Capture and retain heavy rainfalls in Jordan

Capture and retain heavy rainfalls in Jordan Findings of a transdisciplinary German-Jordanian research project Presented by the CapTain Rain project team Editors: Katja Brinkmann, Dörte Ziegler Contributing authors: Ahmad Awad, Ahmad Bariq Allemyar, Christina Maus, Clara Hohmann, Daniel Schuhmann-Hindenberg, Dörte Ziegler, Hanna Leberke, Katja Brinkmann, Linnéa Fölster, Lothar Fuchs, Markus Rauchecker, Martina Winker, Michael Thiemann and Peter Hoffmann Final report 2 Acknowledgments We would like to thank our Jordanian research partners for their valuable contributions to, and excellent cooperation in, our research. The Federal Ministry of Research, Technology and Space (BMFTR) funded the project CapTain Rain as part of the funding measure “CLIENT II – International Partnerships for Sustainable Innovation” in the context of the frame-work program “Research for Sustainable Development” (FONA) under the funding code 01LZ2006A-F. The responsibility for the content of this publication lies with the authors. Imprint Publisher Institute for Social-Ecological Research (ISOE) Hamburger Allee 45 60486 Frankfurt am Main, Germany www.isoe.de The PDF version is freely available at www.isoe.de/en/research-and-teaching/publications (Open Access) CC BY-SA 4.0 international Suggested Citation: Brinkmann, Katja, Dörte Ziegler (Eds.) (2025): Capture and retain heavy rainfalls in Jordan. Findings of a transdisciplinary German-Jordanian research project. Frankfurt am Main. DOI: 10.5281/zenodo.16894182 Funding reference: Federal Ministry of Research, Technology and Space (BMFTR), 01LZ2006A-F. October 2025, Frankfurt am Main Project partners 3 Contents List of figures ..........................................................................................................................................4 List of tables ...........................................................................................................................................6 Frequently used abbreviations .............................................................................................................7 Summary .................................................................................................................................................9 Zusammenfassung .................................................................................................................................9 1 Introduction.................................................................................................................................... 10 2 Objectives and project structure ................................................................................................. 11 2.1 Relevance and objectives ........................................................................................................ 11 2.2 Project structure ....................................................................................................................... 12 3 Stakeholder dialogue and transdisciplinary integration ........................................................... 14 3.1 Research objectives and methods ........................................................................................... 14 3.2 Key findings .............................................................................................................................. 14 3.3 Outlook: Stakeholder dialogue and transdisciplinary integration ............................................. 18 4 Rainfall hazard risk: Analysis of heavy rainfall scenarios in Jordan ...................................... 19 4.1 Research objectives and methods ........................................................................................... 19 4.2 Key findings for Amman and Wadi Musa ................................................................................. 19 4.3 Outlook: Rainfall hazard risk .................................................................................................... 28 5 Exposure and Sensitivity: Flood hazards and risk areas ......................................................... 29 5.1 Research objectives and methods ........................................................................................... 29 5.2 Key findings for Amman and Wadi Musa ................................................................................. 29 5.3 Outlook: Flood hazards and risk areas .................................................................................... 37 6 Adaptive capacity: Selection and localisation of adaptation measures ................................. 39 6.1 Research objectives and methods ........................................................................................... 39 6.2 Key findings .............................................................................................................................. 40 6.3 Outlook: Selection and localisation of adaptation measures ................................................... 48 7 Adaptive capacity: Water and weather data portal for Jordan to improve early warning ..... 50 7.1 Research objectives and methods ........................................................................................... 50 7.2 Key findings .............................................................................................................................. 50 7.3 Outlook: Climate and water data portal for Jordan to improve early warning .......................... 65 8 Integrated vulnerability assessment ........................................................................................... 66 8.1 Research objectives and methods ........................................................................................... 66 8.2 Key findings for Amman ........................................................................................................... 68 8.3 Outlook: Vulnerability assessment ........................................................................................... 71 9 Integrated multi-scenario analysis .............................................................................................. 72 9.1 Key findings for Amman ........................................................................................................... 72 9.2 Development of multi-scenarios ............................................................................................... 73 9.3 Hydrological simulation of rainfall and land use scenarios ...................................................... 77 9.4 Hydraulic simulation of rainfall and measures scenarios ......................................................... 79 9.5 Outlook: Multi-scenario analysis .............................................................................................. 82 10 Key messages and further research need .................................................................................. 83 11 References ..................................................................................................................................... 86 Final report 4 List of figures Figure 1: Overview of the selected study areas in Jordan: The capital Amman as urban region ...... 12 Figure 2: Conceptual framework for the integrated vulnerability analysis of flash floods and associated work packages .................................................................................................. 13 Figure 3: Results of the stakeholder mapping.................................................................................... 15 Figure 4: Group work on planning goals at the second stakeholder workshop in 2023, preparation of the exhibition for the CapTain Rain project’s final event in 2024 ................ 16 Figure 5: Interrelations of structural challenges for flash flood management .................................... 18 Figure 6: Seasonal cycle of daily total and accumulated precipitation in Amman from 1961–2019 .. 20 Figure 7: Long-term trends of the annual mean/maximum daily precipitation in Jordan from 1961–2018 .......................................................................................................................... 21 Figure 8: Scheme of contextualization ............................................................................................... 21 Figure 9: Causal linkage between local extreme rainfall and the large-scale atmosphere circulation ............................................................................................................................ 22 Figure 10: Real-time rainfall estimates from GSMaP satellite product ................................................ 22 Figure 11: Heavy rainfall maps in a design of a Radar ........................................................................ 23 Figure 12: Overview of the available high-resolution climate scenarios for the Euro-Cordex-11 domain having 3-hourly precipitation .................................................................................. 23 Figure 13: Derived heavy rainfall maps for past and future conditions ................................................ 24 Figure 14: Sensitivity of heavy rainfall scenarios 3-hr to 24-hr under past and future climate conditions ............................................................................................................................ 24 Figure 15: Screenshots of the climate service portal for Jordan .......................................................... 25 Figure 16: Long-term monthly mean weather-type characteristics for days above 1 mm and days above 30°C in Jordan ......................................................................................................... 26 Figure 17: Scheme of data processing for training and predicting weather-types in weather forecasts. ............................................................................................................................. 27 Figure 18: Resulting maps, tables and charts of the early warning tool for critical weather-types in Jordan ............................................................................................................................. 27 Figure 19: Flash flood risk assessment approach within CapTain Rain. Risk is the overlay of hazard and damage potential. ............................................................................................ 29 Figure 20: Overview of the study areas Amman und Wadi Musa with the hydrological catchment area and the selected model boundary and the flow paths ................................................ 31 Figure 21: HEC-HMS modules used in the hydrological model setup of the CapTain Rain project .... 32 Figure 22: Hazard map for Downtown Amman modelled with HE2D/FOG2D with recorded rainfall from five stations during the February 2019 event (most intense period) ............... 33 Figure 23: Damage potential map for Downtown Amman, showing buildings with a damage potential class for flash floods without consideration of flooded areas ............................... 34 Figure 24: Flash flood risk map for Downtown Amman ....................................................................... 36 Figure 25: Simulated runoff at the Siq entrance, Wadi Musa with the models HEC-HMS, RRI and HE2D/FOG2D, as well as the mean input rainfall for the event of December 2022. .......... 37 Figure 26: Overview of the stepwise planning process comprising data collection & analysis, planning and localization, and the development & strategy................................................ 39 Figure 27: Multitouch table introduced at GAM in Amman and at the final workshop ......................... 40 Figure 28: Infocard example for the measure “Raingarden” ................................................................ 41 Figure 29: Analysis of landuse overlayed with flood prone map .......................................................... 42 Figure 30: Analysis of landuse overlayed with flood prone map (detailed view). ................................ 42 Figure 31: Location of the focus area within the catchment of downtown Amman .............................. 44 Figure 32: Overview of the different measures and their benefits regarding different planning goals . 45 Figure 33: Photo and examples of results from the workshop entitled “From planning objective to implementation of rainwater management measures” ........................................................ 46 5 Figure 34: Map of the focus area Marj Al Hamam showing flood risk zones and max water depth .... 47 Figure 35: Concept design for the Royal Village Project ..................................................................... 47 Figure 36: Conceptual design of measures in public areas, example secondary school .................... 48 Figure 37: Actors during flood emergencies in Amman and at the national scale ............................... 52 Figure 38: Functionalities covered by the WWDPJ as Part of the Overall Emergency Management Activities ........................................................................................................ 53 Figure 39: Historical Data Locations shown in the WWDPJ Map Interface ......................................... 54 Figure 40: Example Graph of Historical Data in the WWDPJ .............................................................. 54 Figure 41: Selection of Satellite Observations Products via the WWDPJ Interface ............................ 56 Figure 42: Animating a Selected Satellite Observations Product in the WWDPJ Interface ................. 56 Figure 43: PDTRA’s rainfall observations sites shown in the WWDPJ Interface ................................. 57 Figure 44: Displaying Rainfall Observations in the WWDPJ Interface ................................................. 57 Figure 45: Selecting the WWDPJ Rainfall Nowcast Product ............................................................... 58 Figure 46: Selecting Weather Forecast Products in the WWDPJ ........................................................ 58 Figure 47: Selecting a Parameter of a Weather Forecast Products in the WWDPJ ............................ 60 Figure 48: Animating a Selected Weather Forecast Parameter in the WWDPJ Interface ................... 60 Figure 49: Showing a Forecast of Rainfall at an Arbitrary Location in the WWDPJ Interface ............. 61 Figure 50: Forecasted Flow at Three Locations in Amman, shown in the WWDPJ Interface ............. 61 Figure 51: Example Sub-Basin Area for which Rainfall is Averaged in Amman .................................. 62 Figure 52: Area for which Rainfall is Averaged in the Petra Area ........................................................ 62 Figure 53: Depiction of Classified Forecasted Rainfall Averaged over a Sub-basin in the WWDPJ ... 63 Figure 54: History of Threshold Exceedances Shown in the WWDPJ Interface ................................. 63 Figure 55: Definition of Alarm Recipients in the WWDPJ .................................................................... 64 Figure 56: Definition of Alarm Recipient Groups in the WWDPJ ......................................................... 64 Figure 57: Distribution of vulnerability categories for Exposure + Sensitivity and Adaptive Capacity for the districts of the study catchment in Amman ............................................... 69 Figure 58: Combined vulnerability maps showing all domains for sensitivity and exposure and for adaptive capacity for the study area in Amman .................................................................. 70 Figure 59: Results of the Interviews with residents in Amman on local knowledge of flash floods and mitigation measures ..................................................................................................... 70 Figure 60: Overview of the different scenarios within CapTain Rain that have been simulated using hydrological and hydraulic models ............................................................................ 72 Figure 61: Sorted table of heavy rainfall events given as values and patterns for Jordan using different reanalysis products: UERRA, ERA5, W5E5. In addition, the respective largescale circulation patterns (Z500) are given. ........................................................................ 73 Figure 62: Baseline event in Amman on February 28th in 2019 represented by Satellite Rainfall Estimates (GSMaP, JAXA) ................................................................................................. 74 Figure 63: Land use and land cover datasets for our catchment area in Amman for the past (1968), present (2021) and future scenario (2050) ............................................................. 76 Figure 64: Modelled runoff curves at the catchment outlet near Downtown Amman from HECHMS for the heavy rainfall event of February 2019 and different land cover scenarios ..... 78 Figure 65: With HEC-HMS simulated rainfall scenarios (baseline, moderate, intense, and catastrophic) ........................................................................................................................ 79 Figure 66: Focus area Downtown Amman with a) inundation map of the baseline simulation, of b) moderate scenario, c) intense scenario, d) catastrophic scenario ................................. 80 Figure 67: Impact of the ‘public and private space scenario’ in combination with the intense rainfall scenario modelled with the hydraulic model HE2D/FOG2D. .................................. 82 Final report 6 List of tables Table 1: Overview of online-Trainings (webinars) conducted in spring 2022 on WP-specific research topics .................................................................................................................... 17 Table 2: Input data for the hydrological modelling in the study areas Amman and Wadi Musa respectively ......................................................................................................................... 32 Table 3: Flash flood risk matrix (combination of hazard and damage potential) .............................. 35 Table 4: Simulated water at the bridge near the Petra entrance in Wadi Musa for two different rainfall inputs ....................................................................................................................... 37 Table 5: Planning goal for each focus area and a pre-selection of measures .................................. 44 Table 6: Historical Data available in the WWDPJ ............................................................................. 55 Table 7: Satellite Data Products in the WWDPJ ............................................................................... 55 Table 8: Weather Forecast Products in the WWDPJ ........................................................................ 59 Table 9: Selected Indicators for the SEVA for Amman for each component (exposure, sensitivity and adaptive capacity) and domain (social, physical and ecological)................ 67 Table 10: Categorization of the adaptive capacity based on physical, social and ecological aspects ................................................................................................................................ 68 Table 11: Results of the AHP approach showing the relative importance of the different vulnerability indicators from the stakeholders’ perspective ................................................ 68 Table 12: Developed possible future extreme rainfall scenarios (baseline, moderate, intense, catastrophic) for the study areas Amman and Wadi Musa ................................................. 75 Table 13: Description of the selected measures scenarios ................................................................ 77 Table 14: Summary of the results of the measures scenarios for the focus area 1 – Marj Al Hamam ................................................................................................................... 81 7 Frequently used abbreviations AFD Agence Française de Développement AZESA Aqaba Special Economic Zone Authority BGI Blue-Green Infrastructure CSOs Civil society organizations DEM Digital Elevation Model EBRD European Bank for Reconstruction and Development EIB European Investment Bank EWS Early warning system GAM Greater Amman Municipality GIZ German Agency for International Cooperation GCF Green Climate Fund GGGI Global Green Growth Institute GJU German Jordanian University HEC-HMS Hydrologic Modeling System INWRDAM Inter-Islamic Network on Water Resources Development and Management IUCN-ROWA International Union for Conservation of Nature and Natural Resources – Regional Office for West Asia JEA Jordan Engineers Association JICA Japan International Cooperation Agency JMD Jordan Meteorological Department JU The University of Jordan JVA Jordan Valley Authority KfW KfW Development Bank MoA Ministry of Agriculture MoE Minister of Environment MoEd Ministry of Education MoF Ministry of Finance MoI Ministry of Interior MoLA Ministry of Local Administration MoPH Ministry of Public Works and Housing MoPI Ministry of Planning and International Cooperation MoWI Ministry of Water and Irrigation NARC National Agricultural Research Center NCSCM National Center For Security & Crisis Management PDTRA Petra Development Tourism Regional Authority RRI Rainfall-Runoff-Inundation model RJGC Royal Jordanian Geographical Center RSS Royal Scientific Society SDC Swiss Agency for Development and Cooperation US-AID U.S. Agency for International Development Final report 8 UNFCCC United Nations Framework Convention on Climate Change UNDP United Nations Development Programme UNICEF United Nations Children’s Fund UNHabitat United Nations Human Settlements Programme WAJ Water Authority of Jordan WFP World Food Programme WMO World Meteorological Organization. International cooperation entities WP Workpackages 9 Summary The Middle East is particularly affected by climate change and extreme weather events such as droughts and heavy rainfall. In Jordan, repeated heavy rainfall events in recent years have led to flash floods that have caused enormous damage. Minimising such damage, but also maximising the benefits of heavy rainfall through improved water retention in one of the world’s most water-scarce countries, was the research topic of the German-Jordanian project CapTain Rain (“Capture and retain heavy rainfalls in Jordan”; Duration: June 2021 – July 2024; website: www.captain-rain.de). The project was funded by the German Federal Ministry of Research, Technology and Space (BMFTR, former BMBF), as part of the funding measure “CLIENT II – International Partnerships for Sustainable Innovation” in the context of the framework program “Research for Sustainable Development” (FONA). The aim was to help improve current methods and tools for predicting flash floods and preventing of damage. The study areas included the capital Amman with its 4.3 million inhabitants in the metropolitan region and the more rural Wadi Musa region around the UNESCO World Heritage Site of Petra. Both regions have been severely affected by flash floods in the past. To this end, hydraulic and hydrological models were set up, along with a weather data portal. Vulnerability analyses were carried out, measures to reduce the damage caused by flash floods through the use of blue-green infrastructure were identified, and recommendations for urban planning and early warning systems were developed. Future scenarios were simulated for Amman and Wadi Musa to assess the impacts of climate change, urbanisation and how specific measures can help reduce vulnerability. CapTain Rain’s transdisciplinary research methods enabled a holistic analysis of flash flood hazards and facilitated the transfer of scientific knowledge into practical climate change adaptation measures. Climate services (e.g. flash flood risk maps, weather data portal, vulnerability assessment and maps) were developed in close collaboration with Jordanian partners. The report at hand describes the project results and the CapTain Rain climate service products, which have also been made available in a Wiki. Zusammenfassung Der Nahe Osten ist vom Klimawandel und extremen Wetterereignissen wie Dürren und starken Regenfällen besonders betroffen. In Jordanien haben wiederholte Starkregenereignisse in den letzten Jahren zu Sturzfluten geführt, die enorme Schäden verursacht haben. Solche Schäden zu minimieren, aber auch die Vorteile von Starkregen durch eine verbesserte Wasserrückhaltung in einem der wasserärmsten Länder der Welt zu maximieren, war das Forschungsthema des deutsch-jordanischen Projekts CapTain Rain (“Capture and retain heavy rainfalls in Jordan”; Laufzeit: Juni 2021 – Juli 2024; Website: www.captain-rain.de). Das Projekt wurde durch das Bundesministerium für Forschung, Technologie und Raumfahrt (BMFTR, ehemals BMBF), in der Fördermaßnahme „CLIENT II – Internationale Partnerschaften für nachhaltige Innovationen“ im Kontext des Rahmenprogramms Forschung für Nachhaltige Entwicklung (FONA) gefördert. Ziel war es, die aktuellen Methoden und Instrumente zur Vorhersage von Sturzfluten und zur Vermeidung von Schäden zu verbessern. Dabei wurden Starkregenereignisse und mögliche Klimawandeleffekte untersucht, die Auswirkungen von Sturzfluten auf Bevölkerung und Infrastruktur über hydraulische und hydrologische Modelle analysieren, sowie ein Wetterdatenportal eingerichtet. Es wurden Vulnerabilitätsanalysen durchgeführt, Maßnahmen zur Reduzierung von Schäden durch Sturzfluten durch den Einsatz von blau-grüner Infrastruktur ermittelt und Empfehlungen für die Stadtplanung und Frühwarnsysteme entwickelt. Für Amman und Wadi Musa wurden Zukunftsszenarien simuliert, um die Auswirkungen des Klimawandels und der Urbanisierung zu bewerten und zu ermitteln, wie spezifische Maßnahmen zur Verringerung der Vulnerabilität beitragen können. Die transdisziplinären Forschungsmethoden von CapTain Rain ermöglichten eine ganzheitliche Analyse von Lösungsoptionen und erleichterten den Transfer wissenschaftlicher Erkenntnisse in praktische Maßnahmen zur Anpassung an den Klimawandel. In enger Zusammenarbeit mit jordanischen Partnern wurden Klimadienstleistungen entwickelt (z. B. Risikokarten für Sturzfluten, Wetterdatenportal, Vulnerabilitätsbewertung und Karten). Der vorliegende Bericht beschreibt die Projektergebnisse und die CapTain Rain Klimadienstleistungsprodukte, die auch in einem Wiki ausführlich dargestellt werden. Final report 16 a participatory mapping approach in Google Earth. Available studies on flash floods were compiled for each study site, along with a list of knowledge gaps and required actions. The workshop fostered transdisciplinary collaboration and enabled the fine-tuning of project activities and products to the needs of stakeholders. The second stakeholder workshop, which took place on 30 January 2023 at the Geneva Hotel in Amman, was attended by 42 participants. The interim results were validated through a participatory process with the aim of refining the model results subsequently. The workshop also aimed to identify promising adaptation strategies for scenario development through a collaborative process. Alongside the presentation and discussion of the latest research results and the selection of focus areas, group work was conducted to discuss planning goals for reducing flash flood damage as well as potential measures in Amman and the early warning chain in Jordan. The third stakeholder workshop took place on 11 December 2023 at the Al-Hussein Cultural Centre in Amman. Around 30 project partners and Jordanian experts attended the event. Of particular significance were the discussions on scenarios and vulnerability assessment, and the consensus on completing and transferring the CapTain Rain products in collaboration with Jordanian partners. A further workshop was also held at the PDTRA office in Wadi Musa, focusing on the joint selection of planning objectives for the Wadi Musa region and the discussion of potential measures to reduce flash flood damage. The final event took place on 30 June 2024 at the Al-Hussein Cultural Centre in Amman and was attended by over 70 people. After the project results for the various work packages were presented and discussed, an exhibition of the CapTain Rain products was held. Participants gained a comprehensive understanding of the flash flood risk and vulnerability maps, and had the opportunity to engage with climate and water data portals and a digital touch table for urban planning. This was followed by a final event in Wadi Musa on 2 July 2024. A key focus of this event was exploring the potential for further use of CapTain Rain's products in successfully implementing climate change adaptation measures. Figure 4. Group work on planning goals at the second stakeholder workshop in 2023 (left), preparation of the exhibition for the CapTain Rain project’s final event in 2024 (right). Further dialogues and collaboration were held between the stakeholder workshops, including with key donors such as UN-Habitat, the Swiss Development Corporation, GIZ and KfW. 3.2.3 Capacity development To strengthen understanding of flash flood risk reduction in relation to the different work package aspects, online seminars (webinars) were held between 17 March and 8 June 2022 (see Table 1). The stakeholder workshops did not allow enough time to go into sufficient detail, whereas the online format enabled broader participation from both Germany and Jordan. During these capacity development events, the German project partners presented their research activities and discussed potential approaches with interested participants from Jordan. A total of five research topics from different work 17 packages were presented. In the first seminar, German company KISTERS AG presented a powerful data management tool for early warning systems. The second seminar featured itwh GmbH, who provided information on data requirements and possible approaches for analysing flood hazards. Together with Koblenz University of Applied Sciences, they gave an overview of hydrological and hydraulic models for flash flood risk assessment in the third seminar. A particular highlight of this online seminar was the presentation by Dr Qasem Abdelal of the German-Jordanian University, who shared insights into his research in the Wadi Musa catchment area. The fourth seminar was hosted by the Potsdam Institute for Climate Impact Research (PIK), who introduced predictors for heavy rainfall events in Jordan. The series concluded with the fifth online seminar, which was hosted by the German company Hamburg Wasser. They provided an overview of potential measures to mitigate flash flood damage and presented the Sponge City approach. Further training measures (both online and face-to-face) on individual products (e.g. hydraulic modelling, the water and weather data portal and the implementation of adaptation measures) took place in specific work packages in 2023 and 2024, and are explained in the relevant chapters. Table 1. Overview of online-Trainings (webinars) conducted in spring 2022 on WP-specific research topics. Date Topic Presenter 17.03 Features of the state-of-the-art Demonstrator for climatic variables KISTERS AG 11.04 Flood hazard analysis tools – Data needed and possible approaches itwh 16.05 Overview of hydrological and hydraulic models for flash flood risk assessment Koblenz University, GJU, itwh 30.05 Introduction to predictors for heavy rainfall events in Jordan under climate change PIK 08.06 Measures to reduce flash flood risks and examples from the Sponge City Hamburg Hamburg Wasser, ISOE 3.2.4 Structural challenges for flash flood management in Jordan To better integrate flash flood management into the governance structure and develop efficient management solutions, the structural challenges must first be identified and their interrelationships understood. The results of the expert interviews showed that the experts focused more on structural challenges relating to government agencies and international donors than on issues relating to local communities. The main challenge identified was the fragmentation and overlap of responsibilities caused by an increase in flash flood management activities. Stakeholders stated that there was no overarching strategy for dealing with flash floods and that many activities were reactive. A lack of funding for flash flood prevention measures was identified as a problem affecting not only government agencies, but also municipalities. Although flash flood prevention measures are mandatory for new buildings in Jordan, construction companies and the local population often do not comply because the measures are costly. Several experts mentioned the dependence on international donors and projects due to the lack of government funding for flash flood management. Flash flood management projects were considered unsustainable because government agencies did not continue them after completion due to a lack of funding and a lack of links between projects and government agencies. Furthermore, it was found that international donors had their own agendas and capacities when setting up projects. This exacerbates the existing fragmentation and overlap of responsibilities between government agencies. Stakeholders also mentioned that the proliferation of project activities not taken up by government agencies after projects end, coupled with the lack of local community participation in project design and decision-making, would lead to mistrust and result in low compliance, hindering efficient flash flood management. Figure 5 displays the interrelations of the structural challenges mentioned by the experts. The stakeholder analysis showed that decision-makers, the public and research institutions are well aware of the significance of flash floods and the risks they pose due to the frequency of heavy rainfall events. However, the analysis revealed a lack of effective strategies and tools for reducing risk and Final report 18 damage. Although improving flash flood forecasting and risk management is a high political priority in Jordan, it has not yet been sufficiently implemented in practice. The development and implementation of technical and social measures for flash flood prevention and disaster management depends not only on funding and the technical applicability of the measures in a specific flood-prone area, but also on political and administrative support, and on being anchored in the local community of the flood-prone area. The results of the stakeholder analysis demonstrate how structural challenges can hinder the effective management of flash flood risk and highlight the difficulties that research projects encounter when attempting to implement their findings. Figure 5. Interrelations of structural challenges for flash flood management (based on 15 expert interviews). The interviewed experts highlighted a number of measures that could be taken to address the structural challenges. Several of these measures are already partly implemented. To avoid duplication of work, an overarching strategy for flash flood management is needed to better coordinate projects and activities in this area. More recent flash flood management projects, as well as CapTain Rain, have implemented steering and technical committees to improve coordination between state entities and projects. It is important for projects to provide capacity-building opportunities for state officials (see Chapter 3.2.3) to ensure they take ownership of the project results and to strengthen their qualifications so they can build on these results. Very few projects have integrated local communities into their designs and decisionmaking processes. Some state entities provide incentives for local communities to implement flash flood measures. However, it is important to further strengthen the relationship between state entities and local communities, which can be achieved through projects. Furthermore, it is important to improve coordination between state entities, which is achieved for flash flood events by the NCSCM. However, coordination for proactive management is still needed. 3.3 Outlook: Stakeholder dialogue and transdisciplinary integration CapTain Rain’s transdisciplinary research approach integrated scientific as well as practical knowledge of German and Jordanian actors through stakeholder dialogues and contributed to the transfer of scientific results into practice. Examples of this transfer include ‘climate services’, including flash flood risk maps, tools to improve flash flood forecasting, and recommendations for promising adaptation strategies and early warning systems. CapTain Rain’s ongoing stakeholder dialogue and capacity building activities enabled the participatory design of climate services in an understandable and userfriendly form. Although CapTain Rain was a time-limited research project, the continuous stakeholder engagement enabled the results to be prepared for future practical use by local decision-makers, and significant knowledge transfer has already taken place during these interactions and phases of intensive collaboration. 19 4 Rainfall hazard risk: Analysis of heavy rainfall scenarios in Jordan Author: Peter Hoffmann 4.1 Research objectives and methods The objective of this work package was to gain further insight into the natural hazard of heavy rainfall in Jordan by analysing a number of aspects. The preliminary phase of the research involved the identification of critical large-scale weather patterns over the eastern Mediterranean that were identified as the underlying cause of localised heavy rainfall events in Jordan. This approach enables the causal relationship between the large-scale transport of air masses and local weather to be identified and used as a proxy. Based on this, a prototype early warning system was established and tested in collaboration with the Jordan Meteorological Department (JMD), which evaluates operational weather forecasts. In addition, real-time data from satellite-based rainfall estimates for Jordan have been processed and made available in a radar design. In the context of climate change, simulation data from global and regional climate models were then analysed from a number of perspectives. (a) How might the frequency of critical weather patterns change? (b How might heavy rainfall patterns in Jordan change in the future for different return periods? (c) Are there differences in the duration of rain events? (d) What plausible scenarios for possible heavy rainfall events can be derived and used for hydraulic simulations? To implement the results and current climate services, an existing portal was extended to include Jordan. 4.2 Key findings for Amman and Wadi Musa The following subchapters summarise and explain the main results and products of analysing heavy rainfall patterns and trends in Jordan. We start with some general climatological aspects concerning the seasonal distribution of precipitation and trends, followed by specific results. The main results of the heavy precipitation analysis can be summarised as follows:  A causal linkage between local heavy rainfall events in Amman and Petra and critical large-scale weather patterns over the eastern Mediterranean was demonstrated.  Regarding climate projections for the future until 2100, the long-term trends in annual mean precipitation is decreasing while the annual maximum precipitation is increasing in most parts of Jordan  In terms of extreme rainfall, there is little differentiation between Amman and Petra: heavy rainfall can occur anywhere in Jordan.  The heavy rainfall events in the eastern Mediterranean in 2023 (more specific: Libya and Greece) have demonstrated the considerable volume of water that can be discharged in a single day when low-pressure systems develop over the warm Mediterranean Sea. Similar events could also happen in Jordan in the context of climate change.  A detection and monitoring of the critical circulation patterns that may result in intense rainfall should be integrated in operational weather forecasts. This would enable experts to better assess predicted developments of possible extreme rainfall conditions in Jordan days in advance.  Open data of satellite rainfall estimates are suitable proxies to monitor approaching areas of rainfall in near real-time.  Convective and short-term rainfall intensities of several hours in Amman and Petra show a positive climate sensitivity in contrast to longer-term events of 12 hours and more.  Observed extreme events were defined as baseline scenario. Considering a progressive climate warming, such extreme events will be moderately increased by 15% to 20% (on average), however higher values cannot be excluded.  Analysis of climate model simulations showed robust decreasing in the number of extreme rainfall events in Jordan, however the potential for an intensification of single events would increase according to those simulations. Final report 20  A climate service portal was developed for Jordan. The climate service portal enables users to inform about possible future scenarios of different climate indicators (e.g. seasonal mean temperature, number of hot days, maximum precipitation, etc.). 4.2.1 Climatology of Precipitation in Jordan An overview of the seasonal distribution of rainfall events in the Greater Amman area was derived based on UERRA regional reanalysis data (Copernicus, 2019b). This was supplemented by analyses of longterm trends of mean and extreme precipitation. Open data were used due to the difficulty of accessing long-term meteorological measurements in Jordan. Although there may be inconsistencies at the level of individual events, the analyses provide a consistent overall picture at the national level. Figure 6. Seasonal cycle of daily total and accumulated precipitation in Amman from 1961-2019 extracted from the UERRA regional reanalysis data: 1961–1990 (blue) and 1991–2018 (red). The seasonal distribution of daily precipitation values is representative of the typical climatological pattern observed in the eastern Mediterranean region. In Amman, there is minimal precipitation during the summer months, spanning the period from June to September (Figure 6). Maximum values are within the range of 50–70 mm per day. Two distinct periods have been delineated by colour: 1961–1990 (blue) and 1991–2018 (red). The accumulated rainfall totals demonstrate a clear trend towards drier conditions in recent years, with the heaviest events represented by red bars. This is to some extent an expected consequence of climate change, as recently noted by Zittis et al. (2022) in their analysis of changes in extremes in the Mediterranean and Middle East. The region experiences a lower average precipitation level throughout the year; however, when precipitation does occur, it is at a higher intensity. This is supported by the trend analysis shown in Figure 7. Over most of Jordan, annual precipitation has decreased significantly in recent decades (negatively correlated with time). Conversely, there are predominantly increasing trends in annual maxima, particularly in the west, where the highest precipitation levels of about 300 mm per year are recorded. The large-scale weather patterns over the eastern Mediterranean and their changes have a significant influence on the trends. Therefore, the critical weather patterns were also examined in further analysis steps. A recent report on weathering risk in Jordan in 2022 (WeatheringRisk, 2022) describes the current state of the climate and projected future trajectories. As a result of climate change, water stress is expected to increase, with prolonged dry periods becoming more severe as temperatures rise. Meanwhile, rainfall 21 patterns are becoming less evenly distributed, increasing the likelihood of localised flash floods under current and future climate conditions. Annual Mean Precipitation Annual maximum Precipitation Figure 7. Long-term trends (time correlations) of the annual mean/maximum daily precipitation (left/right) in Jordan from 1961–2018 derived from UERRA regional reanalysis data. 4.2.2 Contextualization of Heavy Rainfall in Jordan Any extreme rainfall event in Jordan and elsewhere occurs within a larger-scale atmospheric context, which is defined by the movement of air masses across different geographical regions. Using ERA5 global reanalysis data (Hersbach et al., 2020), we have retrospectively identified the critical circulation pattern associated with extreme local rainfall in Amman (nearest grid cell) for the past (1961–1990) and present (1991–2020) climate periods. A standard atmospheric field in synoptic meteorology is the geopotential height at 500 hPa (Z500), which represents the circulation conditions in the middle troposphere at a height of about 5 km. The curvature of the contour lines is a good indicator of the origin and transport path of air masses. Improving our understanding of the dynamical drivers of local extreme weather events could enhance existing early warning systems and long-term climate risk assessments. Figure 8 shows the basic approach, in which local daily time series (𝑃𝑅𝑑) are combined with the corresponding atmospheric (𝑍500𝑥,𝑦,𝑑). For 30-years climate periods, we filter out days (t) where daily precipitation exceeds the 99th percentile (N = 110d). Figure 8. Scheme of contextualization For these days, we normalised the Z500 and calculated the composite patterns using averaging. This approach was applied to the past (1961-1990) and the present (1991-2020) climate period, separately. Comparison of the patterns indicate possible changes in dynamical drivers. Comparing the patterns indicates possible changes in the dynamical drivers. Changes in intensity can also be identified and Final report 22 attributed by looking at the respective values of precipitation above the 99th percentile. Changes in frequency were identified by applying the 99th percentile threshold to both the past and present periods. Figure 9 illustrates the critical weather patterns associated with extreme rainfall in Amman in both the past and present. Figure 9. Causal linkage between local extreme rainfall (red) and the large-scale atmosphere circulation (contours): past (left), present (right) and present-past (center). A trough-like pattern over the eastern Mediterranean transports cold air masses from north to south, favouring the formation of low-pressure systems and convective rainfall patterns in the Near East region (Dayan, 2015). The difference pattern in the center of Figure 9 shows a slight decrease in frequency in recent years, but not in intensity. This is probably related to higher blocking activity over Eurasia. Similarly, we examined the global climate model ensemble CMIP6 (Copernicus, 2021) up to 2100 for the high-end scenario (SSP585). We find that the models agree on a reduction in the frequency of such events. However, there are larger uncertainties in the changes in intensity. This is not surprising, given that these are global models with coarse spatial resolution. That’s why we also examine regional models with high spatial and temporal resolution from the Coordinated Downscaling Experiment (CORDEX, 2019), in order to capture local phenomena and resulting precipitation patterns more realistically. 4.2.3 Data Products in a Design of Radar The design format of data products is essential for providing user-oriented information. This product integrates operational services combining different open data products (reanalysis data, satellite rainfall estimates and weather forecasts) into a unified design of a radar for Jordan. This enables areas of accumulated rainfall to be monitored and assessed in terms of retrospective analysis of heavy rainfall maps and forecasts for the next 3 days. Figure 10. Real-time rainfall estimates from GSMaP satellite product (Kubota et al. 2020). In practice, the product looks like this (see Figure 11). Heavy rainfall maps are shown on the left. These are derived from regional reanalysis data (UERRA) and illustrate the spatial distribution of heavy rainfall with a return period of 10 years. The range of values extends up to 70 mm/d in the north-west of the country. Following this logic, other real-time data (satellite and forecast) have been integrated. Please note that the data comes from different sources and has not been adjusted for bias. The current status remains experimental beyond the scope of the project. Ideally, it would be possible to monitor and assess the potential risk of expected or predicted rainfall areas and rainfall totals. 23 Heavy Rainfall Maps Satellite Rainfall Estimates Weather Forecast 10-yr return level Nowcast of 12-hr acc. precipitation DWD-ICON 48-hr acc. precipitation Figure 11. Heavy rainfall maps in a design of a Radar: reanalysis (left), satellite (center) and forecast (right). 4.2.4 Heavy Rainfall Maps Local heavy rainfall exhibits considerable diversity and is highly sensitive to long-term temperature increases. To demonstrate this, nine 30-year regional climate simulations were evaluated and compared. The initial dataset is available with a temporal resolution of three hours. From this, a variety of event classes, ranging in length from three to 24 hours, were considered. As shown in Figure 12, these are regionalisations of selected global climate model simulations (CMIP5) from 1981 to 2100 under the highend RCP85 emissions scenario. The data are available in the Copernicus portal’s climate data store (CORDEX, 2019a). The selection process was necessarily pragmatic, taking into account a number of factors, including the choice of domain, the availability of runs, temporal and spatial resolution, and the meteorological phenomenon under consideration. In addition, aggregation over several hours allowed for the recognition that local heavy rain events occur predominantly on a sub-daily basis, thus allowing for the consideration of different influencing factors. The shorter the duration of the events, the more dominant the temperature effect, which is likely to lead to an increase in precipitation intensity. Figure 12. Overview of the available high-resolution climate scenarios for the Euro-Cordex-11 domain having 3-hourly precipitation. Figure 13 shows maps illustrating the sensitivity of 3-hour heavy rainfall in the regional climate ensemble. In most of Jordan, the intensity of precipitation is increasing. The magnitude is estimated to be between 10 and 15 percent for events occurring on average once every 50 years. This magnitude is consistent with that which can be expected on average from theoretical derivations. Final report 24 Figure 13. Derived heavy rainfall maps for past (left) and future (right) conditions: 3-hourly and return period 50-yr. Heavy precipitation can occur anywhere, even in regions that have previously experienced minimal rainfall. It is important to note that Jordan is situated at the border of the domain. It is possible that edge effects may distort the results in this instance. However, this applies equally to both the historical and future simulations. Additional maps for different return periods and duration levels were considered (not shown). In comparison, the short-term events show the strongest climate sensitivity (shown in Figure 14). This finding is relevant for the future planning and adaptation strategies of critical infrastructure in Amman and Petra. The duration of rainfall events significantly impacts the strength of the interaction between dynamic and thermodynamic effects. Consequently, opposing trends may emerge, such as a reduction in the number of events (critical weather conditions) coupled with an increase in their intensity. These are processes and phenomena whose future development is difficult to predict using current climate model simulations. Uncertainties remain because the scale of the phenomena considered varies in space and time. Underestimates of possible developments cannot be ruled out because models, despite their complexity, do not capture all interactions in the climate system. Even the slightest shift in the position of critical weather patterns can distort the results. Figure 14. Sensitivity of heavy rainfall scenarios 3-hr to 24-hr under past (blue) and future (red) climate conditions. 25 4.2.5 Climate Service Portal Climate Impacts Online is a climate service portal for countries run by the Potsdam Institute for Climate Impact Research (PIK), with the aim of putting climate change knowledge into practice. As part of the CapTainRain project, Jordan has been added as a new country. The service provides users with information on the latest climate projections for selected climate indicators up to 2100, for different SSP emission scenarios. This information is presented as zoomable maps, charts and tables in different languages. The baseline scenarios are derived by the ISIMIP initiative at PIK (https://www.isimip.org/). They provide bias-adjusted global climate model simulations (Copernicus, 2021) to researchers studying the impact of climate change worldwide. However, the mesh size of the data is relatively coarse. As a result, Jordan’s regional characteristics are not well represented. The situation is different for the regionally higher resolution climate simulations used in the project to produce heavy precipitation maps. In contrast, the regional distribution of precipitation patterns is particularly evident in the western part of the country. Ensemble simulations at this scale are very large and time-consuming. Here we have accessed and processed available runs. Figure 15. Screenshots of the climate service portal for Jordan (https://climateimpactsonline.com). Final report 32 Figure 21. HEC-HMS modules used in the hydrological model setup of the CapTain Rain project. Further delineation steps, such as sub-basin definition and stream identification, can be performed directly in HEC-HMS based on a DEM. For the meteorological model, we calculated the gauge weights of the rainfall stations using the Thiessen polygon function in ArcGIS. The curve number (CN), which is an empirical parameter used to predict infiltration versus direct runoff, is derived from soil and land cover maps (Awad, 2023). Literature values from USDA (USDA, 2010) were used for the different parameters of the transform and routing modules. The following table provides an overview of the input data used for the HEC-HMS models in Amman and Wadi Musa, respectively. Table 2. Input data for the hydrological modelling in the study areas Amman and Wadi Musa respectively. Input data Amman Wadi Musa DEM 1 m (RJGC) 2 m (PDTRA) Rainfall Time series for five stations in/close to our study region in 5 min resolution for the event in Feb. 2019 (JMD, UN-Habitat 2020) Time series for nine stations in event dependent resolution (PDTRA) and for one station in hourly resolution (MWI) in our study region for the event in Dec. 2022 Land cover 2021: Land cover classification of Sentinel-2 images (Awad 2023) 2021: Land cover classification of Sentinel-2 images (Awad 2023) Soil information HYSOGs250m (Ross et al., 2018) HYSOGs250m (Ross et al. 2018) 5.2.1.3 Hydraulic modelling and flash flood hazard maps Two methods were applied to analyse the flash flood hazards in our study regions of Amman and Wadi Musa: First, a flowpath sink analysis was performed (see Figure 20). This rapid methodology allows watershed and water flow paths to be analysed using GIS software and a DEM. Secondly, hydraulic modelling was performed. This required a software package and input data such as a DEM, buildings and rainfall events. Using personnel and computer resources, the hydraulic modelling produced GIS maps showing flash flood areas, flow velocities and inundation depths. The flash flood hazard was then interpreted in terms of inundation area and depth resulting from a rainfall event of a certain intensity and probability. The hydraulic models and simulations were created using the “Urban Flash Floods” software package (HE2D/FOG2D) from ITWH GmbH, version 8.6.2. The “HE2D” hydraulic model uses 2D 33 shallow water equations to calculate surface flooding processes. These flow equations are solved in space using cell-centred finite volumes and in time using the explicit Euler approach. The model calculates inundation areas with water levels and surface flow velocities. HE2D performs the discharge calculations on an irregular triangular grid. This is generated on the basis of the DEM (1 m respectively 2 m grid) and the buildings using the FOG2D model generator. To ensure that the simulation of the model runs is as efficient as possible, the size of the triangles is defined differently depending on the location. Each triangular element has a constant height value (Z-value), which is determined using a smoothing process. Building polygons are considered as non-flow gaps in the 2D calculation mesh. For the rainfall-runoff calculations, a distinction is made between paved and unpaved areas. Infiltration losses are considered for infiltration on unpaved surfaces using the Horton approach. HE2D uses the Manning-Strickler roughness approach. The roughness coefficients (kSt values) are determined based on the LULC. Flash flood hazard maps, which are produced using 2D hydraulic modelling, show areas that are particularly at risk during periods of heavy rainfall. They illustrate where water accumulates and which areas may be particularly susceptible to flooding. These maps are essential for urban planning, disaster management and risk assessment. Alongside the analysis of damage potential, they provide a foundation for analysing flood risks. For Amman, the area selected for hydraulic modelling is around 120 km². For performance reasons, the hydraulic model boundary was chosen to be slightly smaller than the watershed identified in the flow path–sink analysis. However, it still encompasses key urban areas such as the Roman Theatre. The Wadi Musa hydraulic model area, on the other hand, was chosen to be slightly larger than the watershed and is also approximately 120 km² in size. Figure 22. Hazard map for Downtown Amman modelled with HE2D/FOG2D with recorded rainfall from five stations during the February 2019 event (most intense period). Water levels are classified according to the German DWA-M 119 standard. The results of our model depend heavily on the resolution and quality of the DEM. The 1 m resolution DEM for Amman (RJGC) shows some unrealistic depressions where water accumulates. Furthermore, detailed information on Amman’s sewer systems is lacking. Regarding the hydrological models, data on water levels for calibrating and validating the model results was mostly lacking, except for photographic documentation of the heavy rainfall event that flooded the Roman theatre in 2019. Final report 34 For Wadi Musa we had a DEM with a resolution of 2 m (PDTRA) and of good quality. Only some adjustments were necessary in the area of the Siq. In some parts, the top of the canyon is so narrow that we had to adjust the DEM to create a flow path of the correct size. We used the same approach for the tunnel at the beginning of the Siq. For the subsequent risk assessment, we needed to determine four hazard classes. We opted for a classification according to the DWA-M 119 guideline: low = water depths < 0.1 m, medium = water depths ≥ 0.1 and < 0.3 m, high = water depths ≥ 0.3 and < 0.5 m, and very high = water depths > 0.5 m. 5.2.2 Flash flood damage potential In addition to the hazard analysis, a comprehensive risk assessment requires an analysis of the damage potential, which categorizes buildings and other elements according to their type of use (e.g. residential, commercial, critical infrastructure etc.). The assessment of the damage potential considers the possible impact of heavy rainfall events on various sectors or aspects of life. These include people, particularly vulnerable groups such as children and people with restricted mobility; infrastructure, including critical infrastructure, but also residential, industrial and agricultural areas; the environment, including protected areas; and other aspects such as cultural heritage. The aim of assessing the damage potential is to determine, localise and illustrate protection requirements and damage susceptibility and to assign damage potential classes. Damage potential is evaluated independently of the hazard assessment. Where a high damage potential is present within a hazard area (i.e. an inundation area), a flash flood risk arises. Therefore, the damage potential assessment, together with the hazard assessment, forms the basis of the risk and vulnerability analysis. Figure 23. Damage potential map for Downtown Amman, showing buildings with a damage potential class for flash floods without consideration of flooded areas. In CapTain Rain, the classification of the damage potential was based on the German standard DWAM 119 (DWA 2016), adapted to the local situation through the expert opinion of selected Jordanian stakeholders (n = 5). Four classes were used to categorise the damage potential: 1 – low, 2 – moderate, 3 – high and 4 – very high (critical). Buildings within the study areas were classified based on their predominant use. The data basis for Amman comprised shapefiles from GAM and OpenStreetMap (OSM) datasets, and for Wadi Musa a shapefile from PDTRA, OSM datasets. For Wadi Musa, the data basis comprised a shapefile from the PDTRA and OSM datasets. Additionally, information was added using digitised buildings based on an aerial photograph (PDTRA) and building usage information from 35 Google Maps and Google Earth. However, little information about critical infrastructure was available for Amman and Wadi Musa. The results of the damage potential assessment indicate that densely built-up urban areas are at high risk of damage, since floods can affect many buildings and escape routes can easily become flooded or blocked. The damage potential increases to very high if buildings have basements. The damage potential also increases if people are less mobile and thus less able to leave flooded areas. Therefore, hospitals and childcare facilities are assigned a very high damage potential. The damage potential is also classified as ‘high’ for critical infrastructure, including emergency services and energy infrastructure. As the classification of damage potential can be subjective to a certain degree, it is recommended that it is assessed by expert groups and/or stakeholder workshops. Such a joint assessment can reduce subjectivity. Also, as the determined damage potential is based on a snapshot in time, it should be reviewed at regular intervals to take into account future developments, as well as possible changes in land use and infrastructure. 5.2.3 Flash flood risk The aim of the flash flood risk analysis is to assess the degree of risks in potentially flooded areas through the combined consideration of hazard and damage potential. By identifying high-risk areas and objects, the risk analysis points out existing needs for action and provides the basis for the development of precautionary measures. Particular attention is paid to critical infrastructures and objects. Table 3. Flash flood risk matrix (combination of hazard and damage potential). A location is considered to have a high flood risk if it has a high flood hazard (e.g. high water levels) and an object or area with a high or very high damage potential (e.g. a hospital) is present. The assessment of risks to human health, the environment, cultural heritage, infrastructure facilities and other assets is based on qualitative analysis in four categories according to DWA-M 119 (2016): low, moderate, high and very high. The risk class is determined using a combination matrix in which the flood hazard and damage potential are considered together; see the table below. Final report 36 Figure 24. Flash flood risk map for Downtown Amman. 5.2.4 Multi-model approach to analyse runoff peaks from flash floods The aim of the hydrological modelling was to assess potential runoff curves resulting from heavy rainfall events. Runoff curves enable the modelling of scenarios with regard to climate adaptation measures in larger basins. Based on runoff peaks, it would be possible to evaluate the potential reduction in water levels and inundation areas in these scenarios. However, due to lacking runoff data, it was not possible to calibrate or validate the hydrological models. However, due to a lack of runoff data, it was not possible to calibrate or validate the hydrological models. To ascertain the plausibility of the models’ results, we adopted a multi-model approach for the Wadi Musa study area. By simulating the December 2022 flash flood event with three very different models, we aimed to gain insight into the uncertainty of the results. For this, we used the aforementioned HEC-HMS (hydrological) and HE2D/FOG2D (hydraulic) models, as well as the Rainfall-Runoff-Inundation (RRI) model. The RRI model was setup in a 100 m grid resolution based on the data shared with us by Dr. Sameh Kantoush, Kyoto University. The following figures show the modelled runoff curves at the Siq entrance for all three models. For the model comparison, we used the rainfall data from ten stations, for the event in December 2022 (28 hours), which was selected based on data availability. The models reacted differently to the event: the HEC-HMS model showed the fastest and strongest reaction, while the RRI and HE2D/FOG2D models showed their highest peaks up to two hours later. Runoff modelled with RRI is more inert, with a late peak and a slow descending curve. The hydraulic model HE2D/FOG2D reacts more slowly, showing the first flow nine hours later. 37 Figure 25. Simulated runoff at the Siq entrance, Wadi Musa with the models HEC-HMS (purple), RRI (green) and HE2D/FOG2D (yellow), as well as the mean input rainfall (blue) for the event of December 2022. The models are set up with the same input parameters regarding DEM, slope, soil information, and rainfall. However, the output runoff curves differ due to variations in the model setups and internal processes. While all models provide water level variations at key points within a range of 8 to 30%, the runoff peaks vary between ~ 40 m³/s and > 70 m³/s, i.e. ~ 75% variation. For the December 2022 event, the water levels are almost identical, differing by only 8 %. In the intense rainfall scenario, the water levels of the three models differ up to ~30 %, with the RRI model showing the lowest values (0.8 m), and HE2D/FOG2D the highest water levels (1.2 m). Still all water level results represent a quite close range (0.84 m – 1.21 m). Striking is the behavior of RRI which shows similar water levels for the two rainfall inputs, a behavior which we cannot explain yet. Table 4. Simulated water at the bridge near the Petra entrance in Wadi Musa for two different rainfall inputs. Event HEC-HMS RRI HE2D/FOG2D Dec 2022, 28 h 0.78 m 0.85 m 0.81 m Intense scenario 1.00 m 0.84 m 1.21 m 5.3 Outlook: Flood hazards and risk areas The model results and produced maps are important tools for the Jordanian partners and are now ready to be used for planning purposes. Further refining the hydraulic model for Amman would be beneficial, for example by integrating detailed sewer and stormwater network information and/or an improved DEM. Partners working with data such as DEMs and building information, such as the GAM for Amman and the PDTRA for Petra, should link this data to flash flood hazard and risk maps. They should also actively update these maps and use them to inform other institutions and the public. The HE2D/FOG2D hydraulic model developed by ITWH for Amman and Wadi Musa has been set up so that hazard and risk maps can be easily updated. Other hydraulic models could also be utilised for Amman and Petra, bearing in 0 10 20 30 40 50 60 70 80 90 1000 10 20 30 40 50 60 70 80 90 100 0:00 3:15 6:30 9:45 13:00 16:15 19:29 22:44 2:00 5:15 8:30 11:45 15:00 18:15 21:30 0:45 4:00 7:15 10:30 13:45 17:00 20:15 23:30 2:45 6:00 9:15 12:30 15:45 19:00 22:15 Mean rainfall [mm/h] Simulated runoff [m³/s] Mean Rainfall HEC-HMS RRI HE2D/FOG2D Final report 38 mind potential licence costs, the resources required to set up the hydraulic models, and the resources required for data proofing. In order to calibrate the models and increase their reliability, runoff data is required. Therefore, staff gauges should be installed and maintained, and monitored during periods of rainfall. Suggestions for installing staff gauges have been provided to both GAM and PDTRA. Also, the quality of rainfall data should be improved. For heavy rainfall events, high-resolution (approximately 5-minute) rainfall data should be available. Ground station data needs to be coupled with radar data to obtain special distributed heavy rainfall data. Such quality-controlled rainfall radar data is not currently available in Jordan. Processing radar data and coupling it with ground station rainfall data would require additional resources. Furthermore, the southern part of Jordan is not covered by the Jordan Meteorology Department’s radar system. In future, the quality of rainfall data should be improved to provide a reliable basis for more accurate modelling. This would reduce infrastructure adaptation costs, since this infrastructure could be adapted to quantified rainfall events. 39 6 Adaptive capacity: Selection and localisation of adaptation measures Authors: Daniel Schuhmann-Hindenberg, Linnéa Fölster, Martina Winker and Katja Brinkmann 6.1 Research objectives and methods Adaptive capacity refers to a system’s potential or capability to adapt to flash flood risk. Technical measures alone are insufficient to mitigate the damage caused by flash floods in urban areas. To significantly reduce the damage potential, an integrated set of adaptation measures is needed in both public and private areas. A change of mindset is needed towards jointly assessing promising adaptation measures, involving officials and local stakeholders with different areas of responsibility. In addition to structural measures and property protection, innovative measures for the retention, safe discharge, storage and use of heavy rainfall were identified, evaluated, allocated and approved on the basis of feasibility. The aim was to expand the portfolio of adaptation options to prevent flash flood damage by evaluating innovative measures for the discharge and use of heavy rainfall, and to deliver recommendations for the planning process and implementation of measures. Appropriate and innovative measures to mitigate flash flood risks were identified through literature surveys, GIS-based analysis and participatory methods. Additionally, expert interviews (Chapter 3) were considered. Identifying planning objectives, suitable areas for future implementation (‘focus areas’) and the appropriate measures is an iterative process. In CapTain Rain, we conducted a stepwise participatory planning process divided into three main steps: Data collection and analysis; planning and localisation; and development and strategy (Figure 26). Figure 26. Overview of the stepwise planning process comprising data collection & analysis, planning and localization, and the development & strategy. The process started with the identification of the planning area, followed by collecting and analysing spatial data and field information. All the collected GIS data and relevant information were integrated into a GIS database for analysis. To enable spatially explicit planning and allocation of measures, we used baseline data and maps provided by our Jordanian partners, as well as the results of the hydrological and hydraulic analyses (Chapter 5) and vulnerability assessments (Chapter 8). Successful implementation of flash flood mitigation measures also requires stakeholder involvement. We used the results of the stakeholder analysis (Chapter 3) to consider the different responsibilities, roles and tasks in the 1. Data Collection & Analysis •Identification of planning area and data collection •Categorization of land use types •Assessment of area potential 2. Planning & Localization •Stakeholder engagement •Identification of planning goals and measures •Determination of entry points and focus areas •Selection and localization of measures 3. Development & Strategy •Scenario development •Impact assessment and readjustment •Integration into planning concepts •Development of detailed Plans Final report 40 planning processes according to the different urban areas and sectors involved such as roads, green areas, private and public areas. We also considered water resources and urban structures. Additionally, we analysed the perceptions and knowledge of the local population to identify knowledge gaps and recommend knowledge transfer. To this end, structured interviews were conducted in flash flood hotspot areas in Amman (n = 52) and Wadi Musa (n = 15). In the second phase, ‘Planning & Localisation’, further stakeholder needs were elaborated upon, and focus areas, planning goals, and suitable measures were identified in close collaboration with stakeholders through the second stakeholder workshop, virtual planning workshops, and bilateral exchange. To integrate scientific and practical knowledge, we compiled a database of measures and their potential contribution to flash flood protection and water retention in the MENA region through a literature review, expert interviews and utility experience. The subsequent selection and allocation of measures was carried out in a co-creation process with stakeholders and local decision-makers. To support the participatory planning processes, a multi-touch table was introduced. The multi-touch table (Figure 27) is located at GAM and is the responsibility of the Strategic Planning Department. In the third phase, an impact assessment was conducted based on the integrated scenario analysis within CapTain Rain. The effects of the selected and allocated measures were simulated using the hydraulic models presented in Chapter 5. These results are summarised in Chapter 9. Figure 27. Multitouch table introduced at GAM in Amman and at the final workshop. 6.2 Key findings 6.2.1 Increased knowledge of blue-green infrastructure Measures for rainwater management and flash flood protection encompass a blend of innovative and traditional approaches to effectively manage stormwater and, in this case, reduce the damage caused by flash floods. These measures include:  Blue-green infrastructure, which leverages natural elements such as wetlands, green/blue roofs, and rain gardens to manage rainwater through absorption, filtration, and storage.  Technical measures, or grey infrastructure, which uses engineered solutions like pipes, culverts, and dams to control and direct the flow of stormwater.  Multifunctional tools in which various of the above-mentioned measures as well as other urban requirements (parking lots, playgrounds, sports activities) are integrated on the same plot. A comprehensive literature review and expert interviews (data from stakeholder analysis) were conducted to identify innovative measures for the retention, safe discharge, storage and use of heavy rainfall in arid and semi-arid regions. The results were summarised in a database/catalogue, covering two main aspects: traditional water harvesting methods in arid and semi-arid regions and the types and possibilities of rainwater management in terms of Blue-Green-Infrastructures (BGI). Each measure has unique characteristics and performance, and can be selected according to which planning goal it best 41 addresses, ensuring an effective, tailored rainwater management strategy. While some measures address flash flooding directly, others focus more on the impacts of climate change, such as heat or drought. However, all measures can serve more than one objective. The set of measures was discussed, refined and prioritized in several meetings with the Jordanian partners. In addition, an in-depth analysis of the expert interviews was conducted to gather information on overall challenges and knowledge of, and experience with, flash flood adaptation measures. Based on these results and further expert evaluation, 12 measures were selected from the catalogue for the subsequent planning process and the preparation of information material (infocards, Figure 28) for each measure. The infocards (Schumann-Hindenberg et al., 2025) contain a short description of each measure, its contribution to planning goals and selection criteria such as costs, maintenance needs and implementation conditions. In addition to these infocards, capacity development was conducted for local stakeholders based on a webinar “Measures to reduce flash flood risks/Sponge City Hamburg” in June 2022, which provided insights into possible measures and a practical example for a participatory planning process in Hamburg, Germany. Several online workshops were also conducted as part of the participatory planning process in 2023 and 2024. Figure 28. Infocard example for the measure “Raingarden”. 6.2.2 Increased knowledge through geospatial analysis Geospatial analysis was used to examine land use data and the topography of the catchment area, with the aim of identifying opportunities to integrate BGI into urban landscapes. The sub-catchment areas of Amman were derived from hydraulic and topographic data and used for further processing of the land use data and the potential analysis. The land use map depicts a series of grouped land use categories derived from the GAM typology. These were labelled according to their degree of development and the availability of open space (i.e. undeveloped, non-built-up areas). The latter was derived from land use and land cover maps in a recent study by Awad (2023). Information on flood-prone areas from hydraulic and hydrological models (see Chapter 5) and documentation of actual flooding events were combined with land use information and other cadastral data in a GIS. In this way, potential fields of action were identified and measures assigned. An overall analysis helped to recognise how much space could be available for the implementation of measures per catchment area, whereby a distinction is made Final report 48 Figure 36: Conceptual design of measures in public areas, example secondary school (Source: Sharma, 2024). A multi-scenario analysis was carried out to demonstrate the impact of the measures outlined in these concepts (see Chapter 9.2). To this end, a series of blue-green infrastructure measures were developed within the focus areas, comprising the detailed concepts shown in Figures 36 and 37. The effects of these scenarios, simulated using hydraulic models (see Chapter 5), are summarised in Chapter 9. 6.3 Outlook: Selection and localisation of adaptation measures The planning tools and manuals that have been developed will provide the Jordanian partners with a foundation on which to initiate a participatory process and implement blue-green infrastructure. The adaptive capacity measures and strategies can be applied to other areas and, with the help of technology, provide an innovative approach to reducing the risk of flash flooding. The multi-touch table has been designed to serve as a valuable instrument in future stakeholder meetings, with the Strategic Planning Department and the GIS Department already familiar with its functionalities. The table is intended for use in various urban planning processes in Amman and other regions of Jordan, including the planning of new roads, urban development, green roof strategy, and the implementation of climate change measures. It can be utilised for both stakeholder exchange and internal planning meetings. There is a significant opportunity for impact in deeper collaboration with Miyahuna on comprehensive rainwater management. Miyahuna is involved in the Captain Rain project as a collaborating partner and participated in the stakeholder workshop in January 2023. Several additional meetings have been held to discuss further collaboration within CapTain Rain. Integrating Miyahuna’s data into the hydraulic model, as discussed in WP3, is expected to improve the model’s accuracy. However, this integration was not yet completed, and GAM has been tasked with completing it. Another notable outcome is the identification of synergies between stormwater management and groundwater recharge, an initiative that appears to already exist in Amman. 49 According to GAM, the results of this project will inform further strategic planning processes and climate adaptation strategies, such as ‘Future Amman’ (Three Strategies for Climate-Smart Spatial Transformation) and the Amman Green City Action Plan. They will also inform programmes run by UN-Habitat, such as the development of a preliminary design for flood mitigation and flood risk assessment and flood hazard mapping for Downtown Amman, and by INWRDAM, such as flood water management and risk reduction. These initiatives will enable the city of Amman to effectively reduce the risk of flash floods and establish a more resilient and sustainable urban environment. The tools developed, along with the guidelines containing recommendations for the urban planning process for selecting and implementing measures (Schumann-Hindenberg et al., 2025), will enable GAM to consider alternative options for improving rainwater management. Final report 50 7 Adaptive capacity: Water and weather data portal for Jordan to improve early warning Author: Michael Thiemann 7.1 Research objectives and methods Proactive flood risk and damage assessments, as well as flood impact reduction measures, are critical to managing the occurrence and impact of floods, as investigated in CapTain Rain. Nevertheless, flooding will continue to pose a real risk to Amman and Petra, even if all economically viable measures are implemented. This risk will be exacerbated by rising rainfall variability due to climate change. Therefore, real-time monitoring of the weather and weather forecasts, along with the timely mobilisation of first responders, plays a vital role in managing flooding and its impacts in Amman and Petra. We assessed the current state of flood warning activities in Amman and Petra, analysing the strengths and weaknesses of existing early warning systems (EWS), including their underlying data sources, methodology, and dissemination tools. Recommendations were developed for an EWS adapted to users' needs. The user-friendliness of Petra’s existing EWS was evaluated through expert interviews and focus group discussions with local stakeholders (including questions such as: What information should be included? What media channels should be used? Do warnings reach all people at risk? Are the risks and warnings understood? Are the warnings clear and usable?). This information was used to develop recommendations for the creation of early warning apps for the population. These recommendations formed the basis for the design and implementation of a water and weather EWS portal as a demonstrator. 7.2 Key findings 7.2.1 Assessment of Flood Warning in Amman and Petra Flood warning activities are already performed in Jordan. In order to best implement adaptive capacity measures, we assessed the current status of Flood Warning in Amman and Petra. 7.2.1.1 Amman In Amman, the first step was to document the governmental agencies involved in the event of flooding incidents. There is no early warning system in Jordan or Amman at a national scale. Furthermore, the emergency response chain in the event of flash floods differs at regional (Amman: Greater Amman Municipality, GAM) and national levels (excluding PDTRA). The resulting responsibility and communication diagram is presented in Figure 37. The Jordan Meteorological Department (JMD) sends a climate report twice a day to various institutions. Once the JMD identifies a high risk of extreme rainfall that could cause significant local flooding, the National Center for Security and Crisis Management (NCSCM) implements an overarching emergency response plan. The NCSCM coordinates activities between MWI, MoE, Civil Defense, and also GAM in case of flash floods in Amman. In the case of flash flood incidents in Jordan, the Ministry of Water and Irrigation, the Ministry of Environment, Civil Defense, as well as Police and Military act on the ground. MWI hereby acts according to its emergency response plan and through field teams at at-risk locations. Evacuations and related management during events are to a large degree performed by the Police, the Military and Civil Defense. The MoE only acts in case of the flooding of areas with hazardous materials, which could lead to water pollution, based on its emergency response plan. Amman is a special case because the GAM has its own emergency response structure. According to the GAM emergency response plan, the emergency unit, comprising a main centre, 22 district centres, field teams, the Building Observation Department and the Media Department, is activated in case of 51 flash flood risk, especially during winter. The emergency unit deploys pumps, mobile dams and other equipment to areas at risk of flash flooding in Amman, which GAM has identified based on experience of prior incidents. The GAM Media Department provides information to the population. The Building Observation Department assists the field teams in evacuating people. Three different lines of decisionmaking exist: one for the emergency level (GAM only), one for warning the population (GAM and NCSCM) and one for evacuating people (GAM mayor and minister of the interior). It is important to examine the decision-making lines in more detail, including more than one actor, to determine whether a more efficient structure is required. GAM is responsible not only for emergency response, but also for cleaning out canals before events and cleaning up after them. The key to a successful emergency response is an accurate, timely and location-specific rainfall warning. Using a questionnaire, WP6 assessed flood warning activities in Amman. These were then compared to a set of ideal actions and, based on a gap analysis, recommendations were prepared. Key recommendations include cross-agency integration of real-time rainfall observations, use of the Amman weather radar to identify areas at risk of local flooding, and establishment of a regional, highresolution weather forecast model. Such information would greatly enhance the agencies’ situational awareness before and during extreme rainfall events, making the planning and execution of emergency management measures more effective. Final report 52 Figure 37. Actors during flood emergencies in Amman and at the national scale. 53 7.2.1.2 PDTRA PDTRA operates a network of rainfall observation stations in the Petra area to assess the risk of extreme rainfall and local flooding (Alhasanat, 2017). PDTRA itself primarily handles the coordination, especially within the Petra Archaeological Park. Work Package 6 conducted a status assessment and gap analysis, providing recommendations. These recommendations are similar to those for the Amman area. 7.2.2 Water and Weather Portal for Jordan The Water and Weather Data Portal for Jordan (WWDPJ) was developed to demonstrate the implementation of some of the recommendations via the cloud-based KISTERS datasphere tool for data management & operation. Datasphere provides easy access to real-time weather observations and forecasts via any web-browser and was expanded during the CapTain Rain project to efficiently and reliably manage and visualize open (and free) weather observations and forecasts for use in Amman and Petra. As such it can be used by the Jordanian stakeholders to gain awareness of current weather and to prepare for emergency management activities. In that realm the WWDPJ covers the functionalities marked in yellow in Figure 38. The related access and training were offered to MWI, JMD, GAM, WAJ, and PDTRA. Figure 38. Functionalities covered by the WWDPJ (yellow highlights) as Part of the Overall Emergency Management Activities. Final report 54 7.2.2.1 Historical Data Products While the main purpose of the WWDPJ is to provide situational awareness of current and future weather, it can also manage historical data that can be downloaded for further study purposes. The Jordanian stakeholders provided the historical data listed in Table 6, which was made accessible via the portal. Figure 40 depicts locations for which historical data were made available in the WWDPJ map interface. Figure 41 provides an example data graph of historical precipitation at one site. Figure 39. Historical Data Locations shown in the WWDPJ Map Interface. Figure 40. Example Graph of Historical Data in the WWDPJ. 55 Table 6. Historical Data available in the WWDPJ Parameters Period Spatial resolution Area Provided By Temperature Max, Temperature Min, Temperature Mean, Wet and Dry Bulb Humidity, Vapor Pressure, Dew Point Temperature, Relative Humidity, Wind Direction, Windspeed, Wind Distance, Evaporation, Radiation, Sunshine Hours 1964 to 2017 daily total Jordan Ministry of Water and Irrigation (MWI) Precipitation, Temperature, Relative Humidity daily total Jordan MWI Precipitation, Evaporation, Temperature, Relative Humidity, Radiation, Wind Direction, Windspeed,Pressure hourly 2 stations Mutah University Max Temperature, Min Temperature, Precipitation, Windspeed, Relative Humidity, Radiation 1979 to 07/2014 daily 3 stations USA National Centers for Environmental Prediction (NCEP) Precipitation, Relative Humidity, Temperature, Wind Direction, Windspeed 2022 6H total Jordan World Meteorological Organization (WMO) 7.2.2.2 Observed Remotely Sensed Weather Products Ground observations of weather and water level data can provide a valuable assessment of the local situation. However, larger weather patterns can be analysed via remotely sensed data, typically obtained via satellite observations. Table 7 lists the satellite remote sensing data available through WWDPJ. Table 7. Satellite Data Products in the WWDPJ Provider Products Cover Spatial Resolution Temporal resolution Update Interval Eumetsat H SAF Precipitation (H03B and H60) Global 0.05 ° (~5km) 5 min Every 15 min. Japan Meteorological Agency (JMA) Precipitation (NRT gauge-calibrated und real-time gaugecalibrated) 60°N to 60°S 0.1 ° (~10km) 1 hour NRT every 2 hours; real-time every 6 hours Meteosat IODC (Middle East) Brightness temperature (various) RGB Composite (various) Reflectance (various) Indian Ocean 0.05 ° (~5km) 5 min Every 15 min. These are downloaded from the providing organizations and imported into the WWDPJ at the intervals described. They can then be immediately visualized by the user (Figure 42 and 43). Final report 56 Figure 41. Selection of Satellite Observations Products via the WWDPJ Interface. Figure 42. Animating a Selected Satellite Observations Product in the WWDPJ Interface. Ground Rainfall Observation Products As part of the project, the WWDPJ was coupled with the existing rainfall station network operated by PDTRA in the Petra area. Near-real-time rainfall observations at nine rainfall stations (Figure 43) are imported hourly into the WWDPJ and provided to the end user. 57 Figure 43. PDTRA’s rainfall observations sites shown in the WWDPJ Interface. Figure 44. Displaying Rainfall Observations in the WWDPJ Interface. 7.2.2.3 Rainfall Nowcast Product Rainfall nowcasts provide very short-term forecasts of rainfall by moving observed rainfall fields into the future. This can aid in the rapid mobilization of emergency management personnel as well as in the warning of the population just ahead of extreme rainfall events. Ideally, these rainfall fields are observed by local weather radars that provide rainfall estimates in nearreal time at high spatial and temporal resolutions and with good accuracy. However, such data were not available through JMD and rainfall nowcasting was hence implemented using an existing nowcasting scheme applied to the Eumetsat H SAF satellite precipitation product (Figure 45). Final report 64 7.2.2.7 Alarming The WWDPJ is set up to send notification emails to pre-defined individual recipients (Figure 55) or groups of recipients (Figure 56) if extreme observed or forecasted rainfall events have been detected. Figure 55. Definition of Alarm Recipients in the WWDPJ. Figure 56. Definition of Alarm Recipient Groups in the WWDPJ. 65 7.3 Outlook: Climate and water data portal for Jordan to improve early warning Flood warning and related emergency management is well established in Jordan. However, the data to make data driven decisions ahead of possible extreme rainfall (and possibly flooding) events is somewhat sparse due to the limited availability of real-time rainfall observations and high-resolution radar and weather forecasts. The WWDPJ demonstrates how the use of high-resolution open data and the sharing of weather observations and forecasts among the Jordanian government agencies could improve flood warnings and thus mitigate flood damage during extreme events. The main stakeholders in the project, and especially JMD, utilized the WWDPJ during the project period and voiced interest in continued access to the information provided by the system. The next steps should hence be to explore ways to continuously provide such access. WP 6 is in related discussions with the GIZ, the German Embassy in Amman as well as Swiss and US development agencies to explore the related funding. CapTain Rain project partners are also participating in the Jordanian Water Risk / Flood Mapping Working Group, which coordinates donor funded activities in this field. Final report 66 8 Integrated vulnerability assessment Authors: Katja Brinkmann, Ahmad Awad and Clara Hohmann 8.1 Research objectives and methods Flood vulnerability analysis and assessment is urgently needed to improve urban risk management, reduce flood damages and risks, and protect the local population. The aim of the analysis was to identify vulnerable areas and those with high adaptive capacity. Vulnerable areas and areas with high adaptive capacity were prioritised for the future implementation of flash flood mitigation measures. To identify such areas, we carried out an integrated assessment at different spatial and temporal scales summarizing the results of the various work packages, particularly those relating to heavy rainfall hazards (Chapter 4), flash flood risk (Chapter 5) and adaptive capacity (Chapter 6). The resulting flash flood vulnerability assessment will serve as a decision-support tool for urban planning, helping to identify risk areas and select adaptation strategies (maps, assessments and reports).. Our integrated (spatially explicit) vulnerability assessment (Brinkmann et al., in preparation) has been adapted for data-scarce areas by combining different disciplinary perspectives with local knowledge. An integrated way of understanding vulnerability can be carried out through a Social-Ecological Vulnerability Assessment (SEVA). SEVA is defined “as the extent to which environmental degradation and climate change cause negative changes in exposure, susceptibility and in the capacity of the socialecological system to anticipate, cope with and recover from the hazard” (Depietri, 2020). This approach was based on the general framework promoted by the Intergovernmental Panel on Climate Change (IPCC), which has been widely adopted for vulnerability assessments. In the IPCC definition, vulnerability is defined as the propensity or predisposition to be adversely affected (Ara Begum et al., 2022), often understood as a function of the component’s exposure, sensitivity and adaptive capacity (Thiault et al., 2021). Understanding these components is crucial for effective risk management and resilience building. For each of these components, several indicators were selected to capture the different social, physical and ecological domains (Table 9).This selection was based on data availability, expert opinion and stakeholder needs. Stakeholder perspectives and needs were incorporated into the vulnerability assessment through expert interviews (n = 7) and focus group discussions during three stakeholder workshops. In addition, interviews were conducted with local residents (n = 52) to gather additional information on adaptive capacity in terms of local knowledge of measures to reduce flash flood damage and willingness to implement measures on private land. For the calculation and mapping of indicators we conducted spatial analysis within ArcGIS 10.8 using available GIS data (e.g. buildings, land cover, population statistics at the neighborhood scales) provided by the Municipality of Greater Amman (GAM) and own analysis on exposure and land cover within WP 3. Data gaps, especially on buildings and green spaces, were filled by available open source data and by manual digitization based on recent Google Earth images. To estimate the social domain of sensitivity, which was based on population data at the neighborhood scale, we combined various demographic (very young or very old people, disabled people, refugees) and economic factors (low income, low level of education). To reduce data dimensions of this multivariate dataset, we performed a principal component analysis (PCA) and then used the results of the first principal component as an indicator, which is strongly correlated with the number of refugees and low income. Each indicator map was transformed to a 2 m raster file and classified within ArcGIS 10.8 using natural breaks (Jenks) from 1 = low; 2 = moderate, 3 = high and 4 = very high. 67 Table 9. Selected Indicators for the SEVA for Amman for each component (exposure, sensitivity and adaptive capacity) and domain (social, physical and ecological). Domains: Components: Exposure Sensitivity Adaptive capacity Degree to which a subject (inhabitant, building, ecosystem) is exposed to flooding in case of a flash flood event Degree to which a subject/object is affected by a given (flash flood) exposure Ability of a subject/object to adjust to the hazard event. It reduces the overall level of vulnerability and thus the effects of a flash flood Social Determined based on the proximity of residential buildings to flood prone areas. Classification was based on the risk analysis of WP 3 using the expected water level (cm) during a flash flood event (Baseline Scenario). Estimated based on demographic (very young or very old people, disabled people, refugees) and economic factors (low income, low level of education) as well as building types (Potter et al. 2009; Ababsa and Daher, 2011) Assessed based on the economic capacity of residents to implement measures on their land. Physical Determined based on the proximity of critical infrastructure to flood prone areas. Classification was based on the simplified risk analysis of WP 3 using the expected water level (cm) during a flash flood event (Baseline Scenario). Estimated based on the analysis of damage potential in WP 3 classifying critical infrastructure as sensitive objects (worship places, energy and water infrastructure, cultural heritage sites, schools, kindergarten) Assessment was based on the availability and suitability of open space for the potential implementation of measures to decrease flash flood damages. We considered impermeable open space (non-build up) on public and private land. Ecological Determined based on the proximity to flood prone areas of larger green spaces used for leisure activities and habitats for flora and fauna (parks, floodplain, woodland). Classification was based on the simplified risk analysis of WP 3 using the expected water level (cm) during a flash flood event (Baseline Scenario). Estimated based on the limited capacity for water infiltration. For this we used the impermeable surfaces (build-up areas extracted from Awad, 2023 6) as a proxy, as no detailed soil map data was available. The combination of the individual indicators maps combining exposure (E) and sensitivity (S) resulted in a map of vulnerable areas. All indicators were weighted equally in the following way: Exposure & Sensitivity = Esocial + Ephysical + Eecological + Ssocial + Sphysical + Secological For the assessment of adaptive capacity, we categorized land use parcels within the studied catchment of Amman based on the use (e.g. residential area, road, commercial area), ownership (private, public), residential type (residental villas A/B of better well-off inhabitants, others) and the level of development (developed, undeveloped, see Table 10). Besides the vulnerability assessment for the current situation, we also explored vulnerability for possible future pathways with regard to changes in heavy rainfall events (WP 2), as well as measures to decrease flash flood damages. These scenarios were simulated with hydraulic and hydrologic models (WP 3) and assessed using vulnerability indicators. Final report 68 Table 10. Categorization of the adaptive capacity based on physical (availability of open space), social (open space in private land of well-off residents) and ecological (permeable surfaces with high water infiltration) aspects Category Description Average open space on the parcel (m²) 1 (low) Little space for the implementation of measures (mainly built-up areas) 250 2 (moderate) Medium-sized open spaces on developed private land 950 3 (high) Large open spaces on developed parcels in public space and on private land belonging to wealthy residents (GAM class: residential A/B), as well as on undeveloped parcels in private land (GAM classes: commercial and residential C/D) 3000 4 (very high) Large open areas on undeveloped parcels in public space and on private land belonging to wealthy residents (GAM class: residental villas/A/B) 2000 8.2 Key findings for Amman 8.2.1 Co-creation of vulnerability indicators and assessment To integrate stakeholders’ perspectives on vulnerability indicators and gain insight into the assessment of these indicators, expert interviews and focus group discussions were conducted. The vulnerability indicator “physical sensitivity” classifies the damage potential of critical infrastructure based on the German standard DWA-M 119 (DWA, 2016). Thus, high damage potential is associated with energy and water infrastructure, hospitals, schools, kindergartens, basements or underground infrastructure. This classification was adapted to the local context by incorporating expert opinion from Jordanian stakeholders and adding the categories of ‘cultural heritage sites’ and ‘places of worship’. To this end, semi-structured interviews were conducted with local experts (n = 5) to elicit their views on the potential damage that flash floods could cause in Amman, with particular reference to the different land uses in the city. To find out which type of vulnerability indicators (components and domains) are most important for the integrated assessment from the stakeholders’ perspective, we conducted a focus group discussion during the Captain Rain workshop in December 2024. This was done through an Analytic Hierarchy Process (AHP) where participants rated the relative importance of the different vulnerability indicators. Each participant completed a matrix in which they compared the importance of the vulnerability indicator in pairs. Based on the ratings given, the weight for each indicator was calculated. A total of seven expert judgements were made. The results showed that social and physical exposure as well as social and physical sensitivity were weighted higher than social and physical adaptive capacity. Indicators related to the ecological domains were consistently assigned lower importance/ priority in the context of Amman city (Table 11). Table 11. Results of the AHP approach showing the relative importance of the different vulnerability indicators (components and domains) from the stakeholders’ perspective (n = 7). Vulnerability indicator Priority ranking Avg. Weight Social exposure 1 0.228 Physical exposure 2 0.150 Social sensitivity 2 0.150 Physical sensitivity 3 0.114 Social & physical adaptive capacity 4 0.083 Ecological exposure 5 0.037 Ecological adaptive capacity 6 0.015 Ecological sensitivity 7 0.013 69 8.2.2 Vulnerability assessment of the current situation Our results for the current situation (baseline scenario) showed that especially the sub-districts of the larger Qasabah Amman area in the east of the study catchment are highly vulnerable. This is particularly true for the districts of Al Yarmouk, Basman and Al Madeenah, where more than 20% of the area was classified as very vulnerable to flash floods (Figure 57). Figure 57. Distribution (%) of vulnerability categories for Exposure + Sensitivity and Adaptive Capacity for the districts of the study catchment in Amman. This area is characterized by a high degree of urbanisation, a highly vulnerable resident population (high sensitivity) with many disadvantaged individuals and refuges, and many physically exposed areas (areas in close proximity to a flood path). This is particularly the case for the downtown area of Amman, which is situated in a low-lying basin, surrounded by higher elevations. In addition to the topographical conditions that result in a generally higher exposure (see chapter 5) this area is also characterised by an inadequate drainage infrastructure and the highest proportion of impermeable surfaces (Awad, 2023), which prevent water from being absorbed into the ground, thereby increasing surface runoff. The urban development of downtown Amman has taken place over centuries without the implementation of modern flood management planning. The area has also become a settlement for refugees and a residential area for working class families (Potter et al., 2009; Ababsa and Daher, 2011), a social development linked to the relatively lower cost of living in the older parts of the city. The comparatively low income of residents in these areas results in a comparatively high social sensitivity to flash flood damage. Special attention was given to adaptive capacity, which in our assessment indicates the availability of open space for future implementation of measures. The highest adaptive capacity was found in the southwest in the districts of Umm QuṣayrAl-Muqābalīn and Marj al-Ḥamām with 54% and 37% of the area respectively, and in the northwest in the district of Ṣuwayliḥ with 45%. These areas have open space on public land, but also on private land owned by more affluent residents. Although these districts have less open land than in the past due to rapid urbanisation (Awad, 2023), there are still some undeveloped areas, especially on the outskirts of the city or in areas reserved for future projects. These small, scattered pockets of open land are often used for a variety of purposes, such as small-scale agriculture, informal recreational areas, or are left as vacant plots awaiting future development. On the outskirts of Marj al-Ḥamām, some land remains in its natural state or is used for small-scale agriculture. These areas are more common on the periphery, where urban development has not yet fully encroached. Final report 70 Figure 58. Combined vulnerability maps showing all domains for sensitivity and exposure (left) and for adaptive capacity (right) for the study area in Amman. In Amman, some open spaces are privately owned and reserved for future development, but open spaces are also available on already developed private properties belonging to better-off residents. This indicates great potential for the implementation of future measures in the private sector if sufficient incentives are created here. Our interviews with residents of Amman revealed that many citizens (37%) have already experienced flash flood damage to their property (Figure 59). Many are aware of possible measures (54%) to reduce the damage potential, and some have already implemented them (40%). Overall, the willingness to take further measures to reduce flash flood damage to their own properties in the future is very high (85%), provided it is financially feasible. Figure 59. Results of the Interviews with residents in Amman (n = 52) on local knowledge of flash floods and mitigation measures 71 8.3 Outlook: Vulnerability assessment Our integrated, spatially explicit vulnerability assessment combined different disciplinary perspectives with local knowledge. Despite the challenges posed by scarce data, we identified the areas most vulnerable to flash flooding, as well as those with the greatest adaptive capacity, for future implementation. The lack of data on critical infrastructure was offset by OpenStreetMap (OSM) data and our own measurements. However, basic data on critical infrastructure, social aspects and environmental aspects is still scarce and needs to be improved for a more detailed risk and vulnerability analysis. The SEVA results highlighted that the most vulnerable areas are located in the east of the studied watershed, including downtown Amman (the highest built-up area, with the highest sensitivity of residents and many exposed areas), whereas the highest adaptive capacity was detected on the outskirts of the city in the northwest and southwest (with open spaces on public and private land). Given the great potential of open spaces on private property and the willingness of citizens to implement measures, it is suggested that more incentives should be created to promote the implementation of future measures to reduce flash flood damage. However, given the high demand for housing and commercial space in Amman, any remaining open spaces are under pressure from real estate developers. This suggests that open spaces may continue to decrease as development progresses. Careful urban planning is therefore required to ensure that these areas retain some level of open space and that the development of such areas incorporates the future implementation of measures to reduce flash flooding and green infrastructure solutions. Blue-green infrastructure solutions help to manage and retain rainfall runoff, improving water infiltration. Our results can serve as a decision support tool for ongoing sustainable urban planning in the Amman Green City Action Plan (GCAP 2021), which seeks to enhance the city’s resilience to climate change by implementing measures to mitigate the impacts of extreme weather events, such as flooding. Final report 72 9 Integrated multi-scenario analysis Authors: Katja Brinkmann, Dörte Ziegler, Clara Hohmann, Peter Hoffmann, Christina Maus, Ahmad Awad, Daniel SchuhmannHindenberg, Hanna Leberke and Ahmad Bariq Allemyar Besides the rainfall hazard, exposure, sensitivity, adaptive capacity and vulnerability analysis for the current situation, we also explored possible future scenarios for the study region Amman integrating the results of the various research activities. Using a multi-scenario analysis, the effects of changes in heavy rainfall and land cover changes, as well as measures to decrease flash flood damages were simulated with hydraulic and hydrologic models (Figure 60). These results serve as a decision-support tool for urban planning, showing what could happen in urban areas if the effects of climate change and urbanisation intensify, and demonstrating the opportunities that implementing measures such as bluegreen infrastructure could provide in reducing the risk of flash flooding. The scenarios were defined collaboratively within the CapTain Rain consortium, incorporating scientific expertise and practical knowledge from Jordanian stakeholders. The rainfall and land cover change scenarios were primarily based on scientific knowledge incorporating expert opinions from Jordanian stakeholders. In contrast, the measures scenarios were based on a participatory planning process (Schumann-Hindenberg et al., 2025) and co-created with Jordanian stakeholders. Figure 60. Overview of the different scenarios within CapTain Rain that have been simulated using hydrological and hydraulic models 9.1 Key findings for Amman  Land use and land cover analysis showed that Amman´s urban and built-up surfaces increased 10fold from 1968 to 2021 and that with further urban development the simulated built-up areas in 2050 are expected to increase by around 70 %  The analysis on future changes in the distribution of precipitation showed a decrease of the total precipitation and an increase of extreme precipitation.  Heavy rainfall events shorter than 6 hours show a much stronger reaction to global warming than longer durations.  In future both the severity of intense rainfall and drought conditions in Jordan will have a higher impact on the humans in Jordan  The results of hydrologic and hydraulic modelling show that both climate change and urbanization will increase the hazard of flash floods in Jordan, i.e. flooded areas, water volumes and peak flows.  By relative comparisons, the hydrologic models could demonstrate past and future impacts of land use change on peak flow rates within Amman´s major watershed. Since the scarce soil data seems to indicate quite impermeable soil, the impact of urban development on peak flow rates is much smaller than anticipated. The results showed that changes of precipitation events (increases of intensity and/or duration, e.g. up to an annual rainfall amount in one event) influence the hazard of flash floods up to five times higher than land use changes (Hohmann et al. 2024).  The hydraulic simulation of the measures scenarios indicated that the selected adaptation measures have the potential to reduce inundation areas and potentially flash flood risk for the moderate rainfall scenario by 75%). 73 9.2 Development of multi-scenarios 9.2.1 Rainfall scenarios The rainfall scenarios should depict a range of frequent heavy rainfall events, e.g. once per year, but also the impact of future climate change. They should also demonstrate the consequences of a potential catastrophic event with a low probability, in order to evaluate the effectiveness of various adaptation measures. The German heavy rainfall index (Schmitt, 2015, DWA-M 119, 2018) was used as a reference point, for which rainfall data covering return periods from 1 to over 100 years would be required. However, the data available in Jordan is insufficient to link rainfall intensities to return periods. Therefore, the rainfall scenarios were limited to four: a “baseline” reference situation derived from a comprehensive analysis of historical heavy rainfall events; a climate change scenario derived from climate models (later defined as “moderate”); and two extreme events (one intense and one catastrophic) to cover a broad range of possible heavy rainfall events. To define the intensities and durations of heavy rainfall for these scenarios, a retrospective analysis of historical heavy rainfall events was conducted in AP 2. Then, the impact of climate change on the intensity and duration of heavy rainfall in Jordan was analysed. This enabled us to define the potential range of heavy rainfall intensities and durations in Jordan. The following section explains those analyses: A comprehensive retrospective analysis of historical heavy rainfall events in Jordan was difficult to conduct due to the limited availability of high-resolution precipitation data. Various reanalysis products were used as a basis to identify recent extreme rainfall events including their spatial rainfall patterns and to compare them with available station data. This enabled a baseline scenario to be described for Amman and Wadi Musa, where a significant effect (flash flood) was observed on critical infrastructure and society. Figure 61 shows a screenshot of an event table sorted by date, derived in AP2 for Amman and Wadi Musa, which compares rainfall patterns extracted from different open data sets. The most recent event in Amman occurred on 28 February 2019. Amman Wadi Musa Figure 61. Sorted table of heavy rainfall events given as values and patterns for Jordan using different reanalysis products: UERRA, ERA5, W5E5. In addition, the respective large-scale circulation patterns (Z500) are given. For the baseline scenario, we selected the most intense 5.5 hours of the relatively short February 2019 event, which was recorded by five stations from the Jordan Meteorological Department (JMD) at an hourly resolution. Additionally, time series with a resolution of 5 minutes for this event were obtained from a study commissioned by UN-Habitat (2020). Such an event occurs approximately once a year in Amman. This event exhibited significant spatial and temporal variations; for instance, 58 mm of rainfall was recorded at the Amman Airport station in four hours, which is nearly equivalent to an event with a return period of 25 years, according to the UN-Habitat (2020) study (60 mm/3 h). At one station, however, only 19 mm of rainfall was measured in this time period. The timing and spatial pattern of the event are also accurately reflected in the GSMaP satellite rainfall estimates (see Figure 62). However, the magnitude of the event is clearly underestimated compared to rain gauges. Nevertheless, it was one of the top 10% of events in the last 20 years. Final report 80 Figure 66. Focus area Downtown Amman with a) inundation map of the baseline simulation, water levels in blue, and the differences maps to the baseline in purple of b) moderate scenario (20 % more rainfall than baseline), c) intense scenario (maximum station for the whole catchment, 136 mm in 27 h), d) catastrophic scenario (300 mm in 27 h), the UNESCO world heritage site, Roman theater, is marked with a green star (Hohmann et al. 2024). In order to demonstrate the impact of possible measures (impact assessment), the measures scenarios were simulated for the focus area Marj Al Hamam, a complete sub-catchment of the study region. The identified measures (see Table 13) were positioned along the main flow paths in available open spaces, e.g. beside streets. All measures provide a certain retention volume. This volume was calculated by multiplying the free area in m² by a theoretical depth of 0.5 m; however, real measures may differ from this depth, e.g. for technical and/or covered retention basins. Results of the ‘public space’ scenario: In this scenario, only public spaces were used to allocate measures. The total potential usable space was 432,662 m², corresponding to 4% of the total focus area. Eleven measures were identified and allocated in public spaces: one retention basin, eight infiltration trenches and two bioswales. The measures require an area of 49,048 m² in total, corresponding to 11% of the potential usable space. Nearly half of this area is occupied by infiltration trenches situated along streets, particularly main roads. The measures are collectively capable of holding back a volume of 16,376 m³. Hydraulic simulations of different rainfall scenarios show peak runoff reductions of 34% for the baseline scenario, 21% for the intense scenario, and 6% for the catastrophic rainfall scenario. As most of the measures are located in the lower part of the focus area, the greatest impact in terms of reducing runoff volumes and flash flood damage is seen further downstream. Nevertheless, they also have an impact on critical areas within the focus area itself, such as the significant inundation potential at the intersection of Dead Sea Road and Airport Road. 81 Table 14. Summary of the results of the measures scenarios for the focus area 1 – Marj Al Hamam. Character Very high percentage of agricultural and open land. Focus Public space scenario Public and private space scenario Identified measures 11 measures were identified: 1 retention basin, 8 infiltration trenches along roads and 2 bioswales. 23 measures were identified: the 11 one of the public space scenario as well as 12 additional ones on private land. Overall, they include 5 retention basins, 7 multifunctional areas, 8 infiltration trenches, 2 bioswales and 1 pipe infiltration trench Required Area 49,048 m² 179,524 m²: out of this 130,476 m² located on private land, (multifunctional areas and retention basin have 82 % of 179,524 m²) hold back a volume 16,376 m³ 36,209 m³ (19,833 m³ only private) Potential useable space 3 % of total CA% 26 % of the total CA Implementation area Implementation in 1% of total CA Implementation in 10% of the potential area (2% of total CA) Result simulation Peak run off reduction in rainfall scenarios 34 % for baseline, 21 % for intense, 6 % for catastrophic event 75 % for base scenario, intense 46 %, catastrophic event 13 % Results of the ‘public and private space scenario’: This scenario focuses on private areas and public land. The potential usable space amounts to 2,001,832 m², corresponding to 20% of the focus area. Within this scenario, 23 measures have been identified: the 11 measures from the public space scenario, plus 12 additional measures on private land. The measures include five retention basins, seven multifunctional areas, eight infiltration trenches, two bioswales and one pipe infiltration trench. All of the measures require an area totalling 179,524 m², 130,476 m² of which are located on private land. The measures “multifunctional area” and “retention basin” are particularly noteworthy as they represent four new locations and require most of the area (82%). Overall, the measures require an area of 179,524 m², corresponding to 10% of the potential usable space and 2% of the focus area. The measures can hold back a volume of 36,209 m³. Hydraulic simulations of different rainfall scenarios show peak runoff reductions of 75% for the baseline scenario, 46% for the intense scenario and 13% for the catastrophic event. Final report 82 Figure 67. Impact (decrease of water level) of the ‘public and private space scenario’ in combination with the intense rainfall scenario modelled with the hydraulic model HE2D/FOG2D. The results show that the measures in the ‘public and private space scenario’ have a positive impact. They reduce inundated areas within the sub-catchment. Regarding the runoff curves, a reduction is especially evident in the baseline rainfall scenario. However, looking at the catastrophic scenario with much more rainfall, the reduction is potentially small and does not reduce any of the larger runoff peaks. This highlights the importance of catastrophe management: protection against catastrophic flash floods is not possible, so early warning systems and evacuation measures are required instead. 9.5 Outlook: Multi-scenario analysis The hydraulic simulation clearly showed how an increase in heavy rainfall events could affect inundation areas and put more urban areas at risk. These spatially explicit datasets and maps are extremely useful for storm water management, particularly for planning future measures to prevent flash flooding, as well as for early warning systems and evacuation planning. The rainfall and land cover change scenarios were primarily based on scientific knowledge incorporating expert opinions from Jordanian stakeholders. In contrast, the measures scenarios were developed through a participatory planning process in collaboration with Jordanian stakeholders. The resulting scenarios therefore reflect local possibilities and can be incorporated directly into urban planning work. The measures scenario incorporating bluegreen infrastructure demonstrated significant potential for mitigating floodplain water levels, particularly when measures are implemented in both public and private spaces, where substantial open space is available (see Chapter 8). To promote the implementation of future measures on private property, incentives should be created and incorporated into urban planning. The multi-scenario results serve as a decision-support tool for ongoing sustainable urban planning within the Amman Green City Action Plan (AECOM Limited, 2021). Combined with other modelling approaches and additional data on future demographic trends and land use changes with different building patterns, these simulation results could be further refined and improved. This would enable the two greatest challenges in future urban planning – urbanisation and the increasing risk of flash flooding – to be considered in an integrative way and taken into account in planning. To better assess the scenarios’ potential to reduce vulnerability, the vulnerability assessment presented in Chapter 8 could be carried out for different future scenarios. For the vulnerability component ‘Exposure’, which depicts the degree to which a subject (inhabitant, building or ecosystem) is exposed to flooding in the event of a flash flood, 83 this is feasible. However, detailed data on demographic and socio-economic characteristics and their modelling are required to project future demographic and economic trends for the other vulnerability components (sensitivity and adaptive capacity). 10 Key messages and further research need Authors: Katja Brinkmann and Dörte Ziegler CapTain Rain delivered methods and climate services for flash flood prediction and prevention in a participatory and target group-oriented manner. The overall objective was to improve climate change adaptation in Jordan and reduce vulnerability regarding flash flood events. The project focused on analysis and planning tools to identify options for improving flash flood management and thus climate change adaptation in Jordan. The analysis, planning and early warning tools were developed together with stakeholders so that they can be used in the future in the pilot areas of Amman and Wadi Musa, as well as in other cities in Jordan and in other semi-arid countries in the MENA region. The transdisciplinary research approach integrated scientific and practical knowledge using different research methods such as climate analysis, hydrological and hydraulic modelling, remote sensing techniques, participatory GIS methods, integrated vulnerability assessment, stakeholder workshops, expert interviews and scenario analysis. Despite challenges such as scarce and poor-quality data, regionally adapted solutions for retaining heavy rainfall and reducing flash flood risks were developed in close cooperation with Jordanian stakeholders and practice partners, taking into account scientific and local, practice-oriented findings. These climate services include flash flood hazard and risk maps, tools to improve the prediction of flash floods and recommendations for promising adaptation strategies and early warning systems that are suitable for a semi-arid country with limited monitoring and modelling resources. In order to transfer and further develop the results obtained on climate services and products as decision support in practice, we recommend integrating the climate services on flash floods into urban planning processes. With regard to climate service products, the key results and recommendation of CapTain Rain can be summarized as follows: Rainfall hazard  A climate service portal and future heavy rainfall scenarios that incorporate climate change effects were developed for Jordan.  Analysis of climate model simulations showed a decrease in the number of extreme rainfall events in Jordan, while their intensity would increase.  The integration of critical circulation patterns in weather forecasts would improve the prediction of heavy rainfall events in Jordan. Exposure and Sensitivity  The hydrological models (HEC-HMS, RRI) displayed the runoff curves and allowed an initial assessment of possible changes in heavy rainfall events and impacts of urbanization.  The hydraulic model combined with the spatial analysis of damage potential allowed to show inundation areas and develop flood hazard maps, damage potential maps and flash flood risk maps; it can be used to model impacts of possible adaptation measures.  The hazard and risk maps should be used to communicate potential flooding areas to residents, the public, and to institutions of crisis management and to prioritize measures for climate adaptation. Adaptive capacity: Adaptation measures  A participatory process for the selection and localisation of measures in urban and landscape planning was applied showing the potential for the implementation of blue-green infrastructure  Maps, infocards on measures and a guideline were developed for local decision makers. Final report 84  Blue-green infrastructure should be adapted to the dry climate in Jordan so that evaporation losses and irrigation requirements are minimized. Climate and water data portal  A digital platform to manage rainfall data and to include weather forecasts and improve early warning was adapted to Jordan. Data on climate predictions for Jordan were integrated in the climate service portal.  The use of high-resolution open data and the sharing of weather observations and forecasts among the Jordanian government agencies would improve flood warnings Vulnerability assessment  A trans-disciplinary approach to assess vulnerability towards flash floods was developed, integrating ecological, social and physical aspects.  Vulnerable areas are particularly located around Downtown Amman (highest built-up area, highest sensitivity of residents and many exposed areas); the highest adaptive capacity was found in the outskirts, where there is still space for the implementation of measures, especially on private land  The assessment results and maps serve as decision-support when selecting measures, but also for flood warning and alarm plans. Multiscenario analysis  Changes of rainfall intensity and duration have the largest influence on runoff, since soils have low infiltration capacities.  The catastrophic rainfall scenario serves as stress test to prepare for potential catastrophes.  The simulated scenarios can be used for assessing the impacts of measures on risk reduction under different rainfall scenarios and serve as decision support In Jordan, responsibilities and resources for the management of heavy rain and flash flood events have been assigned and largely clarified. However, there is a lack of resources and continuity, as well as limited access to data. In this context, structural challenges include donor and project dependency. It is crucial to reinforce the bond between state entities and local communities, a process that can be facilitated through the implementation of projects. For the proactive management of flash flood risks, it is essential to establish effective coordination between government agencies. The establishment of a leading agency or state entity would facilitate this process. Furthermore, a transdisciplinary understanding of data requirements and data access across different institutions is necessary. This can be achieved by integrating and sharing data in data portals. This approach would enhance prediction and early warning capabilities, as well as facilitate more targeted risk management and disaster management in the event of a flash flood. Obtaining the data necessary for flash flood risk management remains challenging due to insufficient data analysis and maintenance processes, as well as difficulties accessing the data. Therefore, the quality of data, including rainfall data, digital elevation information, soil data and building information, should be enhanced. However, this requires resources to be allocated to integrated data management and maintenance. For example, installing and maintaining gauges at designated points within watersheds would be a significant step towards improving data on peak flow rates and water tables. Furthermore, utilising open access data, including satellite and OpenStreetMap data, helps to overcome data barriers. It is imperative that existing data gaps are filled and that the data is standardised. This will ensure that future climate, hydraulic and hydrological models achieve a higher level of reliability in their modelling results. Furthermore, the establishment of standards and procedures related to flash flood management is of paramount importance. This involves the development of methodologies into standards with regard 85 to meteorological and hydrological monitoring, as well as the formulation of standards for urban planning and development and building codes for sewers, buildings, or streets. The Greater Amman Municipality (GAM) and the Petra Development and Tourism Regional Authority (PDTRA) are currently implementing initiatives to adapt to climate change in Amman and Wadi Musa, respectively. These initiatives prioritise the implementation of blue-green infrastructure, with rainwater harvesting being of particular significance. When seeking to retain water during more extreme rainfall events, it is crucial to consider the complex effects and interactions of evapotranspiration and groundwater recharge. Given the current limitations of available data, it is recommended that further studies be conducted on soil conditions and groundwater recharge rates. A participatory planning process is advised to facilitate the implementation of blue-green infrastructure, commencing with a joint understanding of planning goals and extending to the utilisation of a comprehensive range of multifunctional measures to adapt to flash floods. This process requires the involvement of various stakeholders in formulating planning goals and measures, which are then formalised in flash flood risk management plans. Thereafter, recurrent assessment of the implementation of measures and their impacts is required. The Jordanian partners and associated stakeholders involved in the CapTain Rain project benefited from capacity building, for example through webinars, which enabled them to enhance their tools and methods for managing heavy rain and flash floods. Cooperation between Jordan and Germany in addressing climate change was strengthened through various visits to Jordan, stakeholder workshops, meetings, virtual conferences and, finally, a joint study tour to Germany in August 2024. There is great interest in further cooperation with Germany, the MENA region and Europe in the future. To secure knowledge transfer, the CapTain Rain project has presented project results to institutions and stakeholders involved in development cooperation and flash flood and water management. These include a donor round moderated by the German Embassy in Jordan, KfW and GIZ, the Swiss Development Cooperation, UN-Habitat and USAID. This has resulted in the formation of a new network, and CapTain Rain is now part of the Water Disaster/Flood Mapping, Analysis and Mitigation Working Group of SDC, regularly attending its meetings. Final report 86 11 References Ababsa, M., Daher, R.F., 2011. Villes, pratiques urbaines et construction nationale en Jordanie. Presses de l’Ifpo, available at https://doi.org/10.4000/books.ifpo.1675. Acosta-Coll, M., Ballester-Merelo, F.,Martinez-Peiró, M., de La Hoz-Franco, E., 2018. Real-Time Early Warning System Design for Pluvial Flash Floods – A Review. In: Sensors (Basel, Switzerland), 18, 7. AECOM Limited, 2021. Amman Green City Action Plan. Available at https://www.amman.jo/site_doc/AmmanGreen2021.pdf. Alhasanat, H., 2014. Flash Flood Assessment for Wadi Mousa City-Jordan. In: Procedia Economics and Finance, 18: 675–683. Alhasanat, H. A., 2017. Early Warning System Guidance to Mitigate Flash Flood Impacts in Petra Region, Jordan. Salford, UK. Al-Qudah, K. A., 2011. Floods as Water Resource and As a Hazard in Arid Regions – A Case Study in Southern Jordan. In: Jordan Journal of Civil Engineering, 5:148–161. Al-Qudah, K. A., Abdelal, Q., Hamarneh, C., Abu-Jaber, N., 2016. Taming the torrents – The hydrological impacts of ancient terracing practices in Jordan. In: Journal of Hydrology, 542: 913–922. Al-Weshah, R. A., El-Khoury, F., 1999. Flood Analysis and Mitigation for Petra Area in Jordan. In: Journal of Water Resources Planning and Management, 125: 170–177. Amin, M. T., Mahmoud, S. H., Alazba, A. A., 2016. Observations, projections and impacts of climate change on water resources in Arabian Peninsula – Current and future scenarios. In: Environmental Earth Sciences, 75: 864. Ara Begum, R., R. Lempert, E. Ali, T.A. Benjaminsen, T. Bernauer, W. Cramer, X. Cui, K. Mach, G. Nagy, N.C. Stenseth, R. Sukumar, and P. Wester, 2022. Point of Departure and Key Concepts. In: Climate Change 2022. Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [H.-O. Pörtner, D.C. Roberts, M. Tignor, E.S. Poloczanska, K. Mintenbeck, A. Alegría, M. Craig, S. Langsdorf, S. Löschke, V. Möller, A. Okem, B. Rama (eds.)]. Cambridge University Press, Cambridge, UK and New York, NY, USA, pp. 121–196, doi:10.1017/9781009325844.003. Awad, A., 2023. Analysis of the spatial-temporal dynamics of land-use changes using a mixed-method approach: A Case study from Amman, Jordan. M.Sc. thesis at the faculty of life Sciences, technical university of Munich, Germany. Awad, A., Hohmann, C., Maus, C., Abu Hamour, W., Al Naimat, M., Brinkmann, K. (in preparation). Understanding urbanization’s impact on flash flood risks from past to future: A Case study of a rapidly growing MENA city. CADRI Partnership Secretariat [ed.], 2018. Capacity Assessment of the Disaster Risk Management System in Jordan – Draft. Available at https://data2.unhcr.org/en/documents/download/67193. Olmedo, C.M.T., Paegelow, M., Mas, J.-F., Francisco, F. (Eds.), 2018. Geomatric approaches fo modelling land change scenarios: Springer (Lecture Notes in Geoinformation and Cartography). Available online at https://doi.org/10.1007/978-3-319-60801-3. Copernicus, 2021. Climate Change Service, Climate Data Store – CMIP6 climate projections. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: 10.24381/cds.c866074c. Copernicus, 2019a. Climate Change Service, Climate Data Store – CORDEX regional climate model data on single levels. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI:10.24381/cds.bc91edc3. Copernicus, 2019b. Climate Change Service, Climate Data Store – Complete UERRA regional reanalysis for Europe from 1961 to 2019. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: 10.24381/cds.dd7c6d66. CORDEX: Copernicus Climate Change Service, Climate Data Store, 2019. CORDEX regional climate model data on single levels. Copernicus Climate Change Service (C3S) Climate Data Store (CDS). DOI: 10.24381/cds.bc91edc3 (Accessed on 31-May-2024). Dayan, U., Nissen, K., Ulbrich, U., 2015. Review Article: Atmospheric conditions inducing extreme precipitation over the eastern and western Mediterranean. Nat. Hazards Earth Syst. Sci., 15: 2525–2544. Depietri, Y. 2020. The social–ecological dimension of vulnerability and risk to natural hazards. Sustain Sci 15: 587–604. https://doi.org/10.1007/s11625-019-00710-y. DWA, 2016. German Guideline, DWA-Merkblatt 119. Risikomanagement in der kommunalen Überflutungsvorsorge für Entwässerungssysteme bei Starkregen (Risk management in municipal flood prevention for drainage systems during heavy rain), DWA German Association for Water, Wastewater and Waste, e.V., Hennef, ISBN: 978-3-88721-393-0. 87 Emergency Events Database, 2022. Available at: https://www.emdat.be/ El Alfy, M., 2016. Assessing the impact of arid area urbanization on flash floods using GIS, remote sensing, and HEC-HMS rainfall–runoff modeling. Hydrology Research, 47: 1142–1160. DOI: 10.2166/nh.2016.133. Field, C. B., Barros, V.; Stocker, T. F., Dahe, Q., 2012. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation – Special Report of the Intergovernmental Panel on Climate Change, Cambridge University Press. New York. Gruntfest, E., Handmer, J., 2001. Coping with Flash Floods, Springer Netherlands; Imprint: Springer. Dordrecht. Hamdy, O, Abdelhafez, M. H. H., Touahmia, M., Alshenaifi, M., Noaime, E., Elkhayat, K., Alghaseb, M., Ragab, A., 2023. Simulation of Urban Areas Exposed to Hazardous Flash Flooding Scenarios in Hail City. Land 12: 353. DOI: 10.3390/land12020353. Hashemite Kingdom of Jordan, 2014. Jordan´s Third National Communication on Climate Change. – Report, submitted to the United Nations Framework Convention on Climate Change. Deposit Number at the National Library (2014/11/5178). HEC-HMS, 2024. Hydrological Modeling System of the Hydrologic Engineering Center within the U.S. Army Corps of Engineers (HEC-HMS), available at www.hec.usace.army.mil/software/hec-hms/ Hersbach, H., et al., 2020, The ERA5 Global Reanalysis, Quaterly Journal of the Royal Meterological Soceity 146: 1999-2049, doi: 10.1002/qj.3803 Hohmann et al. 2024. Hydrological approaches to assess flash flood risks in data scarce and climate-change affected regions like Jordan. Forum für Hydrologie und Wasserbewirtschaftung. Bd. 45, Berlin, S. 25 - 36., https://doi.org/10.14617/for.hydrol.wasbew.45.24 Itwh, 2024. Software package: Urban Flash Flood with FOG 2D and HYSTEM-EXTRAN 2D, Institute for technical scientific hydrology GmbH, https://itwh.de/en/software-products/desktop/urbane-sturzfluten/ Kautz, L.A., Martius, O., Pfahl, S., Pinto, J. G., Ramos, A. M., Sousa, P. M., Woollings, T., 2022. Atmospheric blocking and weather extremes over the Euro-Atlantic sector – a review, Weather Clim. Dynam., 3: 305–336, https://doi.org/10.5194/wcd-3-305-2022. Kubota, T., K. Aonashi, T. Ushio, S. Shige, Y. N. Takayabu, M. Kachi, Y. Arai, T. Tashima, T. Masaki, N. Kawamoto, T. Mega, M. K. Yamamoto, A. Hamada, M. Yamaji, G. Liu, Oki, R., 2020: Global Satellite Mapping of Precipitation (GSMaP) products in the GPM era, Satellite precipitation measurement, Springer, https://doi.org/10.1007/978-3-030-24568-9_20. Kundzewicz, Z. W., Kanae, S., Seneviratne, S. I., Handmer, J., Nicholls, N., Peduzzi, P., Mechler, R., Bouwer, L. M., Arnell, N., Mach, K., Muir-Wood, R., Brakenridge, G. R., Kron, W., Benito, G.; Honda, Y., Takahashi, K., Sherstyukov, B., 2013. Flood risk and climate change – Global and regional perspectives. In: Hydrological Sciences Journal, 59:1–28. Lehmann, J., Coumou, D., Frieler, K., 2015. Increased record-breaking precipitation events under global warming. Climatic Change, 132: 501-515. Lelieveld, J., Hadjinicolaou, P., Kostopoulou, E., Chenoweth, J., El Maayar, M., Giannakopoulos, C., Hannides, C., Lange, M. A., Tanarhte, M., Tyrlis, E., Xoplaki, E., 2012, Climate change and impacts in the Eastern Mediterranean and the Middle East. Climatic Change, 114: 667–687. Lelieveld, J., Hadjinicolaou, P., Kostopoulou, E., Giannakopoulos, C., Pozzer, A., Tanarhte, M., Tyrlis, E., 2014. Model projected heat extremes and air pollution in the eastern Mediterranean and Middle East in the twentyfirst century. Regional Environmental Change, 14: 1937–1949. Mayring, P., 2010: Qualitative Inhaltsanalyse – Grundlagen und Techniken, 11. revised version. Beltz Verlag, Weinheim, Basel. Niyibizi, A., Mpeirwe, A., Ajambo, S., 2013: Vulnerability assessment for rural settings: Applicability to developing countries, in: Leal Filho, W. (Ed.), Climate Change and Disaster Risk Management. Springer, Berlin, Heidelberg, pp. 273–292. Potter, R.B., Darmame, K., Barhamb, N., Nortcliff, S., 2009. “Ever-growing Amman”, Jordan: Urban expansion, social polarisation and contemporary urban planning issues. Habitat International, 33: 81–92. Ross, C.W., Prihodko, L., Anchang, J.Y., Kumar, S.S., Ji, W., Hanan, N.P., 2018. Global Hydrologic Soil Groups (HYSOGs250m) for Curve Number-Based Runoff Modeling. ORNL DAAC, Oak Ridge, Tennessee, USA. https://doi.org/10.3334/ORNLDAAC/1566. Sharma, T., 2024. A comparative scenario analysis of WSUD measures for stormwater management in Amman, Jordan. M.Sc. thesis at the HafenCity Universität Hamburg, Germany. Schoener, G., 2022. Impact of urbanization and stormwater infrastructure on ephemeral channel transmission loss in a semiarid watershed. Journal of Hydrology: Regional Studies, 41, DOI: 10.1016/j.ejrh.2022.101089. Final report 88 Schmitt, T., Krüger, M., Pfister, A., Becker, M., Mudersbach, C., Fuchs, L., Hoppe, H., Lakes, I., 2018. Einheitliches Konzept zur Bewertung von Starkregenereignissen mittels Starkregenindex. Korrespondenz Abwasser, Abfall, 65: 113-12. Schuck-Zöller, S., Cortekar, J., Jakob, D., 2018. Evaluation Transdisziplinärer Forschung und deren RahmenbedingungenVorüberlegungen zur Nutzung im Bereich von Klimaservice. fteval Journal 45: 28–37. Schumann-Hindenberg, D., Awad, A., Brinkmann, K., Fölster, L., Winker, M., 2025. Integrated planning for reducing flash flood risk through blue-green infrastructure: A guideline using the city of Amman in Jordan as an example. DOI: 10.5281/zenodo.17351802 Thaler, T., Levin-Keitel, M., 2016. Multi-level stakeholder engagement in flood risk management—A question of roles and power: Lessons from England. Environmental Science & Policy, 55: 292–301. The Jordan times, 2019. Flooding results in road closures, complaints and rescues on Thursday. Aivailable at https://www.jordantimes.com/news/local/flooding-results-road-closures-complaints-and-rescues-thursday. Thiault, L., S. D. Jupiter, J. E. Johnson, J. E. Cinner, R. M. Jarvis, S. F. Heron, J. M. Maina, N. A. Marshall, P. A. Marshall, Claudet J., 2021. Harnessing the potential of vulnerability assessments for managing socialecological systems. Ecology and Society 26:1. https://doi.org/10.5751/ES-12167-260201 UN-Habitat, 2020. Developing a Preliminary Design for Flood Mitigation and Performing a Flood Risk Assessment and Flood Hazard Mapping for Downtown Amman. Report available at https://procurement-notices.undp.org/view_file.cfm?doc_id=219741. USDA (United States Department of Agriculture), 2010. Natural Resources Conservation Service – Part 630 Hydrology – National Engineering Handbook – Chapter 15 Time of Concentration, available at https://rashms.com/wp-content/uploads/2021/11/TOC_NRCS-Part-630-Ch-15.pdf. Vries, A. J. de, Ouwersloot, H. G., Feldstein, S. B., Riemer, M., El Kenawy, A. M., McCabe, M. F., Lelieveld, J., 2018. Identification of Tropical-Extratropical Interactions and Extreme Precipitation Events in the Middle East Based On Potential Vorticity and Moisture Transport. Journal of Geophysical Research: Atmospheres, 123: 861–881. Westra, S., Alexander, L. V., Zwiers, F. W., 2013. Global Increasing Trends in Annual Maximum Daily Precipitation. In: Journal of Climate, 26: 3904–3918. WeatheringRisk (2022): Climate Risk Profile for Jordan. Available at https://www.weatheringrisk.org/sites/default/files/document/Climate_Risk_Profile_Jordan.pdf. Winker, M., Frick-Trzebitzky, F., Matzinger, A., Schramm, E., Stieß, I., 2019. Die Kopplungsmöglichkeiten von grüner, grauer und blauer Infrastruktur mittels raumbezogener Bausteine. Ergebnisse aus dem Arbeitspaket 2, netWORKS 4. netWORKS-Papers, 34. Berlin: Deutsches Institut für Urbanistik Difu, Germany. Available at https://networks-group.de/de/publikationen/networks-paper.html (accessed 6.12.2024). Winker, M., Matzinger, A., Anterola, J., Frick-Trzebitzky, F., Pillen, J., Schramm, E., 2022. Infokarten für die Planung blau-grün-grauer Infrastrukturen. In: Forschungsverbund netWORKS (Hg.): Frankfurt am Main. Ministry of Water and Irrigation, 2023. National Water Strategy 2023 – 2040. Available at https://www.mwi.gov.jo/EBV4.0/Root_Storage/AR/EB_Ticker/National_Water_Strategy_20232040_Summary-English_-ver2.pdf, last accessed on 20.06.2024. WION, 2023. ‘Unprecedented phenomenon’: In just one day, Greece surpasses average annual rainfall in Magnesia region. Available at https://www.wionews.com/world/unprecedented-phenomenon-in-just-one-daygreece-surpasses-average-annual-rainfall-in-magnesia-region-633011, last accessed on 22.03.2024. Zhang, Y., Wang, Y., Chen, Y., Liang, F., Liu, H.,2019. Assessment of future flash flood inundations in coastal regions under climate change scenarios-A case study of Hadahe River basin in northeastern China. The Science of the total environment, 693: DOI: 10.1016/j.scitotenv.2019.07.356. Zittis, G., Almazroui, M., Alpert, P., Ciais, P., Cramer, W., Dahdal, Y., Fnais, M., Francis, D., Hadjinicolaou, P., Howari, F., Jrrar, A., Kaskaoutis, D.G., Kulmala, M., Lazoglou, G., Mihalopoulos, N., Lin, X., Rudich, Y., Sciare, J., Stenchikov, G., Xoplaki, E., Lelieveld, J., 2022. Climate Change and Weather Extremes in the Eastern Mediterranean and Middle East. Reviews of Geo-physics, 60: e2021RG000762.