NI: Enhancing global hydrological models with local knowledge

Lead Research Organisation: University of Aberdeen
Department Name: Geography and Environment

Abstract

The assessment of water resources at both the global and regional scales is currently a major concern for scientists and policy makers due to the increasing incidence of water scarcity and extreme water-related hazards affecting the globe in the last 50 years. This concern will become even more relevant should future climate projections for the XXI Century become a reality in the coming years. The hydrological modelling community is aware that there is a need to look at the large-scale fluxes of water but also to consider the use of water at smaller scales. However, much of the work done until this moment on this regard has focused on achieving hyper-resolution of models, while the ability of these new modelling approaches to provide relevant and reliable information to catchment water management has received relatively less attention. This project initiates a collaboration between the large-scale modelling and the catchment modelling communities, aiming to fill a gap whose existence is often recognised but sparsely studied. The overall research objective of this project is to explore and design an approach for integrating Catchment Water Management Models (CWMM) with Large-Scale Hydrological Models (LHM), and to assess its potential and limitations to enhance the quality of information LHMs provide at a regional scale. This project will deliver an integrated programme of knowledge exchange - on selected the CWMM and LHM, AQUATOOL and CWatM respectively, and their capabilities - and research - developing a proof-of-concept to couple both models in the heavily managed, data-rich Ebro River Basin (Spain), in collaboration with three leading institutions on global and regional hydrological modelling. The main project output will be a proof-of-concept of an advanced CWatM model for the Ebro River basin that efficiently couples water management and other human interactions with the hydrological system. The new version of CWatM will allow for a better evaluation of human impacts on water availability, overcoming the current limitations LHMs to faithfully represent regional hydrology. This will translate into substantial progress not only for the hydrological science community, but for the climate and earth system science communities. The outcomes of this project will be the onset of further collaborative research to produce a generalised version of the proof-of-concept which will be applicable to understand and address pressing regional water issues, such as transboundary conflicts, which are traditionally hampered by data securitisation. The generalised version can also serve as a benchmark for the benefits of modelling-based water resources assessment and management in developing countries with low water governance, which may encourage investments in capacity building and data collation to conduct their own analyses, which this partnership can support.
 
Description This call focused on the development of new long-term partnerships with international scientists by means of a research project. Our project proposed to set up a new collaboration the University of Aberdeen and Cranfield University in the UK with the International Institute for Applied Systems Analysis in Austria, and Universitat Politecnica de Valencia and Estacion Experimental Aula Dei in Spain. The scientific objectives revolved around the evolution of large hydrological models (LHM) to better represent the human impacts on the water cycle at regional and local scales with the support of catchment-scale water management models (CWMM) so they can become effective tools in the management and decision making of regional water management. Our key research questions were:
1. What level of detail should be included in LHMs to make them suitable to inform decision making at the regional level?
2. What are the best coupling strategies? Can CWMMs be embedded in LHMs to improve their regional performance? How should the flow of information take place between CWMMs and LHMs to be effective? And
3. Do the generalised learnings from data-rich regions perform satisfactorily in data-scarce regions? What would be the minimum data required in the later to allow transferability?
To achieve this, the project explored the integration of catchment management models with large hydrological models, being the main outcome the development of a CWatM model of the Ebro River basin (Spain) including locally sourced data, including weather, hydrology, land use, and water management instead of data sourced from global databases.
Throughout the development of this model, we succeeded in answering our research questions in the following way:
1. We confirmed that current LHMs fail to appropriately represent a trustworthy tool for water management, even at the national scale, due to the large size of their cells (~50 km2). We also confirmed that hyper-resolution LHM trend to use 1 km2 grid cells is sufficient to capture catchment dynamics relevant to short-, medium- and long-term management of regional water resources. Moreover, the considered scale is consistent with the availability of data in most developed catchments.
2. While CWMMs can provide useful information to LHMs, their direct linkage can be cumbersome mostly due to technical differences regarding coding and data structure. However, indirect linking can provide valuable information to complement the data structures of LHMs. In this case, the use of a CWMM of the Ebro River basin allowed to refine reservoir release curves and to differentiate between releases to demands and releases to the environment, improving the overall performance of the model.
3. While there was not enough time to test our learnings in a data scarce region, we found that in highly managed systems, the availability of reservoir- related data (storage and releases) is probably one if not the most important piece of data to chase to obtain trustful results, with the identification of the largest water consuming regions (irrigation) coming next.
Exploitation Route The results of this project contribute to ongoing debates regarding hyper-resolution of hydrological models with a case study containing a large volume of data, sometimes beyond the scope of the debate. Hence, the developed model, which will be openly available, can be used by other researchers to test additional hypotheses on this regard.
In addition, the model we developed can be used to continue working on understanding the Ebro River basin. The project did not contemplate the run of climate or socio-economic scenarios, hence the model can still provide very valuable information for the management of the catchment, both from scientific and managerial perspectives. The model can also be used as a base for the development of monitoring and early warning tools for the Ebro River basin in conjunction with short- and medium-term weather forecasts in the management of droughts or seasonal water allocation.
Sectors Agriculture

Food and Drink

Environment
 
Title CWatM Ebro River Basin 
Description A distributed hydrological model of the Ebro River Basin developed with CWatM. The model includes the operation of all management-relevant reservoirs in the Ebro River basin as well as a detailed account of irrigated areas and how they are supplied. 
Type Of Material Computer model/algorithm 
Year Produced 2024 
Provided To Others? Yes  
Impact This model has allowed us to advance our understanding in the implementation of hyper-resolution large hydrological models, especially with regard to heavily managed systems like the Ebro. We have been able to assess the minimal amount and quality of additional data these kind of models require to return reliable output that can be useful for water resources planning and decision making. 
 
Description EEAD Ebro 
Organisation Estación Experimental de Aula Dei
Country Spain 
Sector Public 
PI Contribution We are developing AQUATOOL and CWatM models of the Ebro basin.
Collaborator Contribution The EEAD-CSIC contribution to this project will consist of: 1. Support to Dr. Haro-Monteagudo and Dr. Momblanch regarding the modelling tasks of the Ebro River basin, through data provision and assessment of model outcomes. 2. Liaison with the Ebro River agency and key stakeholders to organise a 1-day workshop in Zaragoza around month 16 of the project, including the provision of meeting spaces. 3. My participation in project meetings with all other partners. 4. Preparation of communications for scientific meetings and co-writing of research articles.
Impact Submission of abstract to EGU22 conference
Start Year 2021
 
Description IIASA CWatM collaboration 
Organisation International Institute for Applied Systems Analysis
Country Austria 
Sector Academic/University 
PI Contribution We are developing a hydrological model of the Ebro River basin (Spain) with the CWatM model, developed by IIASA. We have collated all the necessary data to build the model and we will implement it later in the project.
Collaborator Contribution 1. Co-guidance of Dr Haro-Monteagudo and Dr Momblanch with regard to the CWatM modelling tasks. 2. Participation of Dr. Kahil and Dr. Burek in project meetings with all other partners. 3. Preparation of communications for scientific meetings and co-writing of research articles. So far, IIASA has provided support to initiate modelling work with CWatM model for the Ebro River basin. They have contributed to drafting a conference abstract and presentation. They will continue to provide support and guidance in the project's modelling efforts.
Impact Abstract submission to EGU22 conference
Start Year 2021