Please describe how your professional (and/or personal) experience relates to space technologies and their applications to water resources management.
I am an expert in hydroinformatics, mainly involved in research projects and research supervision of MSc and PhD students. My research focusses on physically based models for inland waters (rivers and lakes). One of the major fields where modelling is used in water resources is flooding. In order to have adequate representation of floods, most models require large amounts of data, both for model building and model usage. This is especially true for pluvial flood modelling, which focuses on pluvial floods, or flash floods, which occur in areas where flow gauging stations are rare. Remote sensing (RS) technologies are a part of the solution, as they offer spatially distributed information.
Please tell us more about your current project, EIFFEL – “Revealing the role of GEOSS as the default digital portal for building climate change adaptation and mitigation applications”. How will the proposed novel methods (super resolution, data fusion) contribute to addressing water and land use management needs?
EIFFEL, which is a Horizon 2020 European Union (EU) project, involves several European research institutions and universities that work together to contribute to Europe’s implementation of the Paris Agreement (PA) for Climate Change (CC). As part of the EIFFEL project, my IHE Delft colleagues and I will demonstrate how to apply modelling for the CC adaptation and mitigation applications for small catchments.
In addition to national, European and global policies that are contributing to this overarching goal such as the EU Water Framework Directive, the EU Floods Directive, the European Green Deal, and the Sendai Framework for Disaster Risk Reduction (SFDRR); the Copernicus programme in 2014 launched the Climate Change Service (C3S), for “providing information to increase the knowledge base to support adaptation and mitigation policies”. C3S contributes to Essential Climate Variables (ECVs), climate analyses and indicators at temporal and spatial scales relevant to adaptation and mitigation strategies for EU sectoral and societal benefit areas. Though relevant big data exists, the fragmentation of temporally and spatially high-quality data to monitor and support CC measures remains a barrier. This is due to the lack of dynamic and context-aware mechanisms that could exploit relevant results and existing Earth Observation (EO) datasets, including satellite, in-situ, crowd-sourced and complementary sources. The models that we use are able to apply any of GEOSS dataset, but as I mentioned, in EIFFEL we will showcase these for particular catchments.
Recently you were involved in the SCENT project - Smart Toolbox for Engaging Citizens into a People-Centric Observation Web. Please tell us more about the project and share with us the opportunities and challenges of using data contributed by citizens in flood mapping and modelling.
The SCENT research project, funded by EU under Horizon 2020 programme, explored how citizens can be engaged in data collection – how they can become the ‘eyes’ of policy-makers. The project developed a gamified smartphone application to collect data on land use, water levels and velocities in the form of photos and videos.
The main research question was whether such collected data can contribute to the improvement of a flood model. The collected citizen data in SCENT proved that such data is effective. It found that there was a sufficient amount of well-distributed data from crowdsourcing. A comparative analysis of different data sources indicated that crowdsourced data have value as they complement the limited data available from classical sources in the modelling context. Possible usages of crowdsourced data in the modelling of floods include validation datasets that calibrates the model with the measured data, or vice-versa. Fine-tuning the model to use both measured and crowdsourced datasets in combination improves its performance, and demonstrates that crowdsourced data is a valid dataset that can contribute to the modelling process.
These initial results are promising; however, many challenges remain. Will crowdsourcing approaches gather a sufficient amount of data at the right time and the right place? Will there be continuity in applying the methods for collecting crowdsourced data after the main project funding stops? Therefore, there is a need for better involvement of the citizens in these actions. Authorities would be well-advised to invest in such actions and continuous campaigns, in projects that are close to local communities’ concerns.
How will the Water scenarios For Copernicus Exploitation (WaterForCE) project facilitate closer cooperation between in-situ, remote sensing (RS) and modelling communities?
In the project’s first two years, we have analysed current and coming policies, end-users needs, innovation needs, and the need to support water-related SDG’s. A state-of-the-art gap analysis and recommendations were done on Copernicus and other services supporting water management, with focus on flood, drought, water allocation, SDG6 and water accounting. We particularly focused on data that represent human interactions and interventions (e.g., dam operations, irrigations, etc.). This improved our understanding of how models using RS data can improve adaptive decision-making and policy implementation, as well as technical requirements for future needs. In short, we have engaged many stakeholders and raised awareness about what can be done with the available data. For example, a webinar in February 2022 discussed how EO can contribute to determine the status of SDG6 indicators and work for their improvement – have a look at the SDG 6 Clean water and Sanitation webinar recordings here.
What would you say are the main Earth observation gaps identified with regards to water and how could they be addressed?
There is a wealth of information available in terms of Earth observation, and yet there is a need for more. Computer processing and data storage capacities are advancing and so is the capability of professionals to use data for better water management. We need to look for a better frequency of spatial and temporal measurements and for algorithms that will process and offer services coming from this data in simpler and ready- to-use formats.
With a background in hydroinformatics, could you elaborate on the development of models? What are the most important models used in hydrology using EO data? What are the main ingredients and decisions to be taken to develop well-fitted models? How do you start, how do you know you need to adjust, and how can you assess if a model fits your region good enough to use it?
Models in water resources cover hydrology, hydraulics, water allocation - there is no single modelling tool that is the most important. What matters is the objective of the modelling and the problem that needs to be solved – based on this, the best tool to address the needs can be chosen. EO provide a wealth of data to improve the capability to model simulations of what is happening. This is a good starting point to explore possible solution for the given problem. No modelling can be done without talking with those who are trying to find a solution for their problem or without talking with the stakeholders in a region and involving them in the process. For a model to be successfully used by a decision maker, the obtained results must be checked with observed data and presented to stakeholders – they need to be validated with users and with those who will benefit from the tested solutions in a model.
A lot of your work is focused on capacity building for students and water professionals. Your aim is to enable them to research on large and complex questions, to use big complex data structures and to train their skills in using relevant tools to handle the data. Why is this important for water resources management?
Water is an important resource for everyone. We need it in all aspects of our life, for multiple uses, including drinking, food, energy, and we also need to focus on hazard reduction (droughts or floods). What I see as most important is a world in which people manage their water and environmental resources sustainably and equitably – this is also the mission of my employer, IHE Delft Institute for Water Education. The first step to achieving such a dream goal is educating and training professionals to 1) expand their knowledge base through “complex questions”, 2) use the latest available technologies and data, and 3) build the capacity of sector organizations, knowledge centres and other institutions active in the fields of water, the environment and infrastructure. This applies to all, no matter the country they are coming from.
I hope global efforts toward this goal will be accelerated at the UN Water Conference to be held March 2023 in New York.
Can you elaborate on your research findings about the importance of citizen science data for SDG 6? How can it be leveraged for informed and effective decision making? What do you need to consider working with citizen science data to derive scientifically accurate and relevant data?
I think citizen science is very powerful, as it helps citizens not only become aware of environmental needs, but to serve as the “eyes” of the decision makers. In this way, citizens can help define needs well and work towards meeting these needs. At the beginning, citizen science data may not be as accurate data from a sensor, but sensors are expensive, and any extra data is helpful. As time passes and citizens gain experience in collecting data, the quality also improves a lot. Although challenging, the implementation of citizen science has demonstrated its usefulness. It can help improve many aspects related to environmental monitoring and in the end lead to better protected and managed delta areas.
What are the current software tools used for water resources modelling and web-based decision support systems? What are their advantages and limitations?
There are many tools used for water resources modelling, all very good. Which tool to use depends on the user’s expertise, and on the problem to be solved. The most important is the problem to be solved, the main objective of the model one would create is to answer a specific research or operational question. There are many tools currently available to set-up and use hydrological models with applications ranging from small catchments to global models, from highly gauged catchments to ungauged ones. Each tool and consequently the instantiated model has its own unique characteristics; some physically-based (i.e., using the physics of underlying hydrological processes), some data driven (i.e., using correlations between input and outputs); and distributed in space and time or not.
The advantage of using water resources models is that they can support flood forecasting, proper water resource management, sediment transport analysis, water quality evaluation, estimating effects of climate change, etc. The disadvantage is that they require large amounts of data, which bring uncertainties in results; or need storing and managing; or data for validating a model is missing. In order to overcome these challenges, it is important to include rapid advances in remote sensing technologies, risk analysis, etc.
Having served as a member of several committees for masters programmes, how do such committees approach the design of a study programme? How can you develop an idea of what the academics of tomorrow will need to address the upcoming research challenges?
Nowadays, the volume of information that water graduates are expected to know is increasing far more rapidly than the ability of water-related curricula to “cover it.” Water graduates are increasingly finding employment in non-traditional water fields, such as computer engineering, health and safety engineering, and even business and finance. Therefore, the demands for masters’ programmes today are different than they were 10 or 15 years ago. To be effective across the broad spectrum of employment possibilities, the graduates should understand concepts in physics, mathematics, ecology, geography, computer and software engineering that are well beyond the range of the traditional water-related curriculum. At the same time, the work done by any one water professional tends to occupy a relatively narrow band. Structuring the curriculum that meets the needs of most students appears to be an increasingly elusive goal. The solution is to institute multiple tracks for different areas of specialization – as we have done at IHE Delft Institute for Water Education with our new MSc in Water and Sustainable Development, which offers students flexibility in designing their course of study.
In order to answer your second question, I would say that water problems usually cut across boundaries, and alliances that link professionals in many locations are becoming more common. The growing complexity and increased interdisciplinarity of water-related projects require a wide academic education and practical training. Apart from topics that are addressed, the design committees consider new methods of transferring knowledge, like group learning via collaborative engineering as well as online courses. These were not included in curricula in the past. The new approach will prepare graduates to adapt and discuss with everyone when searching for solutions.
What do you need to innovate?
It is very difficult to say what I need to innovate, as there is no methodology for it. I think, however, that one needs to have the chance to work with specific problems and try to solve them with the data available – while also considering relevant stakeholders. This will eventually lead not only to innovative solutions but also to practical ones.
Early in your career you had the chance to study and work in several universities. Is there any professor or researcher who especially influenced your career path, and if so, what inspired you particularly?
There is no one person in particular who influenced my career path; I am privileged to work with professors, researchers and students who are enthusiastic about their work. This has influenced me and my approach to research. I have been particularly inspired by the projects to which I have contributed, each one with its own challenges. I have also been inspired by my colleagues with whom I can explore ideas, and the students with whom I work who are curious about the topics they learn, and from whom I always learn something new about a particular problem or case study.
Would you like to share any advice to young professionals starting their career in the field of hydro-informatics?
First and foremost, understand the problem you need to solve. Then, adapt and always be open to new technologies and new data.
What courses would you recommend to decision makers working on sustainable water (resource) management and hydrology and aquatic ecosystem preservation topics?
I think life-long learning is important. I would advise decision makers in these fields to always be open to new technological advances. Moreover, it is very important to seek reliable information, to respect the opinions of stakeholders, and to always assess the consequences of possible decisions when using models. There is no one particular course one would need to follow, but a combination of courses. It is also important to stay up-to-date with the latest developments, and to always look for sustainable solutions.
What is your favourite aggregate state of water and why?
What an interesting and challenging question! I never thought of a water state as my favourite. If I think about it now, I guess the flowing liquid form. It is fascinating how it flows and no matter what we think, when we spill it, it will always surprise us and go on paths that one would not expect.