disrupts ferries and nautical tourism	Storm surges linked to extreme winds threaten small ports and coastal zones
Community engagement	Local community engaged in park management; limited capacity for broader governance	Strong local associations, but adaptation planning fragmented	Moderate; coordination between DMOs, utilities, and municipalities still weak	Stakeholders call for stronger integration between local gov’t, DMOs, and utilities
Adaptation measures	Trail reinforcement, visitor flow management, diversification beyond snow tourism	Shaded routes, hydration points, flexible scheduling of events	Infrastructure upgrades, water-saving policies, promotion of shoulder seasons	Pilot water-saving projects, eco-tourism, renewable energy interest
Opportunities	Branding Plitvice as year-round nature destination	Wellness, cultural, and eno-gastro tourism	Extending season, eco-sports, nautical diversification	Leadership in eco-tourism, sustainable water and energy solutions

). These findings are based on a systematic analysis of stakeholder interviews and reflect stakeholders’ common and specific perceptions of climate risks and destination-specific responses.

4.2.1. Climate Hazard Impacts on Tourism Destinations

Across all destinations, stakeholder perceptions of climate hazard impacts closely reflect observed climatic trends, reinforcing the validity of the qualitative findings. However, the extent to which climate hazards translate into operational disruptions varies considerably depending on the level of institutional preparedness and the integration of sustainability standards within destination management.

Mountain and protected-area environments in Ličko-senjska County (Plitvice region) exhibit pronounced vulnerability to hydrological and forest-based climate hazards. The reduction in snow cover has contributed to a decline in winter visitation, while more frequent storms and floods cause recurrent damage to trails, bridges, and tourist infrastructure. Heat stress increasingly restricts afternoon activities during summer season, and concerns over declining lake and waterfall water levels highlight the sensitivity of nature-based tourism to changing hydrological regimes. Similarly, in the continental wellness-oriented destination of Sveti Martin na Muri in Međimurska County, climate stressors manifest primarily through elevated summer temperatures and reduced river flows. High midday heat lowers the attractiveness of outdoor activities, prompting a shift toward indoor or wellness-oriented experiences. Although infrastructure damage is less severe than in other regions, extended drought periods affect the usability of river-based recreational services. Destinations with established environmental certifications, such as the Plitvice region and Sveti Martin na Muri, demonstrate a higher capacity to anticipate and mitigate climate-related impacts, particularly in relation to visitor flow management and ecosystem protection. In contrast, destinations with more fragmented governance structures experience climate hazards primarily as reactive management challenges rather than as drivers of strategic transformation.

Island destination Krk in Primorko-goranska County is exposed to a broader combination of hazards. Increased temperatures and prolonged dry periods intensify water scarcity, leading to seasonal restrictions on water use. Coastal storms result in repeated damage to promenades, beaches, and marina infrastructure, while strong winds, particularly the Bora, intermittently disrupt ferry connections and nautical tourism. Similar patterns appear in coastal destination Pakoštane in Zadar County, where heat stress significantly reduces tourist comfort during peak summer months. Water scarcity is acute, with the Vrana Lake ecosystem showing signs of stress during high-demand periods. Storm surges and extreme wind events repeatedly damage ports and beaches, increasing the financial and operational burden on local authorities. Across all destinations, stakeholder perceptions of climate hazard impacts closely reflect observed climatic trends, reinforcing the validity of the qualitative findings. However, the extent to which climate hazards translate into operational disruptions varies considerably depending on the level of institutional preparedness and the integration of sustainability standards within destination management.

4.2.2. Community Engagement

Community involvement in climate resilience planning varies among regions. In the Plitvice area (Ličko-senjska County), local engagement is largely concentrated within national park management structures, with limited capacity to coordinate adaptation measures beyond protected-area governance. Stakeholder interviews indicate that formalized sustainability frameworks and certification schemes play an important role in structuring community engagement in climate adaptation. In Sveti Martin na Muri, the presence of EU Ecolabel certification, EMAS registration, and ESG-oriented strategies within the thermal resort has contributed to stronger awareness of environmental responsibility and facilitated cooperation between tourism operators, local authorities, and civil society actors. Conversely, the destination of Sveti Martin na Muri (Međimurska County) benefits from strong civil society participation, although adaptation initiatives remain fragmented and lack an overarching regional framework. Similarly, the Green Destination certification in the Plitvice region provides an institutional framework for environmental management and stakeholder coordination, although community engagement remains largely concentrated within protected-area governance structures rather than broader destination-wide collaboration.

On Krk (Primorko-goranska County), community engagement is moderate and characterized by partial coordination between destination management organizations (DMOs), municipalities, and tourism service providers and infrastructure managers (e.g., ferries, promenades, beaches, nautical tourism facilities). Nevertheless, gaps in integrated planning limit the effectiveness of adaptation efforts. In Pakoštane (Zadar County), stakeholders recognize the urgency of climate impacts, particularly those related to water scarcity and coastal hazards, and have increasingly called for stronger collaboration between local government, infrastructure and nature park managers DMOs, and tourism practitioners. These efforts are strongly supported by the management structures of the Vrana Lake Nature Park, which implement structured protection measures in the park area and raise awareness among residents and tourists about the need for climate adaptation. Overall, community engagement in climate resilience appears strongest where sustainability certification is embedded in destination governance, while weaker coordination persists in destinations where adaptation relies primarily on voluntary or ad hoc initiatives.

4.2.3. Current Adaptation Measures

The destinations demonstrate varying levels of adaptation maturity. The Plitvice region has implemented several operational measures, including reinforcement of hiking trails, improved visitor flow management, and diversification away from snow-dependent tourism. However, these efforts remain predominantly reactive. The analysis suggests that destinations with a higher prevalence of certified environmental management systems exhibit more structured and forward-looking adaptation approaches. In Sveti Martin na Muri, adaptation measures implemented by the thermal resort are closely aligned with EU Ecolabel and EMAS requirements, emphasizing resource efficiency, visitor comfort, and long-term risk reduction. In destination Sveti Martin na Muri, adaptation focuses on enhancing visitor comfort through shaded routes, hydration points, and flexibility in scheduling outdoor events. These measures align well with the destination’s growing emphasis on wellness and cultural offerings.

The island of Krk shows more advanced adaptation planning, with significant investments in climate-resilient infrastructure, water-saving policies, and strategic promotion of shoulder-season tourism. On the island of Krk, the presence of numerous hotels holding certifications such as Green Key and ISO standards supports incremental improvements in energy efficiency, water management, and operational resilience. However, stakeholder interviews reveal that these measures are often implemented at the enterprise level and are not yet fully integrated into a coherent destination-wide adaptation strategy.

Pakoštane is similarly active in piloting water-saving initiatives and renewable energy sources, including water and nature conservation policies and promotion eco-tourism, driven by the high seasonality and periodically local water shortages.

These findings highlight a distinction between operational adaptation driven by individual actors and strategic adaptation embedded in destination governance, underscoring the need for stronger coordination mechanisms to upscale certified practices beyond individual tourism enterprises.

4.2.4. Opportunities for Climate-Resilient Tourism Development

Across pilot regions, stakeholders identified a set of emerging development opportunities that align with their distinct environmental and socio-economic contexts. Taken together, the stakeholder perspectives illustrate that while climate hazards are universally experienced, adaptive capacity is uneven and closely linked to institutional maturity, governance coordination, and the strategic use of sustainability certifications. In Plitvice region, climate-induced seasonality shifts and reduced dependence on snow-based activities are seen as an opportunity to strengthen the destination’s position as a year-round nature-based tourism area, emphasizing hiking, ecological interpretation, and immersive landscape experiences beyond the traditional summer peak. Tourism practitioners from Sveti Martin na Muri reported a growing potential for wellness, cultural, and ecological and gastro tourism, building on the region’s thermal resources, rural heritage, and gastronomy, special tourism products that are less sensitive to heatwaves. Such products could support a more evenly distributed visitor flow throughout the year.

Stakeholders consistently emphasized that existing sustainability certifications provide a foundation for advancing climate-resilient tourism development. In destinations such as Sveti Martin na Muri and Plitvice, certified environmental management systems are perceived not only as compliance tools but also as strategic assets that support product diversification, branding, and long-term competitiveness under changing climatic conditions.

On the island of Krk, respondents highlighted opportunities to extend the tourist season through eco-sports, cycling, hiking, and diversified nautical products, taking advantage of increasingly favorable shoulder-season conditions and strong infrastructural connectivity. Finally, respondents form Pakoštane sees strategic potential in positioning itself as a leader in eco-tourism and sustainable water- and energy-management solutions, particularly given the region’s exposure to drought and water scarcity. This includes promoting nature-based experiences connected to Lake Vrana while leveraging ongoing investments in water-saving technologies and renewable energy. In Krk and Pakoštane, respondents identified significant opportunities to leverage existing certified accommodation providers and emerging eco-tourism initiatives as catalysts for broader destination-level climate strategies, particularly in relation to water management, energy transition, and seasonality management.

4.3. Triangulating Climate Data and Narratives with Regional Contextualization

The mixed methods framework (Figure 2) followed established protocols [42,43,44], but extended them through explicit regional differentiation for the triangulation of climate data and stakeholder narratives. Rather than treating Croatia as a homogeneous whole, this analysis systematically compared four different tourism typologies-mountain/protected areas (Plitvice Lakes National Park), continental wellness destinations (Međimurje), coastal/island locations (Primorsko-Gorski Kotar County), and coastal lake ecosystems (Pakoštane) and integrated additional qualitative information.

Examining convergences, divergences, and regional contextualization that link climate change with geographic features and tourism typology, it was found that tourism in Ličko-senjska County (Plitvice area) is under hydrological stress. The tourism identity of the Plitvice region relies on water resources. It is the oldest Croatian National Park with a UNESCO protected system of cascading karst lakes and tributaries. Triangulation revealed strong convergence between climate data (Table 2 and Table 3) and stakeholder observations (Table 4). The increase in the number of dry months according to the sc-PDSI directly coincides with stakeholder reports of “visible declines in water levels in lakes, rivers, and waterfalls” and concerns about “esthetic degradation of karst lakes” (Table 4), which are tangible manifestations of the multi-month soil moisture deficits recorded by sc-PDSI (Table 3). A significant divergence occurs between stable annual precipitation trends and stakeholder reports of frequent infrastructure damage. This finding reveals a fundamental limitation of trend analysis based on annual precipitation aggregates. Infrastructure damage does not arise from the total annual precipitation amount, but from the intensity, duration, and temporal concentration of individual precipitation events (characteristics that the annual total does not capture). The stakeholder emphasis on flood damage highlights that even in the absence of long-term trends in total precipitation, changes in the characteristics of extreme events (increased storm intensity, changes in seasonal distribution, or the occurrence of rain on snow instead of gradual melting) create significant operational challenges. This divergence demonstrates that annual precipitation statistics are insufficient for planning tourism infrastructure in mountainous environments where dynamics at the level of individual events determine maintenance costs and visitor safety. Effective planning requires data on the intensity of extreme precipitation, the frequency of floods, and the temporal distribution of precipitation events relative to tourism seasons.

Finally, regional contextualization highlights the high sensitivity of the destination to water availability. This sensitivity, combined with the limited possibilities of substituting the tourism product (nature-based tourism), creates a significant vulnerability for the destination, despite its moderate exposure to climate change compared to coastal regions.

The Međimurje region is known for its differentiated continental tourism based on thermal springs, cultural heritage, riverside recreation and gastronomy. Data triangulation reveals moderate concerns among respondents, despite the increase in temperatures (Figure A2a) and dry spells (Table 3). Stakeholders describe the impacts as manageable: “drought reduces river flow, affects activities on the Mura and Drava rivers” and “increased summer heat reduces outdoor activities during the day” (Table 4), rather than as existential threats. This convergence is highlighted by a shift towards “indoor wellness and cultural activities”, which represents a “tacit” adaptation that reduces exposure to outdoor heat. Divergence occurs around local hydrological sensitivity. Stakeholders report that “smaller tributaries show a pronounced decrease in flow during the summer months, which affects fishing and recreation more than larger rivers”, even though regional climate data do not show an increasing trend in precipitation (Figure A3a). This points to local factors, such as soil properties, groundwater abstraction, micro-catchment characteristics that are not captured by regional meteorological stations.

Regional contextualization shows how tourism diversification in the Međimurje region ensures climate resilience. Moreover, tourism has evolved from multiple resource bases like cultural and health tourism that are not highly sensitive to climate shifts. The increase in dry periods (sc-PDSI, Table 3) suggests that the destination may not maintain some of its characteristics if groundwater levels or river flows fall below ecological thresholds.

The Primorsko-goranska County, including the City of Rijeka and the islands of Krk and Mali Lošinj is a developed region of sun and beach tourism, characterized by high seasonality, water-intensive infrastructure and limited freshwater resources. Triangulation reveals an exceptionally strong convergence between warming signals and thermal discomfort. Mali Lošinj and the coastal station of Rijeka show the strongest increase in summer temperatures (Table 2, Figure A2b). Stakeholder narratives agree with this trend: “the peak summer months are becoming less pleasant”, “guests are increasingly looking for air-conditioned accommodation” and “visitors are less mobile during the peak midday heat, concentrating beach activities in the early morning and evening”. In addition, stakeholders note that summer months are becoming less comfortable and air conditioning has become a necessity for visitors. A critical convergence occurs around water scarcity in Mali Lošinj, where sc-PDSI (Table 3) indicating significant water stress. Stakeholders agree with sc-PDSI: “seasonal water scarcity on islands is acute, with constraints during peak tourism periods” and “limited freshwater resources constrain development”. This finding confirms that sc-PDSI better detects water stress relevant to tourism than indices based solely on precipitation (Figure A3b). It should be noted that in this, as in other coastal destinations in Zadar County, high temperatures drive evapotranspiration that exceeds precipitation input. Stakeholders also report increased damage to infrastructure from storms: “more frequent and intense climactic weather events cause coastal erosion, damage to marinas and coastal flooding”. This divergence reflects the Plitvice pattern, where annual precipitation aggregates “mask” changes at the extreme weather event level.

Regional contextualization highlights island vulnerability as manifested in limited water availability and reduced resilience of infrastructure that is often damaged by severe storms. In addition, the connection between water and energy is becoming critical. Air conditioning is extremely important for thermal comfort, and developed tourist infrastructure (e.g., swimming pools, promenades, catering) also request significant amount of energy. However, autumn remains warmer than spring. This is an opportunity for tourism to adapt to climate change. The extension of the tourist season and the offer of other content, such as outdoor activities, should be accompanied by active promotion.

In Zadar County, the focus of this study was on the coastal settlement of Pakoštane, known for its lake ecosystems and amenities in Vrana Lake Nature Park (Croatia’s largest natural lake). Triangulation indicates significant changes in the hydrological regime, leading to ecosystem degradation. The area of the Pakoštane municipality has been experiencing significant warming for a long time, with longer dry periods (Table 3), where increased sc-PDSI signals drier conditions. Stakeholder narratives confirm this transformation: “the Vrana Lake ecosystem is visibly stressed”, with “a pronounced drop in water levels affecting ecological functions and tourist aesthetics”. They describe “water restrictions during the peak season that damage the image of the destinations” and report that “air conditioning is now an essential infrastructure”, which coincides with the strongest warming signal.

A significant divergence appears, indicating a critical hydrological transformation that is occurring despite the lack of a significant precipitation trend (Table 2), as in Ličko-senjska and Primorsko-goranska Counties. The mechanism is clear: warming leads to increased evapotranspiration that exceeds stable precipitation inputs. Historically the wettest area, it now faces arid conditions not from reduced rainfall but from higher atmospheric water demand. This finding emphasizes that water resource planning in coastal Mediterranean regions should consider temperature trends alongside precipitation. Namely, temperature trajectories determine future water availability, where evapotranspiration dominates.

Regional contextualization reveals specific vulnerabilities. Lake Vrana is a crypto depression (its bottom lies below sea level) and is separated from the Adriatic by a narrow limestone barrier. This makes the lake vulnerable to potential saltwater intrusion if the water level drops sufficiently. Climate-induced hydrological stress threatens not only the lake level but also the fundamental character of the ecosystem. For a destination whose identity emphasizes the unique “blue and green” combination of beach and protected lake, ecosystem degradation threatens positioning in a key market.

5. Discussion

5.1. Climate Change Signals and Stakeholders Perception in Croatian Regions

Recent regional research [14,15,16] highlights increasing seasonal shifts, such as more intense heath wavs, short-duration rainfall events, prolonged dry spells, and altered timing of annual precipitation peaks, which directly affects water-dependent landscapes and visitor experiences. According to a recently published European Commission report [71], Croatia is facing increasingly prolonged and frequent droughts, which have major impacts across multiple sectors, especially agriculture, fisheries, forestry, energy, and tourism.

The comprehensive analysis of air temperature, precipitation, and drought/wetness indices (SPI-9 and sc-PDSI) demonstrates that Croatia maintains a robust historical climate dataset extending beyond 64 years, with temporal resolution spanning from hourly to decadal scales. This extensive observational record provides a solid foundation for assessing long-term climate variability and trends in the region. Considering the results of the mean annual air temperatures in the study area, a statistically significant positive trend was observed (Figure 3). This warming pattern shows consistency from approximately the early 1990s to the present, aligning with the accelerated global warming documented by the IPCC community [1,72,73].

Stakeholders across all destination types (coastal, inland, protected areas, islands, and rural landscapes) reported that high summer temperatures increasingly constrain outdoor activities, shorten the effective time window for excursions, and shift visitor activity patterns toward the early morning and late evening periods (Table 4). These stakeholder observations are consistent with quantitative evidence of positive trends in average annual air temperatures across Croatia, reflecting the global warming trajectory.

By using a triangulation of quantitative and qualitative data, this study shows how shared climate signals translate into differential vulnerabilities across destinations, with stakeholder narratives providing context-specific interpretations of quantitative climate trends. In the case of continental destinations, climatological data indicate an increase in temperature (Figure A2a), which is confirmed by stakeholder narratives (Table 4). They reported reduced summer outdoor comfort due to warming, as well as a change in visitor patterns due to observed positive temperature trends. However, continental location shows relatively milder warming compared to coastal destinations (Figure A2b).

Notably, quantitative analysis revealed spatial differences in warming patterns, with particularly strong summer warming at the coastal and island destinations. As tourism practitioners from those destinations reported, such temperature trends have increased arrivals in pre- and post-season, while peak summer is less comfortable. This situation increases demand for infrastructure and confirms climate projections that indicate that extreme events such as heat waves double with each Celsius degree of global warming [73], directly translating warming trends into challenges with operational costs and visitor comfort. This climatological and narrative evidence is consistent with studies across the Mediterranean that show enhanced warming in maritime climates, especially during summer [74]. This spatial pattern helps explain why coastal and island stakeholders expressed particularly acute concerns about thermal discomfort and cooling infrastructure requirements and the need to rapidly implement adaptation measures, while continental destinations reported more moderate, though still significant, heat-related challenges. The reported behavioral adjustments (Table 4) reflect documented thermal discomfort and declining climatic suitability in similar Mediterranean coastal and island destinations [75].

In contrast, precipitation changes did not show significant long-term linear trends (Figure 4 and Figure A3), but did show pronounced decadal variability, which is consistent with broader European findings of high interannual precipitation volatility [76]. This modest increase, however, does not translate to improved water availability. Tourism practitioners from selected Croatian destinations experienced changes in precipitation as less disruptive for most of the year compared to the impact of rising temperatures (Table 4). However, significant challenges were described in coastal and island destinations that are seeking to extend their summer season. This trend, combined with higher water consumption, can result in temporary water shortages, which directly affect tourism activities. This situation is consistent with the vulnerability profiles of Mediterranean islands [77]. Regarding extreme precipitation events and storm surges, tourism practitioners from all Croatian destinations reported significant infrastructure, coastal and nature attraction damage. These findings are consistent with studies documenting such disruptions [78,79].

Global observations indicate declining soil moisture despite stable or increasing precipitation, primarily attributed to significantly elevated evapotranspiration rates driven by rising temperatures [73]. Our findings confirm this pattern (Figure 5), with the frequency of dry months increasing over approximately the past three decades in the selected Croatian regions. These findings also show convergence with the stakeholder observations presented in Table 4. They reported increasingly frequent and severe drought episodes during the summer season, as well as water scarcity and described visible declines in water levels in lakes, rivers and waterfalls, with implications for the esthetic value, recreational use and ecological functioning of attractions such as karst lakes, wetlands and river corridors. These findings are consistent with the observed drying trends of the Mediterranean basin [73,74] and their increasing impact on tourism infrastructure and seasonality [80]. This pattern has also been documented in many European drought studies [81,82,83]. For destinations whose identity and market positioning depend heavily on freshwater features and snow cover, these shifts were considered potentially transformative. Such climate-induced hydrological shifts and snow cover reduction are described by Csete and Szécsi in their study of mountain tourism [84]. In general, water constraints have negative impacts on visitor satisfaction and destination image. As Gösslinget al. [85] reported, water scarcity directly affects tourists’ perceptions of destination reliability and quality.

Overall, the experiences of stakeholders and the results of quantitative drought indicators in the climatically diverse and topographically complex study area strengthen confidence that the observed trends represent strong climate signals rather than localized anomalies.

Additionally, stakeholders across all destination types recognized a clear link between drought and wildfire risk as a cascading climate effect, which is consistent with scientific evidence reported by Ruffault et al. [20]. Namely, increasing drought frequency and severity create conditions conducive to wildfire ignition and spread. Even destinations with historically low fire incidence expressed rising concern due to drier vegetation and expanding wildland–urban interfaces.

The observed divergence between precipitation trends (Figure 4) and perceptions of actual water availability (Table 4) highlight the complex interplay between temperature, evapotranspiration, and effective moisture in the regional water balance. The spatial coherence of the results of this study is very high despite the complexity of the study area, as demonstrated by the sc-PDSI results (Figure 5). Compared to SPI-9, the sc-PDSI provided a better basis for spatial comparisons between regions characterized by different climate regimes (Figure A4). As shown by Wells et al. [34], the self-calibration algorithm adjusts the behavior of the index and thus more accurately reflects local climate anomalies, improving both temporal stability and spatial comparability compared to the PDSI. As a result, the sc-PDSI is increasingly used in large-scale drought monitoring programs [86,87] and in assessments of climate impacts on socio-ecological systems. Finally, drought metrics confirmed important differences between the SPI-9 and the sc-PDSI. While SPI-9 captures short-term to seasonal precipitation deficits, the sc-PDSI reflects multi-month soil moisture anomalies and evapotranspiration processes [34]. This is evidenced in this study by the relatively low correlation between the two indices for the driest pilot meteorological station Mali Lošinj and the relatively higher correlation coefficients for the pilot station Ličko Lešće (Figure A5). Such evidence supports previous observations that the agreement between these two indices decreases in drier locations due to different hydrological sensitivities [69,70].

The clear increase in the dry category of the sc-PDSI in the months after 1993 at most pilot stations supports the claim that warming-induced evapotranspiration contributes to increased hydrological stress even in the absence of strong precipitation reductions. This pattern has also been documented in many European drought studies [29,81,82,83]. Overall, for assessing the vulnerability of tourist destinations to drought, the sc-PDSI offers several advantages: (1) Its self-calibration allows for meaningful comparisons across climatically diverse destinations, (2) its explicit inclusion of temperature effects indicates tourist comfort and water stress, and (3) its standardized output facilitates communication with non-technical stakeholders in the tourism sector.

The strong convergence between quantitative climate indicators and stakeholder perceptions across all destinations validates both data sources and demonstrates that experiential knowledge effectively translates climate trends into operational realities. The spatial coherence of results across Croatia’s topographically complex and climatically diverse regions reinforces the robustness of detected climate signals. These findings underscore the value of integrating quantitative analysis with stakeholder narratives, enabling assessments that are simultaneously scientifically rigorous and contextually relevant. This integrated approach validates regional climate trend assessments while highlighting the place-specific vulnerabilities of climate-dependent tourism, thereby enabling more targeted rather than generalized adaptation strategies. Global warming, hydrological stress and the increasing occurrence of related climate hazards emphasize the necessity for climate adaptation strategies not only for continental and coastal tourism but also for sectors that accompany it, like water resource management, agriculture, and nature protection.

5.2. Building Climate-Resilient Tourism: Evidence-Based Planning and Community Engagement

For selected tourist destinations, climate change creates operational challenges such as increased risk of heat waves, flooding due to heavy rainfall, reduced outdoor activities and increased pressure on water management systems during extended summer dry spells. These challenges can reshape tourism seasonality, infrastructure and supply. Nevertheless, stakeholders have identified new opportunities associated with warmer spring and autumn periods, which are becoming increasingly attractive for outdoor activities, wellness and cultural tourism. Similar shifts toward season extension are reported across southern Europe, driven by warming shoulder seasons and changing tourist preferences [6,80]. Yet, stakeholders underscored that capitalizing on these opportunities requires improved water management, adaptation of outdoor infrastructure to intense autumn rainfall, and clear communication of climate risks. Overall, while climate impacts vary across environmental and tourism contexts, Croatian destinations share a set of interlinked vulnerabilities that require coherent, multilevel adaptation strategies.

This finding emphasizes the call for adaptation opportunities, such as leveraging more favorable conditions in shoulder seasons, developing flood-resilient visitor facilities and heat-resilient infrastructure, and diversifying tourism seasonally and tourism offers to maintain safety, comfort, and economic stability [88]. Engaging local communities and tourism stakeholders in developing evidence-based adaptation strategies can take place at any stage of the process. This engagement could begin from the initial step the creation of potential impact scenarios (so called impact chain) through to the identification of key factors related to exposure, sensitivity, and adaptive capacity that influence vulnerability. As Zovko et al. stated [3], when conducting a vulnerability and climate risk assessment according to climatological data, involving stakeholders could be essential for shaping assessments tailored to specific locations or areas of interest. Consequently, by considering evidence based on climate data, their perceptions shape support for adaptation measures and informal actions [37].

In this research, community engagement is present in all destinations but tends to be uneven and insufficiently integrated. Adaptation measures are expanding, with coastal and island regions showing the strongest momentum due to the immediacy of their climate risks. Together, these findings demonstrate that climate resilience in Croatian tourism is largely location-specific and requires coordinated, multi-sectoral adaptation strategies tailored to each destination’s hazard profile, community capacities and tourism development pathways.

5.3. Value of Mixed-Methods Triangulation for Climate–Tourism Assessment

This study demonstrates that combining quantitative climate data with qualitative stakeholder narratives offers a scientifically robust and operationally relevant understanding of climate risks for tourism destinations. The integration of these two knowledge streams follows established principles of mixed-methods triangulation (Figure 2), which enhances analytical validity, improves contextual interpretation, and supports rapid and evidence-based adaptation planning [89,90].

Many scholars of different expertise [42,43,44,46,62,89,90] utilize a combination of data derived from quantitative observations alongside data gathered through participatory methods (discussions, interviews, workshops, and focus groups). The model presented in this research illustrates how climate related data and stakeholder narratives were systematically compared to derive integrated climate-tourism insights. This approach is consistent with climate-service and socio-ecological system frameworks promoting co-production and multi-scalar validation of climate risk information [1,25,64].

This study also identifies cases where climate hazards appear in the data but are not recognized by stakeholders, or where stakeholders report emerging challenges that may not yet be visible in long-term climate indices. It should be emphasized that a participatory process introduces transparency into the subjective definition of climate challenges, thereby reducing potential sources of conflict and increasing the acceptance of final adaptation outcomes. Engaging in the local community early in the process fosters a sense of shared ownership and enhances the acceptance and implementation of the findings.

If the aim is to achieve objective and verifiable results, prioritizing quantitative models is essential. On the other hand, if the objective is to enhance awareness or pinpoint adaptation priorities for specific areas of interest, participatory approaches should be favored, as they are regarded as more effective [25]. However, the combination of quantitative and qualitative data regarding climate and tourism strengthens the validity of the objective and subjective findings and ensures that adaptation recommendations are based on both empirical climate data and the lived experiences of tourism practitioners.

From an operational standpoint, this research identifies several systemic impediments: resource constraints, fragmented governance structures, truncated planning cycles, and inadequate adaptive responses. Where time and financial resources are limited for long-term and extensive measurements and analyses of climate parameters, community engagement could be an efficient way to recognize and quantify climate impacts [91,92]. This approach is often the case at a very local level, which is rarely covered by detailed statistical data and where climatic and hydrological characteristics are too specific to be captured by modeling. However, such information could be highly subjective. Moreover, it is difficult to replicate and has limited precision and spatial resolution [93]. Some researchers have indicated that utilizing local data, considering relevant contexts, and employing clear language were critical for building trust with participants [94,95]. Bormann et al. [96] discovered that participants placed greater trust in local knowledge compared to scientific knowledge and chose to engage in discussions with a relatively short time frame.

Finally, this study demonstrates that synthesizing climatic evidence with stakeholder narratives establishes a robust foundation for scientifically sound, timely, and contextually appropriate adaptation policy, consistent with established climate resilience frameworks [1,25,73]. However, three key advantages emerge:

• Faster risk identification: quantitative data and indices efficiently flag emergent climatic hazards (e.g., rising drought frequency), while stakeholder narratives prioritize those affecting tourism operations most urgently
• Higher policy relevance and feasibility: local narratives identify operational constraints, acceptable measures, and behavioral trends, elements that are essential for designing implementable adaptation strategies
• Support for iterative adaptive management: the combined use of climate indices and stakeholder observations provides the feedback structure necessary for monitoring and adjusting adaptation measures over time.

6. Conclusions

This study provides a comprehensive integration of long-term climate data (1961–2024) with systematic stakeholder narratives in selected inland, coastal, and island Croatian tourism regions. The convergence between 64 years of controlled meteorological data and independent stakeholder observations shows that climate change is actively reshaping Croatian tourism. The results showed that Croatian destinations face challenges, which are both universal (all regions experience warming and hydrological stress) and deeply differentiated (impacts depend on destination-specific geography, resource dependence, and adaptive capacity). This study demonstrates that effective adaptation requires the integration of quantitative precision and qualitative context. The convergence between long-term climate data and stakeholder narratives confirmed the hypothesis that the observed trends represent real signals with tangible operational impacts. This study also highlighted that the adaptation pathway needs to be scientifically sound, operationally relevant, and politically robust.

In response to research question 1 of how long-term trends in temperature, precipitation, and drought indices are evolving at selected Croatian destinations, this study revealed universal and statistically significant warming trends in all pilot regions, with the strongest implications in coastal and island destinations. Annual precipitation showed stable or insignificant trends in all regions, but the sc-PDSI drought indices revealed fundamental regime shifts from historically wet to increasingly dry conditions and hydrological transformation in all selected destinations. The findings of this study indicate that temperature-driven evapotranspiration increasingly dominates water stress in warmer coastal and island destinations, overriding stable precipitation inputs.

We provided several responses to research question 2, which is related to tourism worker narratives. Stakeholders accurately perceived warming trends, with acute thermal discomfort, increased demands on cooling infrastructure, and changed visitor behavior patterns. Perceptions of water stress also align with the sc-PDSI results, confirming that water balance indices better capture tourism-relevant stress than precipitation-only measures. Differences between climate data and stakeholder narratives revealed quantitative data limitations rather than stakeholder errors. For instance, annual precipitation aggregates fail to capture the dynamics of extreme weather events that cause infrastructure damage, and regional meteorological stations miss local hydrological sensitivities. Second, systematic triangulation revealed strong convergence between quantitative climate indicators and stakeholder perceptions across all pilot regions, providing cross-validation of both data sources.

Our response to research question 3, which concerns the impacts of climate change on tourist behavior, operational practices, and adaptation strategies at different types of destinations, pointed to changing tourist patterns (Table 4) and the efforts of tourism practitioners to mitigate unwanted behavioral changes. To adapt to climate signals and provide an adaptive response, destinations are facing a transformation of tourism. For instance, the continental Međimurje County is turning to tourism products (wellness and gastronomy) that are not directly affected by climate change. Such strategies enable “tacit” adaptation without acute operational disruptions. Overall, the findings suggest that climate change is increasingly acting as a catalyst for rethinking tourism development pathways.

The empirical contribution of this study lies in establishing a link between statistically validated climate trends and proven impacts reported by tourism practitioners in representative Croatian destinations. All pilot regions show significant warming, with the strongest signals in coastal and island locations. The analysis reveals a hydrological paradox: stable or insignificant precipitation trends mask a dramatic increase in drought frequency, with an increase in dry months according to sc-PDSI across the regions. This indicates that temperature-driven evapotranspiration, rather than precipitation reduction, is causing hydrological stress. Furthermore, a systematic comparison of the SPI-9 and sc-PDSI drought indices provides empirical evidence that water balance indices that include temperature effects better match perceptions of stress than indices that rely only on observations of precipitation trends. These results ensure evidence-based selection of adaptation measures, some of which may be highly technologically and financially demanding, like infrastructure for sustainable water use, ecosystem protection measures, including measures related to the health of residents and visitors.

Theoretical contributions are reflected in the differentiation of vulnerability and the use of valuable knowledge of stakeholders from different tourism regions. This study therefore advances climate tourism theory through two main contributions. First, the differential vulnerability framework shows that the vulnerability of a destination is a function of climate exposure, the sensitivity of tourism to climate-dependent resources and adaptive capacity. For instance, water-dependent heritage sites (Plitvice) face potentially transformative impacts despite moderate exposure to climate change, while coastal (Pakoštane) and island destinations (Mali Lošinj and Krk) face extreme changes in the context of their geographical constraints (limited freshwater resources, high infrastructure costs). Coastal lake ecosystems experience changes in the water regime and face a potential problem of salinization, which threatens the identity of the destinations. Diversified continental destinations (Sveti Martin na Muri, Varaždin) mitigate hydrological stress and high air temperatures through alternative products, such as wellness, culture and gastronomy.

This practical framework challenges previous research that has analyzed the impact of regional climate patterns on tourism, treating destinations as uniformly vulnerable. Instead, the proposed multidimensional framework of this study suggests that tourism typology and existing adaptive capacities (e.g., transformation potential and sustainability of resource use) play a significant role in adaptation planning. Second, the study reframes the relationship between quantitative climate data and qualitative stakeholder knowledge, which have been shown to be complementary rather than hierarchical. Convergences provide mutual validation because both knowledge systems capture real climate signals. Furthermore, divergences reveal the limitations of quantitative data. For example, regional stations miss local hydrological sensitivity, and indices based solely on precipitation trends underestimate water stress caused by evapotranspiration. The identification of “tacit” adaptation practices that emerge from experiential learning demonstrates that tourism professionals possess valuable adaptive knowledge that adaptation programs should document and reinforce, rather than replace with external expertise. Therefore, it could be said that such a theoretical contribution positions this research as a transdisciplinary integration of physical and social sciences, and not as a mere application of climate projections in the tourism context.

The practical value of the study lies in providing baseline climate evidence for specific destination with robust empirical justification. Investments in water infrastructure in island destinations are justified based on observed increases in dry spells according to sc-PDSI criteria and stakeholder reports on seasonal constraints. Cooling infrastructure in coastal destinations is justified by observed temperature trends and stakeholder evidence that air conditioning has moved from a convenience to a competitive necessity. Ecosystem-based adaptation is supported by sc-PDSI evidence of hydrological stress and stakeholder observations of visible esthetic degradation. These evidence-based priorities enable tourism destinations and regional authorities to justify adaptation investments that may be politically contentious or financially significant. The study also identifies the need for climate projections and water balance indices, as well as the importance of local hydrological monitoring networks. These specifications guide the development of climate information systems relevant to tourism. Furthermore, this paper recognizes that effective adaptation should consider destination-specific geography, tourism typology, and its adaptive capacity, rather than the sole application of uniform climate and environmental related regulations. The work highlights the importance of differentiated adaptation strategies, such as ecosystem-based approaches, water infrastructure upgrades, product diversification, and transformative repositioning.

Incorporating these robust and repeatable mixed methods for triangulating quantitative and qualitative climate and tourism data and information could be important for future research on climate impacts on tourism. Primarily because tourism destinations around the world face obstacles in implementing adaptation strategies due to political sensitivities, stakeholder skepticism or resource constraints. On the other hand, future research priorities require expansion in several directions. Integrating climate projections with documented historical trends would allow for scenario-based adaptation planning. Furthermore, it would be desirable to expand the sample of respondents to include visitors and stakeholders of tourism sub-sectors. Also, a systematic comparison with other Mediterranean destinations (Spain, Italy, Greece) would avoid generalization. A longitudinal evaluation of adaptation effectiveness would assess whether documented practices reduce the climate vulnerability of destinations. Economic quantification of climate impacts would complement the operational challenges faced by tourism practitioners and investors.