Using System Dynamics Modelling to visualize the effects of resource management and policy interventions on biodiversity at a regional scale

Ecological Indicators 145 (2022) 109630 Available online 3 November 2022 1470-160X/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Short Note Using System Dynamics Modelling to visualize the effects of resource management and policy interventions on biodiversity at a regional scale Chrysi Laspidou a , * , Konstantinos Ziliaskopoulos b a Department of Civil Engineering, University of Thessaly, Greece b Department of Environmental Sciences, University of Thessaly, Greece ARTICLE INFO Keywords: Biodiversity Red List Index System Dynamics Model Policy Co-creation SDG15 ABSTRACT A methodology and a System Dynamics Model is constructed, using published data from the IUCN Red List Index database, that can be used to quantify the biodiversity status. The methodology is implemented for the Nestos River catchment in Greece. It is intended to help the authorities run scenarios of different interventions, addressing the specific problems and threats to the local ecosystem that each community might be facing. The methodology enables them to see and quantify biodiversity improvements as a result of these interventions. It compares and contrasts four specific threats to the ecosystem, identified from stakeholder consultation workshops, namely solid waste, agriculture, domestic wastewater and dams and water management/use. The effects of each one of the interventions to the species in the region and to the modified Red List Index overall are presented and compared, showcasing the dams as the most deleterious threat to the local ecosystem. Finally, an easy to use interface is developed and introduced to better connect the stakeholders with the scientific analysis and facilitate informed decision-making that could lead to smarter policy implementation. This would result insetting priorities for interventions and investments concerning improvement of the ecological status and biodiversity. 1. Introduction Global biodiversity is declining rapidly, largely due to climate and land use change. According to the Global Assessment Report on Biodiversity and Ecosystem Services by IPBES (https://ipbes.net/), there is overwhelming evidence that presents an ominous picture, since the health of ecosystems is deteriorating at an alarming rate. The effects of these changes on biodiversity responses differ among ecosystems and transnational regions, especially when they have important differences in species composition. Newbold et al. (2020) have identified the Mediterranean region as one that has sustained the strongest negative responses to both land-use and climate pressures. A critical step in taking action for halting biodiversity loss is understanding how biodiversity is changing—especially how human activities are impacting it—and quantifying how much it is changing. Even if what is impacting biodiversity is well established, a methodology to put this impact in some sort of measurable perspective for authorities is lacking. For example, it may be known that use of pesticides in agriculture affects biodiversity, but the extent of this damage is unknown. It is also difficult to quantify how much biodiversity will improve if every-one turns to organic farming. The proper solutions usually exist but will not be easily implemented if people are not convinced of the need for these solutions Furthermore, since biodiversity loss is directly or indirectly linked to human activities, it is the decisionand policymakers that need to regulate how people relate to their natural environment, through prioritizing their regulations, interventions, infrastructure, and resource use. Despite the gravity of the biodiversity loss crisis, local communities, which are best placed to make decisions about the protection of their habitats, remain ill-informed on the significance of safeguarding biodiversity, as it remains something that they cannot easily measure and see it as tangible. Even more so, local authorities have questions on how to measure biodiversity and how to assess current and future conditions, especially given the Water Framework Directive and other relevant directives such as the one on Urban Wastewater that authorities need to comply with and report on. It is important to develop metrics and indicators that synthesise and communicate the current status and future trends in biodiversity, especially when planning specific interventions. These metrics need to not target only scientists and engineers, but need to be visual, easy to comprehend and easy to communicate in layman’s terms, since interventions to halt biodiversity loss need to be co- * Corresponding author. E-mail address: [email protected] (C. Laspidou). Contents lists available at ScienceDirect Ecological Indicators journal homepage: www.elsevier.com/locate/ecolind https://doi.org/10.1016/j.ecolind.2022.109630 Received 21 April 2022; Received in revised form 10 October 2022; Accepted 29 October 2022 Ecological Indicators 145 (2022) 109630 2 designed and co-created with local stakeholders to be successful. Citizen science projects and other bottom-up approaches that involve citizens in assessing and improving the status of our environment (Fraisl et al., 2020) are important to raise awareness and keep the citizens engaged in safeguarding biodiversity. Therefore, the need for a metric that measures and translates the complexity of the problem in simple and clear terms is pressing. Despite their importance in decision making, only a few biodiversity indicators have been introduced and tested for their performance and behaviour (Rowland et al., 2020) and little effort has been made in making these indicators easy to comprehend and communicate with the public. One such biodiversity indicator is the Red List Index (RLI), which uses information from the IUCN Red List to keep track of trends in the projected overall risk of extinction of species. The Red List Index is used as an indicator for Sustainable Development Goal No. 15 “Life on Land”; it is Tier 1 indicator 15.5.1. An important advantage of the RLI is that it is widely known and recognized for the measurement of progress towards reaching SDG 15 on biodiversity. However, it has been criticized for its shortcomings and further improvements have been proposed in the literature (Butchart et al., 2004, 2007). Yet, various researchers have utilized the Red List Index metric as a means of quantifying biodiversity: O’Reilly et al. (2022) have based the development of a metric to assess the relative habitat use of species on the RLI; Irwin et al. (2022) uses the RLI to calculate global extinction risk footprints and associate them with consumption sectors, thus identifying the large contributors to extinction. Finally, Levin et al. (2022) used one RLI metric—the Extent of Occurrence—along with data from the Global Biodiversity Information Facility, to develop a rapid species occurrence assessment framework. The RLI is reportedat a national scale by IUCN. In this article, we develop a methodology and a model for calculating a modified RLI for a scale that is smaller than national—regional or local, customized to the specific needs, conditions, pressures, and/or threats of the region. Specifically, we demonstrate the methodology for the Nestos River basin in Greece, even though it could be applicable to any region. We then define specific pressures that are applicable to the region under study, and we include them in the model. New RLI values are obtained for a range of interventions that will mitigate the effects of the specific pressures that the ecosystem under study withstands. A visual interface is also constructed that allows stakeholders to experiment with different interventions and observe the effects they have on biodiversity through the RLI. 2. Materials and methods The work presented in this article takes stock of stakeholder consultations during a workshop in the frame of the Horizon2020 project NEXOGENESIS (https://nexogenesis.eu/), which focuses on a series of transboundary catchments. In this specific case study, the project models the Greece-Bulgaria transboundary Nestos River catchment (Fig. 1) under a nexus framework, in order to strengthen cooperation and participation of the two communities in the decision-making process regarding resource use. About thirty representatives of local government, businesses and the private sector, energy and water resource operators and managers, environmental organizations, NGOs, and academic/research institutes were among the workshop participants. The overarching aim of this series of workshops was the co-creation and codesign of policies targeting the effective management of water resources through the integrated management of the Water-Energy-FoodEcosystems (WEFE) nexus. The Water-Energy-Food (WEF) nexus interlinkages have been studied and explored before in the literature (Laspidou et al., 2019; Laspidou et al., 2020) and in various research projects (e.g. Horizon2020 project SIM4NEXUS—https://sim4nexus.eu/). However, the “ecosystems” element has been understudied and presented a challenge during stakeholder consultations, since it was difficult to assess and quantify for the stakeholders, proving that indeed the sector of ecosystems and biodiversity needs to have a more tangible quantification tool that the citizens can comprehend and relate to easily. With the conclusion of the first workshop, it was possible to conclude on the major pressures that the Nestos river basin ecosystem sustains. These can be grouped in four categories: (i) agriculture with the increased use of fertilizers, pesticides and herbicides; (ii) solid waste, since some of the upstream villages on the Bulgarian side do not have a solid waste management plan and large quantities of garbage end up in the river downstream; (iii) domestic wastewater, since some of the settlements do not have a wastewater network; (iv) hydropower dams that alter and degrade the ecosystem and the availability and/or fluctuation of adequate ecological flow. We obtained data from the IUCN Red List Index database (https://iucnredlist.org/) for mainland Greece. Each species is listed in the data base with its Red List Category with the following scores (Butchart et al., 2004): 5—Extinct (EX), 4—Critically Endangered (CE), 3—Endangered (EN), 2—Vulnerable (VU), 1—Near Threatened (NT), and 0—Least Concern (LC). The list of relevant species was identified on the basis of the habitat of interest. In the case of the Nestos catchment, relevant habitats were “5.1-Wetlands (inland) - Permanent Rivers/ Streams/Creeks (includes waterfalls)” and “5.13-Wetlands (inland) - Permanent Inland Deltas”. This initial list contained 305 species. We then proceeded to further refine the list, by cross-mapping the species on our initial list with the four types of threats that we identified as important from the stakeholder workshop. In the IUCN database, the species are listed along with the corresponding threats that could potentially pressure them, and we identified the threats from the database that correspond to the ones pinpointed by the stakeholders. In order to create the tool that will allow the stakeholders to comprehend the interlinkages between their activities related to resource management, food production and biodiversity, we develop a System Dynamics Model with STELLA software (https://www. iseesystems.com/), a high-level visual-oriented programming and simulation language. The model is developed in a generic format and can be applied to any case study and any relevant habitat, as long as the specific methodology is followed and sufficient data is available. A description of the model is provided herein. First of all, the data is implemented in the model, as an array converter and the vulnerability of each species to each of the four threats is represented by a binary parameter, where 0 denotes no vulnerability and 1 indicates that the species is vulnerable to the threat. A modified RLI is calculated for the region, including the 52 species and associated RL Categories listed in Table 1, using the formula of Butchart et al. Fig. 1. Geographical depiction of the case study region. C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 3 (2007). Each threat is considered to have the same weight in terms of its effect on the species. The four threats enter the model as “interventions” by the authorities to improve conditions on the ground. Each intervention is modeled as an exogenous variable ranging from 0 to 1, where 0 denotes no improvement from the current situation and 1 denotes the complete elimination of the threat from the case study region. This is defined for the stakeholders in operational terms that relate to where the funding might be directed towards. Each species RL category score is then improved by a percentage of its overall threat improvement, using a simple algorithm. SRLCi=RLCi ×(1−WWVi×WWI +SWVi×SWI +AgrVi×AgrI +DVi×DI WWVi+SWVi+AgrVi+DVi+Otheri) Where, (S)RLC is the (Simulated) Red List Category Score of species i, WWV i , SWV i , AgrV i , and DV i stand for the vulnerability of species i to each of the four threats analyzed (wastewater, solid waste, agriculture and dams, respectively), while OtherV i is the number of threats of species i to other threats beyond these four. Vulnerability, as explained before is a binary parameter and takes values of either 0 or 1. Finally, WWI, SWI, AgrI and DI denote the level of intervention in each of the four threat categories; as explained before, these exogenous variables range from 0 to 1.Having calculated SRLC i for all species, we can then calculate a new simulated RLI (SRLI) for the region using the same RLI equation (Butchart et al., 2007). To show the results of the model, we ran the simulation five times, once with each intervention equaling 1 and the other three interventions equaling 0 and once with all interventions equaling 1. In the former case, we explore the full potential of each intervention separately to the biodiversity of the region. In the latter case, we show the ideal scenario, when all four important pressures in the region are eliminated. The nature of the RLI is such that even minute numerical changes can signify a shift in ecosystem condition: For a frame of reference, according to the published list of IUCN Red List Indices for 2021, the difference from the top-score country to the country that ranks 40th in Table 1 Final list of selected resident species in the Nestos River basin Case Study. Species RL Category Dams Wastewater Agriculture Solid Waste Other Threats Azure Bluet 0 x x N/A Balkan emerald 0 x x x N/A Balkan Goldenring 1 x x x 5 Banded Darter 0 x N/A Banded Demoiselle 0 x x N/A Beautiful Demoiselle 0 x N/A Blue chaser 0 x x N/A Blue Featherleg 0 x N/A Braune Sumpfschnecke 0 x N/A Bulgarian Emerald 2 x x x 4 Caliaeschna microstigma 0 x N/A Common Clubtail 0 x x x N/A Common Darter 0 x N/A Dainty Bluet 0 x x N/A Desmoulin’s Whorl Snail 2 x x 5 Duck mussel 0 x x N/A Eastern willow Spreadwing 0 x N/A Ebner’s Cone head 3 x 2 Fagotia daudebartii 0 x N/A Fen pondweed 0 x N/A Few flowered spike rush 0 x x N/A Flowering rush 0 x N/A Greek Stream Frog 0 x x x N/A Green Eyed Hooktail 0 x x N/A Green Gomphid 0 x N/A Hairy Hawker 0 x N/A Heart Shaped Lip Dactylorhiza 0 x x N/A Keeled Skimmer 0 x N/A Large Redeye 0 x N/A Lesser Water plantain 1 x 2 Marsh Fern 0 x x N/A Odalisque 0 x N/A Opposite leaved pondleaf 0 x x N/A Ornate Bluet 0 x x N/A Petite Massette 0 x N/A Pisidium milium 0 x x N/A Potamon fluviatile 1 x x x 5 Potamon ibericum 1 x x x 3 Radomaniola curta 0 x x x x N/A Ruddy Darter 0 x N/A Salix xanthicola 2 x 2 Segmentina molytes 0 x N/A Segmentina servaini 0 x x N/A Slender tufted sedge 0 x N/A Sombre Goldenring 1 x x x 5 Southern Skimmer 0 x N/A Swan Mussel 0 x N/A Theodoxus varius 0 x x N/A Turkish Goldenring 0 x N/A Turkish Red Damsel 3 x 6 Waterwheel 3 x x x 7 Yellow Spotted Emerald 0 x N/A C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 4 the list is only 0.1. In order to be able to better visualize the improvements of the RLI as a result of interventions to the stakeholders, we normalize both the current and the simulated modified RLI scores. This is only done for visualization purposes and does not alter the value of the simulated RLI score. An overview of the methodology in a step by step fashion is shown in Fig. 2a, while the model as it was done in STELLA software is shown in Fig. 2b. 3. Results and discussion The data for the analysis are obtained from the IUCN Red List Index database. Having cross-mapped the species listed at the national level with the ones found in the geographical area under consideration, namely the Nestos River basin in Greece, we continue to further condense the list by cross-mapping species with the corresponding threats. These threats are cross-listed in the RLI database as impacting each species in the specific Nestos River habitat. In this case study, the relevant threats have been identified by the stakeholders during the consultation workshops. These threats are: 9.3 - Agricultural & forestry effluents; 9.4 - Garbage & solid waste; 9.1 - Domestic & urban wastewater; and 7.2 - Dams & water management/use. After this crossmapping, some of the initial species were eliminated, since they were not found sensitive to these four types of anthropogenic interventions. Out of the initial 305 species, the resulting list contained 84 species. For most of these species, there is associated information in the IUCN database about their geo-location; thus, only the species that are found in the Nestos River basin were selected. Therefore, after selecting only the geographically relevant species out of the list of 84 species, we obtained the final list of 52 species. In Table 1, we list the 52 species coming from the IUCN database, along with their Red List Category and we identify the ones that are indeed affected by one or more of the threats we consider (marked by ‘x’ in the corresponding cells in Table 1). In Fig. 3, the species with an RL category above 0 (Near Threatened categorization and above) and their corresponding threats from the four Fig. 2. An overview of the (a) step by step methodology and (b) System Dynamics Model. C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 5 Fig. 3. Near threatened and above species in the case study region and their corresponding threats. Note that only the four simulated threats identified by the stakeholders are included. Individual threat bars (Wastewater, Solid Waste, Agriculture, Dams) have equal height. Fig. 4. Results of five simulation runs analyzing the impact of all modeled interventions to the RLC per species. In the legend, the different interventions are shortened as “Int”, so “WasteWater Int” stands for Wastewater Intervention equal to 1, etc. The term “All Int” signifies that all intervention variables equal to 1, while “Red List Category” denotes the current situation with all intervention variables equal to 0 (no intervention). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 6 previously identified are depicted. That way, the vulnerability of each species to these four threats is visualized. In other words, species with the tallest bars (such as the Balkan Goldenring and the Waterwheel) are especially vulnerable to the simulated threats. Naturally, these species may be affected by several other threats (other than the four that we identified as relevant). In the last column of Table 1, we include the number of other threats that the specific species is affected by, always according to the IUCN database. For the species that are classified as LC (score of 0), no further improvement is possible in regards to their Red List Categorization; therefore no additional processing of these species is necessary (indicated as “N/A” in the last column of Table 1). The results of these simulations to the RLC per species are shown in Fig. 4. Please note that only the species with an RLC of 1 and up (NT to EN, since no EX or CE species were identified in the case study region) are included in Fig. 4, since no further RLC improvement is possible with species with an RLC score of 0. When comparing the effects of the interventions on individual species, we see that out of the four interventions analyzed, the one with the least potential biodiversity improvement is “solid waste”, since only two species improve their RLC, namely Potamon fluviatile and Desmoulin’s Whorl Snail. All other threat interventions improve more species. We can also see that the level of improvement of each species depends on the total number of threats they are influenced by. If they are influenced by fewer threats, then an improvement in one of these threats would confer Fig. 5. Comparison between normalized RLI (%) and SRLI (denoted as new RLI,%) for all interventions separately (shown in (a) through (d)) and simultaneously (shown in (e)). The computed modified RLI and SRLI are shown on top of each bar. Please note that the black bar (RLI %) is the baseline RLI % value of the case study region, i.e. all interventions set to 0 and it is repeated in each graph for comparison purposes. C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 7 a larger overall improvement to the species’ RLC. For example, while an intervention in Dams would result in an improvement to both the Waterwheel and Ebner’s Cone head, the latter’s RLC score would improve more overall, since it has fewer other threats in the ecosystem under study (as shown in Table 1, the number of Other threats for Waterwheel is 7, while for Ebner’s Cone head is 2). For a view of how each intervention impacts the overall ecosystem of the region, we can compare the RLI and the SRLI computed from equation 1 after each intervention. It should be noted that the current modified RLI of the case study region is 0.923, as computed by this methodology. For reference, the overall RLI for Greece in the year 2021, according to the published list by IUCN, is 0.83. This difference is expected since the case study region is mostly rural and includes a large part of protected designated NATURA areas. In Fig. 5, we compare the normalized RLI and SRLI for each of the five aforementioned simulations. This way, it is easy to compare which intervention would potentially have the greatest impact on the overall ecosystem, while also showing where the ecosystem is most vulnerable—in this case, it would be the dams, since a largest improvement with an intervention also implies the largest potential harm with further deterioration of current practices. As for whether each of these interventions are realistic or feasible to reach the value of 1, or 100 % elimination of the threat in the ecosystem, it depends on each intervention. For example, wastewater creates problems in the region, because not all settlements are equipped with wastewater networks, but they operate with septic tanks. If the authorities decide on an intervention that establishes a wastewater network with an adjacent wastewater treatment plant for 100 % of the population that is currently without a network, then the variable “wastewater intervention” will achieve the value of 1. If only a fraction of the population is serviced, then the variable will take the corresponding fractional value. A similar reasoning is followed with the other 3 threats, even though some of them may not be so straight forward to quantify. For some threats, such as agricultural and forestry effluents, the value of 1 might not be realistic to achieve, since this would mean halting any agricultural practices in the region. The value of the tool is nevertheless important, since the stakeholders can see the trends of ecosystem improvement as a result of their interventions, simply and concisely. The interface developed has been uploaded to the iseesystems website and can be accessed publicly through the following link. In Fig. 6 we see the layout of the interface; it allows the user to vary the level of each intervention in a scale of 0 to 1 and then to run the model and see the corresponding % normalized SRLI each time: https://exch ange.iseesystems.com/public/chrysi-laspidou/red-list-index-calculatio n-for-nestos-rbd-greece. 4. Conclusions After identifying the need for the quantification of biodiversity and consulting with stakeholders to identify the important threats to their case study region, a systems dynamics model was developed in STELLA software utilizing the IUCN database and the red list index as methods of assessing ecosystem status and biodiversity. This article develops a methodology that enables regional authorities and local stakeholders to link biodiversity status with anthropogenic pressures and ecosystem threats and to simulate potential improvements to the biodiversity status after interventions through the red list index. Using a simple algorithm, potential interventions in each identified threat were simulated on the health of each species and on the overall ecosystem. The results distinguished the most vulnerable species and the most influential threat to the local community and ecosystem. An interface was developed in order to better communicate this information with stakeholders and facilitate a smart and informed decision-making process. This digital tool forms the basis for a social innovation in the context of co-creation and co-design workshops conducted with stakeholders, giving them the opportunity to “play” with different scenarios and lead to informed decision-making. CRediT authorship contribution statement Chrysi Laspidou: Conceptualization, Methodology, Writing – original draft, Writing – review & editing, Project administration, Funding acquisition. Konstantinos Ziliaskopoulos: Methodology, Software, Data curation, Visualization, Writing – original draft. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Data availability We have used publicly available data Acknowledgement The work described in this paper has been conducted within the Fig. 6. A showcase of the developed interface suitable for stakeholder use. C. Laspidou and K. Ziliaskopoulos Ecological Indicators 145 (2022) 109630 8 project NEXOGENESIS. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement No. 1010003881 NEXOGENESIS. This paper and the content included in it do not represent the opinion of the European Union, and the European Union is not responsible for any use that might be made of its content. References Butchart, S.H., Resit Akçakaya, H., Chanson, J., Baillie, J.E., Collen, B., Quader, S., Turner, W.R., Amin, R., Stuart, S.N., Hilton-Taylor, C., 2007. Improvements to the red list index. PLoS One 2 (1), e140. Butchart, S.H.M., Stattersfield, A.J., Bennun, L.A., Shutes, S.M., Akçakaya, H.R., Baillie, J.E.M., Stuart, S.N., Hilton-Taylor, C., Mace, G.M., Reid, W.V., 2004. Measuring global trends in the status of biodiversity: Red List Indices for birds. PLoS Biol. 2 (12), e383. Fraisl, D., Campbell, J., See, L., Wehn, U., Wardlaw, J., Gold, M., Moorthy, I., Arias, R., Piera, J., Oliver, J.L., Mas´ o, J., 2020. Mapping citizen science contributions to the UN sustainable development goals. Sustain. Sci. 15 (6), 1735–1751. https://doi.org/ 10.1007/s11625-020-00833-7. Irwin, A., Geschke, A., Brooks, T.M., Siikamaki, J., Mair, L., Strassburg, B.B., 2022. Quantifying and categorising national extinction-risk footprints. Sci. Rep. 12 (1), 1–10. https://doi.org/10.1038/s41598-022-09827-0. Laspidou, C.S., Mellios, N., Kofinas, D., 2019. Towards Ranking the Water–Energy–Food–Land Use-Climate Nexus Interlinkages for Building a Nexus Conceptual Model with a Heuristic Algorithm. Water 11, 306. https://doi.org/ 10.3390/w11020306. Laspidou, C.S., Mellios, N.K., Spyropoulou, A.E., Kofinas, D.T., Papadopoulou, M.P., 2020. Systems thinking on the resource nexus: Modeling and visualisation tools to identify critical interlinkages for resilient and sustainable societies and institutions. Sci. Total Environ. 717, 137264 https://doi.org/10.1016/j.scitotenv.2020.137264. Levin, M.O., Meek, J.B., Boom, B., Kross, S.M., Eskew, E.A., 2022. Using publicly available data to conduct rapid assessments of extinction risk. Conserv. Sci. Pract. 4 (3), e12628. Newbold, T., Oppenheimer, P., Etard, A., Williams, J.J., 2020. Tropical and Mediterranean biodiversity is disproportionately sensitive to land-use and climate change. Nat. Ecol. Evol. 4 (12), 1630–1638. https://doi.org/10.1038/s41559-02001303-0. O’Reilly, E., Gregory, R.D., Aunins, A., Brotons, L., Chodkiewicz, T., Escandell, V., Foppen, R.P.B., Gamero, A., Herrando, S., Jiguet, F., Kålås, J.A., Kamp, J., Klvaˇ nov´ a, A., Lehikoinen, A., Lindstr¨ om, Å., Massimino, D., Jostein, Ø.I., Reif, J., ˇ Silarov´ a, E., Teufelbauer, N., Trautmann, S., van Turnhout, C., Vikstrøm, T., Voˇ ríˇ sek, P., Butler, S.J., 2022. An assessment of relative habitat use as a metric for species’ habitat association and degree of specialization. Ecol. Ind. 135, 108521 https://doi.org/10.1016/j.ecolind.2021.108521. Rowland, J.A., Lee, C.K.F., Bland, L.M., Nicholson, E., 2020. Testing the performance of ecosystem indices for biodiversity monitoring. Ecol. Ind. 116, 106453 https://doi. org/10.1016/j.ecolind.2020.106453. C. Laspidou and K. Ziliaskopoulos