FORWARD

FORWARD - Operational monitoring and FOrecasting system for Resilience of agriculture and forestry under intensification of the WAteR cycle: a big Data approach

Project Website

Coordinator
Alberto Fernández Villán

Executive Coordinator

Projects  Partner and Institution:
TSK (Spain)
Technical University of Denmark (Denmark)

Sumaqua (Belgium)

Key words: Resilience, Ecohydrological, Big Data, Process-based Model, Data-driven Model, Machine Learning, Agriculture, Forestry.

Water stress becomes increasingly problematic due to various changes, including the increasing urbanization, land use changes, population growth and the increasing demand for (fresh) water. On the other hand, climate change induces more variability and extremes in rainfall and temperature. This intensification of the water cycle will highly affect agriculture and forestry, giving rise to the urgent need to assess the resilience of agricultural and forest ecosystems under varying hydro-climatic conditions and more specifically in water limited regions. The objectives of the project have been mainly three.

• First, to understand the resilience of agricultural and forest ecosystems to climatic extremes.
• Second, to develop improved forecasting models of ecohydrological variables.
• Third, to deploy the other objectives in a Big Data Framework capable to collect data from various sources, analyse them and make predictions through data-driven, process-based models and machine learning techniques combination.

FORWARD had tackled these challenges by combining Big Data, data mining and advanced analytics, with different model types (both data-driven and process-based) and assimilation techniques. The entire course of the project has been characterized by an intensive multidisciplinary approach, as it was necessary for each member to address some fields of knowledge quite different from its own. Collaboration was then crucial to understand the work processes performed by other members with mainly scientific and academic profile, on one side, and industrial technological and software development profile, on the other side.

The most remarkable result of the project has been the development and implementation of an extensible and tailorable Big Data framework able to manage and process multivariable information sources, including large volumes of geospatial datasets, data-mining techniques and models at several scales. Such powerful functionalities have been built on top of a set of data-driven algorithms and process-based models using Earth Observations (EO). These models enable for prediction capabilities, as well as real time monitoring and forecasting. The enhanced forecasting system, expected knowledge gain into concepts of resilience of forestry and agriculture, and the modular Big Data framework goes beyond the state-of-the-art in ecohydrological sciences, and exploits all available sources of information and techniques. 

The consortium managed to collect the stakeholders interests and most urgent needs in order to address them in the project. This way, the project succeeded to engage with very diverse stakeholders, such as international organisms, grassroots non-governmental organizations, private companies and Ministries from national government and several researchers/universities from all around the world. Their feedback was taken into account in order to decide the most relevant indicators to be produced as project outputs.

The project outcomes have a considerable potential to produce impacts from the scientific and societal point of view through different mechanisms, especially on research, industry, end users and policy. The algorithms developed for data analysis of variables relevant for agriculture and forestry will be important for monitoring and warning, data mining, and gaining insight into vegetation anomalies. Moreover, with Big Data analytics, problems can be detected faster and at a greater scale with more in-depth statistics, which helps end-users to make reliable and better-informed decisions. There is impact for industry applications as well, considering the insurance industry stands to benefit from the generated data mining techniques for anomaly detection, for instance. Another important remark of FORWARD project was the engagement in public plans using open software (GEE), which was developed by DHI Gras Company in close collaboration with some members of the consortium. This way, the scientific methods and model outputs has been made readily available to different stakeholders (scientific, companies, government agencies, international agencies) by making the code available in google Earth Engine (GEE). 

Project structure: 

Implementation: The FORWARD project is divided into 5 WPs:
WP1. Big Data framework: data management and processing.
This WP will focus on data gathering, pre-processing and integration. Novel integration approaches based on Big Data technologies and not used by the traditional scientific community will be created according to the volume, velocity and variability of the sources of information.
WP2. Data-mining and data-driven modelling.
Data mining and data-driven modelling techniques will be explored to quantify variability, extremes and resilience in space and time.
WP3. Historical Process-based models using EO data.
Historical process-based modelling using EO time series. Satellite and other spatial climatic databases compiled in WP1 will be used as input into process-based model to estimate evapotranspiration and water use efficiency (WUE).
WP4. Integration and analysis of results.
A final integrated analysis will be performed studying and validating the sensitives across regions, regimes, and vegetation types. After completing all WPs, an integrated framework will emerge indicating how extreme and variable climate conditions affect the eco-hydrological behaviour of ecosystems, highlighting spatial differences across regions and response changes due to local factors.
WP5. Communication and dissemination.
To ensure that the objectives and approach of FORWARD address the stakeholders’ needs, to maximize dissemination of the outcomes of FORWARD and to create the potential of making the tools of FORWARD industry standard in resilience management and Big Data.

Outcome/deliverables:
M1.1: Data sources defined and integrated (month 5).
M1.2: Indicators defined (month 12).
M1.3: Architecture deployed (month 12).
M2.1: All tools/techniques/data inventorized/collected (month 2)
M2.2: Semi-finalized set of data mining techniques developed (month 8)
M2.3: Preliminary set of data-driven modelling techniques implemented (month 18).
M4.1: Implementation of WP2 and WP3 techniques in the Big Data framework (month 18).
M4.2: Assembled data and configured framework for scenario simulations (month 20).
M5.1: Defined the questionnaire and identified list of contacts (month 3).
M5.2: Published at least 5 international papers and attended 5 conferences (month 24).
M5.3: Secured at least 1 additional valorisation opportunity in which all partners can participate (month 24).

Deliverables:
D1.1: Data sources and indicators finalized (month 12).
D1.2: Finalized architecture deployment (month 24).
D2.1: Generalized data mining techniques and anomaly algorithm & indicator (month 12)
D2.2: Maps indicating hotspots, vulnerability and identified “hot time periods” (month 14)
D2.3 Generalized data-driven modelling techniques (month 20)
D4.1: Derived climate change scenarios for drivers/predictors at relevant scales (month 18).
D4.2: Deployed Big Data framework (month 20).
D4.3: Simulated defined seasonal and long term scenarios and analysed the impacts (month 22).
D4.4: Derived key indicators (month 24).
D5.1: Set-up website (month 1).
D5.2: Extracted list of most urgent needs and collected data from stakeholders (month 6).

References coordinator and  leaders of  each WP:
WP1. Leader: Alberto Fernández Villán.
WP2. Leader: Vincent Wolfs.
WP3. Leader: Mónica García.
WP4. Leader: Alberto Fernández Villán.
WP5. Leader: Mónica García.

Main outputs:

• Moyano, M.C.; Garcia, M.; Palacios-Orueta, A.; Tornos, L.; Fisher, J.B.; Fernández, N.; Recuero, L.; Juana, L. 2018. Vegetation Water Use Based on a Thermal and Optical Remote Sensing Model in the Mediterranean Region of Doñana. Remote Sensing. 2018, 10, 1105
• Li Y., Kustas WP., Huang C., Nieto H., Haghighi E., Anderson M., Domingo F., Garcia M., Russell S. Evaluating Soil Resistance Formulations in Thermal‐Based Two‐Source Energy Balance (TSEB) Model: Implications for Heterogeneous Semiarid and Arid Regions Water Resources Research 55 (2), 1059-1078.
• Wang, S, Ibrom, A, Bauer-Gottwein, P & Garcia, M. 2018. Incorporating diffuse radiation into a light use efficiency and evapotranspiration model: an 11-year study in a high latitude deciduous forest. Agricultural and Forest Meteorology, 248, 479-493.
• Vicente-Serrano, S.M., Azorin-Molina, C., Peña-Gallardo, M., Tomas-Burguera, M., Domínguez-Castro, F., Martín Hernández, N., Beguería, S., El Kenawy, A., Noguera, I., García, M., 2019. A high-resolution spatial assessment of the impacts of drought variability on vegetation activity in Spain from 1981 to 2015. Nat. Hazards Earth Syst.

More results on the project: Data and resources

Contact Point for  Communication/Dissemination activities: Mónica García.

Contact Point for Open Data/Open Access activities: Alberto Fernández Villán.