Integration of Remote Sensing Data to Facilitate Multi-Hazards Risk

Assessments in Coastal Regions

Eduardo R. Oliveira

, Leonardo Disperati

and Fátima L. Alves

COPING TEAM – Coastal and Ocean Planning Governance, CESAM - Centre for Environmental and Marine Studies,

Department of Environment and Planning, University of Aveiro, Aveiro, Portugal

Dipartimento di Scienze Fisiche, della Terra e dell’Ambiente, Università degli Studi di Siena, Siena, Italy

1 RESEARCH PROBLEM

Risk management is amongst the priorities of

Portugal’s national strategic policies, including

regional and national spatial planning programmes

(PROT and PNPOT). According to Europe 2020

Strategy, there is a global need to strengthen

economies towards climate risks, disaster prevention

and response.

Following the EU Internal Security Strategy target of

establish a risk management policy linking threat and

risk assessments to decision making in each member

state, the Sectorial Plan for Risk Prevention and

Reduction has been developed to include risk

assessment and cartography (ANPC, 2014).

However, several difficulties have been recognized

by the Portuguese Directorate-General for the

Territory (DGT) and the National Civil Protection

Authority (ANPC), regarding complexity and lack of

consensus from multiple scientific domains (DGT,

2013).

Multi-Hazard Risk Assessments (MHRA) deal with

the combination of multiple hazardous sources and

multiple variable elements overlapping in time and

space. Such approaches can go beyond the simple

aggregation of single-hazard assessments, by

considering several types of interactions, which may

amplify their consequences (Delmonaco, Margottini

and Spizzichino, 2007; Marzocchi et al., 2012).

MHRA are solutions capable of supporting spatial

planning decisions and emergency strategies. One of

the most important considerations concerns the

availability of ready and reliable data, which if often

expensive or unavailable to risk assessment experts.

Earth Observation (EO) using satellite Remote

Sensing (RS) have been widely used to provide data

for single hazard risk assessments, as complement to

ground-based data. However, the main research

question arises: ‘Can multi-hazard risk assessments

be based on freely available RS data?’

2 OUTLINE OF OBJECTIVES

This thesis aims at providing practical methodologies

and tools to improve and facilitate multi-hazard

assessments in populated and natural territories. The

main objective is to develop an innovative

methodology for multi-hazard risk assessments, using

satellite multi-spectral RS images and digital

elevation data as the main data source. The

methodology will be applied to the Aveiro region (in

the Norwest of Central Portugal), using alternative

data sources for validation purposes.

To achieve this objective, the following specific goals

should be accomplished:

- to systematize detailed geospatial information about

coastal and riverine floods, wave overtopping,

wildfires and soil erosion hazards in the Aveiro

region, based on the available literature and

databases;

- to study the effects and interactions between land

cover types, elevation, temperatures, humidity and

water levels with hazard events;

- to develop a methodological approach for multi-

harzard risk assessments, based on satellite RS data;

- to test, apply and validate the methodology in the

Aveiro Region;

- to discuss the potential uses of this new

methodology, including applications with different

hazard types, semi-automatic routines for primary

risk assessments, and application on large/remote

areas.

R. Oliveira, E., Disperati, L. and L. Alves, F.

Integration of Remote Sensing Data to Facilitate Multi-Hazards Risk Assessments in Coastal Regions.

In Doctoral Consortium (GISTAM 2018), pages 3-7

3 STATE OF THE ART

The coexistence of modern industrial societies

together with fragile natural territories increases the

vulnerabilities and exposure to both technological

and natural risks, placing new challenges for risk-

management at local and regional scales (de Souza

Porto and de Freitas, 2003). The effects of global

climate change are contributing to increase the

frequency and intensity of weather related hazards

(Adger et al, 2007; Deleu, Tambuyzer and Stephenne,

2011), requiring decision support information tools in

order to establish effective disaster mitigation

strategies (Grünthal et al., 2006).

Traditional Single-Hazard Risk Assessment (SHRA)

approaches include vulnerability and exposure

analysis of the affected elements by one hazard (e.g.

buildings, people, cars, land uses, infrastructures).

Nonetheless, most natural and anthropogenic risks are

likely to occur at one same location and, not rarely, at

the same time (Carpignano et al, 2009). Experiences

with decision makers show that a territorial

perspective is desirable for spatial planning decisions

and emergency strategies (Grünthal et al., 2006), in a

way that Multi-Hazard Risk Assessment (MHRA)

combines multiple hazardous sources and multiple

vulnerable elements overlapping in time and space,

which may be as close as possible to the reality of

spatial management for decision-makers (Carpignano

et al., 2009). Beyond the territorial perspective,

MHRA can be element oriented, concentrating on the

potential impacts from various events in the same

element at risk (Delmonaco, Margottini and

Spizzichino, 2007). Another highlighted aspect is the

interaction among different risks (Marzocchi et al.,

2009; Selva, 2013), and the so-called “cascade

effects”, which are often neglected in SHRAs

(Marzocchi et al., 2012).

Mapping is usually amongst the first steps to take

preventive measures, allowing decision makers to

identify the spatial distribution of hazard intensities,

exposed population and values, as well as expected

losses. Complete MHRAs enable significance

comparison of different hazard types, contributing to

raise awareness and develop tailor-made mitigation

strategies (Carpignano et al., 2009; EC, 2010).

Spatial and statistical data have different relevancies

for each hazard type. Nevertheless, data requirements

for natural hazard assessments include land use,

vegetation, slope, oceanographic and meteorological

factors (Van Westen, 2013). Remote Sensing (RS)

has provided a synoptic perspective for many of these

measurements, for variable spatial scales and

temporal resolutions, contributing for a wide range of

disciplines (Tralli et al., 2005). Satellite earth

observations have been used in many SHRAs,

enabling the possibility to reconstruct recent-history

catastrophic events and providing data for prediction

and mitigation planning actions (e.g., Lu et al., 2004;

Grünthal et al., 2006; Chuvieco et al., 2010; Leifer et

al., 2012).

4 METHODOLOGY

This thesis is being developed within a research group

which as participated in several projects concerning

the Aveiro region and its relation to hazards and

global change scenarios (e.g. ADAPTARia (FCT),

LAGOONS (FP7), SPRES (EU-INTERREG IV),

ClimAdaPT.Local (MFEEE/EEA-Grants)). This

experience is considered relevant to the thesis

development, providing insights and relevant inputs

from these projects databases.

Given the complex interactions between different

hazard types and the innovative character of this

study, the list of selected hazards was restricted to:

floods (river and coastal), wildfires and soil erosion.

Not only are they amongst the most significant and

studied hazards affecting this study area, but they are

also in terms of RS, directly related with two of the

most studied and easily-identifiable land cover

elements - water and vegetation.

To minimize the costs of this thesis (which will also

affect its potential applications), the methodological

development and application will be based on free

available data and freeware/open-source software. In

RS, this represents a significant constraint in terms of

available resolutions – temporal, spectral, but mostly

spatial. However, they should be suitable for regional

assessments, allowing the compliance with national

and European strategies.

This innovative methodology will provide solutions

to reduce efforts, costs and time of traditional field

monitoring and campaigns and surveys for MHRA.

By delivering a simplified methodology based on

freely available resources and easily accessible to risk

managers and the public, socioeconomic benefits

should be generated, promoting risk awareness and

contributing for increasing the resilience of

populations and ecosystems.

DCGISTAM 2018 - Doctoral Consortium on Geographical Information Systems Theory, Applications and Management

4.1 Literature Review

The initial steps of this PhD consist in the elaboration

of an exhaustive literature review about the main

topics relevant to this thesis. Several methods

regarding single and multi-hazard risk assessments

were classified in terms of the possibility of

integration of RS methods and applicability to the

study area. This work has been summarized and

submitted as a review paper in a relevant peer-review

scientific journal.

4.2 Single Hazard Risk Assessments

This sub-section includes the collection of hazard or

risk assessment studies comprising the study area, as

well as other literature with similar approaches to this

working programme.

Relevant geospatial data regarding floods, wildfires

and soil erosion, have been collected to support both

single and multi-hazard risk assessments.

Historical hazard-related events occurring in the

study area will be reconstructed and characterized,

including present conditions and global change

scenarios. The available databases of previous

research projects are important information sources

(e.g. geomorphologic variables; meteorological data;

hazard occurrences, consequences and responses;

socioeconomic variables), which can be

complemented with press-archives and other studies.

The characterization will continue with data

acquisition from satellite imagery databases, for those

periods around hazard occurrences (prior, after and if

possible during), to determine possible correlations

RS data and the occurrence of hazard events, which

will be crossed with

Land cover changes will be analysed using freely

available RS multispectral images, from medium

spatial resolution satellites (e.g. Landsat, Sentinel,

ASTER). MODIS (TERRA and AQUA) will be used

whenever higher temporal resolutions are required,

providing information about temperatures, humidity

and water levels.

SRTM, ASTER DEM and ALOS data will be used to

obtain elevation data.

By the end of this step, single risk assessments will

be established for each of the selected hazards, using

either data-driven or physically based methods.

Each historical event occurring since the late

seventies (which corresponds to the oldest records of

Landsar series will be carefully analysed in order to

identify usable satellite images. Potential interactions

between multiple hazards will be analysed, focusing

on simultaneous or cumulative occurrences. These

results will be published in relevant peer-review

scientific journals, together with further

methodological developments.

4.3 Development of an Innovative

Methodology for MHRA based on

Satellite RS Data

The methodology to be developed in the context of

this PhD is based on the co-relation of changes

detected in previous hazard-related events,

privileging RS data to provide fast and reliable risk

assessment solutions.

4.3.1 Exposure Assessment

The first step will be dedicated to identify methods

for determining levels of multi-hazard exposures.

Each event will be analysed to identify predisposing

conditions for single hazards’ direct effects, and in

terms of triggering effects which may generate

secondary hazards (e.g. floods occurring in recently

burned areas, or floods following wave overtopping

events) (Van Westen, 2013).

4.3.2 Vulnerability Assessment

Subsequently, a global vulnerability index will be

developed for covering individual aspects for each

specific hazard, including expected effects on

physical, biological and socioeconomic dimensions.

Source data will be directly (or indirectly) obtained

through RS methods (e.g. geomorpholical features,

land covers, buildings).

4.3.3 Multi-Hazard Risk Assessment

By the end of this step, an innovative risk index will

be proposed to relate multi-hazard exposures and

vulnerabilities, using qualitative relations (matrixes)

or algebraic operations between different indexes.

Another scientific publication will resume these

developments.

4.4 Methodological Application to the

Case Study Area and Validation

The Aveiro region will be used to test and validate

single and multi-hazard risk assessments. The

application of RS based methods will be compared to

others obtained from alternative data sources in order

to validate results. These can be found in several

projects related with risk assessments in the Aveiro

Integration of Remote Sensing Data to Facilitate Multi-Hazards Risk Assessments in Coastal Regions

Region (e.g. ADAPTARia, Lagoons, MISRaR, Plano

de Ordenamento da Orla Costeira de Ovar-Marinha-

Grande, Secur-Ria, PESERA).

4.5 Exploring the Relationship between

Semi-Automatic Remote Sensing

Data and Multi-Risk Analysis

All results and conclusions obtained in the previous

stages of the doctorate will be compiled in the

dissertation and a final scientific publication. The real

uses of the developed methodologies will be

discussed, allowing the compliance with national and

European sectorial plans and strategic goals.

The topics for discussion include the application in

large or remote areas, accuracy levels, and

incorporation of semi-automatic routines.

5 EXPECTED OUTCOME

The main deliverable of this thesis will be an

innovative methodology for multi-hazard risk

assessments based on RS methods, which remain

underexploited. Its highest potential concerns

regional scale applications, in line with the

requirements of national and European strategies and

sectorial plans for risk prevention and management.

The outputs of this work will provide spatial

managers and decision-makers with an integrated

reliable and facilitative approach to support private

and public sectors in increasing territorial resilience

to risk, and to generate socioeconomic benefits from

sustainable use of resources.

6 STAGE OF THE RESEARCH

At the time of this writing, the thesis is at is half-way

point. A paper containing the literature review has

already been submitted to a scientific journal.

The PhD candidate has collaborated in a publication

about a RS based method for modelling shoreline

evolution to support coastal risk management,

including an application to the study area (Cenci et

al., 2015, 2017).

Another paper is being concluded, consisting in a RS

method for river flood detection (again with

application to the study area).

The following step should address the remaining

single hazard assessments (wildfires and soil erosion)

and the application of an overall vulnerability index.

The final task will consist in the definition and

application of a multi-hazard risk assessment, which

will consider all the previous steps of this thesis,

including every single-hazard risk assessment. The

validation of results is being made together with each

individual hazard assessment.

REFERENCES

Adger, W. et al. (2007) ‘Assessment of adaptation

practices, options, constrains and capacity’, in Climate

Change 2007: Impacts, Adaptation and Vulnerability.

Contribution of Working Group II to the Fourth

Assessment Report of the Intergovernamental Panel on

Climate Change, pp. 717–743.

ANPC (2014) ‘Avaliação nacional de risco’, Anpc, p. 221.

Carpignano, A. et al. (2009) ‘A methodological approach

for the definition of multirisk maps at regional level:

first application’, Journal of Risk Research, 12(3–4),

pp. 513–534. doi: 10.1080/13669870903050269.

Cenci, L. et al. (2015) ‘Remote sensing for coastal risk

reduction purposes: Optical and microwave data fusion

for shoreline evolution monitoring and modelling’,

2015 IEEE International Geoscience and Remote

Sensing Symposium (IGARSS), pp. 1417–1420. doi:

10.1109/IGARSS.2015.7326043.

Cenci, L. et al. (2017) ‘Integrating remote sensing and GIS

techniques for monitoring and modeling shoreline

evolution to support coastal risk management’,

GIScience & Remote Sensing. Taylor & Francis, 0(0),

pp. 1–21. doi: 10.1080/15481603.2017.1376370.

Chuvieco, E. et al. (2010) ‘Development of a framework

for fire risk assessment using remote sensing and

geographic information system technologies’,

Ecological Modelling, 221(1), pp. 46–58. doi:

10.1016/j.ecolmodel.2008.11.017.

Deleu, J., Tambuyzer, H. and Stephenne, N. (2011) ‘Multi

Risk preparedness supported by Earth Observation’,

Eurosense Gi4DM Conference, 3-8 May, Antalaya,

Turkey.

Delmonaco, G., Margottini, C. and Spizzichino, D. (2007)

‘ARMONIA methodology for multi-risk assessment

and the harmonisation of different natural risk map’,

ARMONIA PROJECT: Applied Multi-Risk Mapping of

Natural Hazards for Impact assessment, Deliverable

3.1.1, Rome, Italy.

DGT (2013) Linha de Investigação em Riscos. Available at:

http://www.dgterritorio.pt/

ordenamento_e_cidades/ordenamento_do_territorio/pr

evencao_e_reducao_de_riscos/linha_de_investigacao_

em_riscos/ (Accessed: 11 January 2018).

EC (2010) ‘Risk Assessment and Mapping Guidelines for

Disaster Management’, European Comission staff

working paper, 21 December, Brussels, Belgium.

Grünthal, G. et al. (2006) ‘Comparative Risk Assessments

for the City of Cologne - Storms, Floods, Earthquakes’,

Natural Hazards, 38, pp. 21–44. doi: 10.1007/s11069-

005-8598-0.

DCGISTAM 2018 - Doctoral Consortium on Geographical Information Systems Theory, Applications and Management

Leifer, I. et al. (2012) ‘State of the art satellite and airborne

marine oil spill remote sensing: Application to the BP

Deepwater Horizon oil spill’, Remote Sensing of

Environment, 124, pp. 185–209. doi:

10.1016/j.rse.2012.03.024.

Lu, D. et al. (2004) ‘Mapping soil erosion risk in Rondônia,

Brazial Amazonia: Using RUSLE, Remote Sensing and

GIS’, 512, pp. 499–512.

Marzocchi, M. et al. (2009) ‘Principles of multi-risk

assessment - Interation amongst natual and man-

induced risks’, Project Report, EUROPEAN

COMMISSION, Directorate-General for Research,

Directorate I — Environment, Unit I.5—Climate

Change and Environmental Risks, (Research*eu,

Brussels, Belgium).

Marzocchi, W. et al. (2012) ‘Basic principles of multi-risk

assessment: a case study in Italy’, Natural Hazards,

62(2), pp. 551–573. doi: 10.1007/s11069-012-0092-x.

Selva, J. (2013) ‘Long-term multi-risk assessment:

statistical treatment of interaction among risks’,

Natural Hazards, 67(2), pp. 701–722. doi:

10.1007/s11069-013-0599-9.

de Souza Porto, M. F. and de Freitas, C. M. (2003)

‘Vulnerability and industrial hazards in industrializing

countries: an integrative approach’, Futures, 35(7), pp.

717–736. doi: 10.1016/s0016-3287(03)00024-7.

Tralli, D. M. et al. (2005) ‘Satellite remote sensing of

earthquake, volcano, flood, landslide and coastal

inundation hazards’, ISPRS Journal of

Photogrammetry and Remote Sensing, 59(4), pp. 185–

198. doi: 10.1016/j.isprsjprs.2005.02.002.

Van Westen, C. J. (2013) ‘Remote Sensing and GIS for

Natural Hazards Assessment and Disaster Risk

Management’, in Schroder, J. F. and Bishop, M. P. (eds)

Treatise on Geomorphology. Academic P. San Diego,

USA.: Elsevier Inc., pp. 259–298.

Integration of Remote Sensing Data to Facilitate Multi-Hazards Risk Assessments in Coastal Regions