Space-based Information Support for Prevention and Recovery of Forest Fire Emergency in the Mediterranean Area

G. Laneve (1), A. Sebastián López (2), M. L. Tampellini (3), B. Hirn (4), F. Ferrucci (4), A. Kostaridis (5), G. Leventakis (6), S. Clandillon (7), A. Bento Gonçalves (8)


(1) DIAEE Università La Sapienza. – Italy
(2) GMV Aerospace and Defence S.A.U, – Spain
(3) CGS S.p.A. Compagnia Generale per lo Spazio – Italy
(4) IES Consulting SrL – Italy
(5) SATWAYS – Greece
(6) KEMEA Center for Security Studies – Greece
(7) Universitè de Strasbourg – France
(8) University of Coimbra – Portugal

Overview 

    Fig. 1. In gray, the countries where the project partners are located, and in red, the project test areas. Image Credit: Authors.

Fig. 1. In gray, the countries where the project partners are located, and in red, the project test areas. Image Credit: Authors.



The Prevention and Recovery of Forest Fire Emergency in the Mediterranean Area (PREFER) project is funded under the EU FP7 (G.A. 312931), to be developed during 2013-2015. PREFER  aims at responding to major fire prevention needs in Southern Europe by extensive exploitation of space-borne sensors. All reports on the state of Europe’s forests indicate that the Mediterranean area is systematically affected by uncontrolled forest fires with large impacts on ecosystems, soil erosion, slope instability, desertification, and local economies as a whole, with a negative mid-to-long term prospect because of climate change. In this scenario, the need to improve forest fire prevention is widely recognized. Fire prevention is still the most cost-effective strategy when compared to firefighting. PREFER will contribute by: 1) providing timely, multi-scale and multi-payload information products based on all available space-borne sensors within the next four years; 2) offering a portfolio of Earth observation products depicting the forest fire emergency cycle in the EU Mediterranean area; 3) preparing the exploitation of new space-borne sensors available by 2020 (e.g., Sentinels) and 4) contributing to the definition of user requirements for new Earth observation missions. PREFER will process and distribute information to end users, developed and maintained by a regional research and development cluster of core users, industries and research institutes. This regional service is intended to stimulate coordination among countries on forest fires prevention in the EU Mediterranean region.

Project Objectives

The main objective is to set up space-based, end-to-end information services to support prevention/preparedness for and recovery from forest fires  in the Mediterranean region. The PREFER portfolio will consist of two main services:

1. Information Support for Fire Preparedness/Prevention, and
2. Information Support for Fire Recovery/Reconstruction

These information services will include a harmonized set of user’s requirements, defined by users in Portugal, Spain, Italy, France and Greece and taking into account the different legal frameworks existing in such countries. They will be as general as possible to be usable throughout the Mediterranean region.

Additionally, they will:

• Demonstrate an interoperable service provision infrastructure based on the Open Geospatial Consortium’s INSPIRE that will allow easy access.
• Complement products provided by Copernicus, the European Earth observation program and the European Forest Fire Information System.
• Integrate Earth observation, digital terrain model, socio-economic, in-situ, and meteorological data.

The PREFER consortium has the ambitious objective to start up the formation of a cluster of research institutes, industries and small and medium enterprises providing space-based information services and products for forest fire emergency management in the Mediterranean region.

PREFER Area of Interest and End Users

PREFER focuses on the Mediterranean region (Fig. 1). This area was chosen because the climate of Southern Europe and the Mediterranean basin is projected to warm at a rate exceeding the global average, exacerbating wild fires.

The following users manage forest fires emergencies in their respective regions and have signed a letter of commitment to support and collaborate with the PREFER consortium:

• Portugal: Portugal National Authority for Civil Protection
• Spain: Regional Environment Government of Andalusia, Environment and Water Agency
• France: Office National des Forêts, Direction Générale Corse
• Italy: Corpo Forestale e di Vigilanza Ambientale – Regione Autonoma della Sardegna
• Greece: Prefecture of Peloponnesus (including associated civil protection resources)

PREFER can help harmonize the management of forest fires at a transnational level by providing regional authorities with the same portfolio of basic products. Some PREFER products (e.g., burned areas, daily hazard map) will be also generated at European level,  but  a feature higher spatial resolution that matches most users’ requirements at the national level.

Information Support for Preparedness/Prevention

    Fig. 2. Practice of prescribed burning in Spain. Image Credit: Authors.

Fig. 2. Practice of prescribed burning in Spain. Image Credit: Authors.



The list of products provided by this service will include seasonal fuel maps, daily and seasonal fire hazard maps, seasonal risk maps and prescribed fire maps.

In the past, fuel mapping using Earth observation data has been accomplished at a variety of temporal and spatial scales. PREFER will take a dynamic approach to fuel mapping based on continuous monitoring of ever-changing vegetation during the fire campaign. We will use of a visible/infrared imager radiometer suite sensor for downscaling from local to regional purposes.

As for fire risk and hazard assessment, many indexes have been developed in recent years. Hardly any of them uses data derived from satellite images. The Fire Probability Index (FPI) (Burgan et al. 1998, Sebastian-Lopez et al. 2002) is an exception to this rule and, in addition, makes use of meteorological data. In spite of being a very complete index, the FPI still allows room for improvement. PREFER’s innovative approach to FPI will allow taking into account the effect of solar illumination conditions in determining the humidity present in dead vegetation and, therefore, its susceptibility to burning. PREFER will also adapt the index to take into account water present in living vegetation  by measuring equivalent water thickness (Ceccato et al. 2002, Laneve et al. 2011, Huesca et al. 2009). PREFER will use daily FPI forecasts to produce a seasonal fire hazard index by including in the model the human factor as captured by the fire occurrence statistics.

PREFER will produce seasonal fire risk maps by combining seasonal hazard data with vulnerability and exposure maps.

Finally, prescribed burning (PB) (Fig. 2) represents the controlled application of fire to vegetation under specific environmental conditions to attain planned resource management objectives (FAO, 2007). The main objective of PREFER for PB is to support the user in the identification of specific areas and times where and when PB is applicable and safe. Taking into account the results of other European projects (such as FIREPARADOX,), PREFER can  integrate  advanced remote sensing techniques not yet been developed to identify the right time to intervene with a PB in the area of interest.

RapidEye_GA_

Fig. 3. Example of very high spatial resolution satellite image (RapidEye) used for estimating a burned area. Acquired through the PREFER project’s service project and dataset management.



Information Support for Recovery/Restoration

The list of PREFER products will include post-fire vegetation recovery maps; burn scar maps at high spatial resolution (Fig. 2) based on optical and SAR images; biomass burning aerosol maps; 3D fire damage assessment maps; and damage severity maps.

The automatic contextual method (Hirn et al. 2005) applied (based on optical and SAR images) by PREFER for assessing the burned areas at a scale ranging from 1/25,000-1/2,000 is able to provide burned area contour at cadastral scale and can be used for defining the effective vegetation re-growth and biomass increase. Of course, the vegetation recovery analysis is restricted to areas previously damaged by fire.

Close monitoring and quantitative assessment of vegetation re-growth after depletion by fire are essential for assessing (i) the new availability of fuel accompanying the restoration of biomass, and the resulting changes in fire risk; and (ii) the reduction of slope instability risk associated with re-growth of significant vegetation in erosion-prone areas. Systematic, quantitative analysis of vegetation re-growth will be an important part of post-fire monitoring of restored wildlife habitat.

The objectives of the 3D fire damage assessment maps are: (1) to provide 2D and 3D information about the damage caused by forest fires to highlight areas prone to soil erosion and increased risk of flash flooding and debris flows, (2) to include volumetric indicators about the burnt areas and (3) to document burnt area evolution over time.

The severity damage map will provide information on the status of post-fire vegetation (De Santis et al, 2009) for planning the recovery phase. The PREFER team will develop a damage severity index . Using post-event, hyper-spectral images, the index will estimate the impact on vegetation by distinguishing different levels of damage, namely the loss of biomass products. We expect the product to be composed of two layers, both extracted from post-fire images. The first layer will assess immediate conditions, while the second will evaluate the long-term damage. Fully automatic, the damage severity index  will cover the entire European continent.

Conclusions

Notwithstanding the improvements in fire fighting,  the fires themselves are not showing any tendency to decrease. Therefore, the European Commission has recently adopted  an approach to natural and man-made disasters that emphasizes common efforts over separate national ones: for example, developing knowledge, linking actors and policies, and efficient targeting of community funds to prevention like done in the Firesmart project, for example. Prevention is still the most cost-effective strategy when compared to fire fighting and suppression, able to efficiently mitigate this major environment threat. PREFER intends to contribute to such a pragmatic need of southern Europe’s forests by providing timely information products based on the exploitation of all available spacecraft sensors and offering a portfolio of products focused on pre- and post-crisis forest fire emergency suitable for the users throughout the European Mediterranean region.

Therefore, PREFER will set up a regional service to process and distribute this information to end users. The service will be a new, powerful tool for forest fire management.

References

Burgan, R.E.; Klaver R.W.  and Klaver, J.M. (1998) “Fuel models and fire potential from satellite and surface observations,” Int. J. Wildland Fire 8 (3), 159-170.

Ceccato, P., Gobron, N., Flasse, F., Pinty, B., & Tarantola, S. (2002). Designing a spectral index to estimate vegetation water content from remote sensing data: Part 1, Theoretical approach. Remote Sensing of Environment, 82, 188-197.

De Santis, A., & Chuvieco, E. (2009), “GeoCBI: a modified version of the Composite Burn Index for the initial assessment of the short-term burn severity from remotely sensed data,” Remote Sensing of Environment, 113 (3), 554-562.

FAO 2007: Fire management – global assessment 2006. FAO Forestry Paper 151, Rome. 156 p.

Hirn B., Ferrucci F., (2005), “MYME2: a multi-payload integrated procedure for the automated, high-resolution remote sensing of Burn Scars,” Proceedings IEEE (IGARSS 2005), Seoul.

Huesca, M.; Litago, J.; Palacios-Orueta, A.; Montes, F.; Sebastián-López, A.; Escribano, P. (2009), “Assessment of forest fire seasonality using MODIS fire potential: A time series approach,” Agricultural and Forest Meteorology 149, 1946–1955.

Laneve, G.; Jahjah, M.; Ferrucci F.; Battazza, F.  (2011), “The Development of a Fire Vulnerability Index for the Mediterranean Region,” IGARSS 2011 Proceedings, pp. 4146 – 4149.

Sebastián-López, A.; San-Miguel-Ayanz, J.; Burgan, R.E. (2002), “Integration of satellite sensor data, fuel type maps and meteorological observations for evaluation of forest fire risk at the pan-European scale,” Int. J. Remote Sensing, 23 (13), 2713-2719.