Transcription

1 RISK ASSESSMENT AND EVALUATION ArcGIS toolbox USER S MANUAL CRATER COASTAL RISK ANALYSIS OF TSUNAMIS AND ENVIRONMENTAL REMEDIATION

2

3 RISK ASSESSMENT AND EVALUATION ArcGIS toolbox USER S MANUAL CRATER COASTAL RISK ANALYSIS OF TSUNAMIS AND ENVIRONMENTAL REMEDIATION

4 Italian Ministry for the Environment and Territory Via Cristoforo Colombo Roma Italy Asian Disaster Prepardness Center P.O. Box 4, Klong Luang, Pathumthani Thailand CONTRIBUTING AUTHORS Dr. Filippo Dall Osso Alessandra Cavalletti. PhD Eng. Paolo Polo Contact:

5

6

7 CONTENTS: 1. Tsunami vulnerability and risk analysis Introduction Vulnerability level Multi-criteria analysis TUTORIAL: Creating a built environment vulnerability map Creating a population vulnerability map Creating a socio-economic aspects vulnerability map Creating an environment vulnerability map Hazard mapping Risk level TUTORIAL: Creating a built environment risk map... 70

8

9 1. TSUNAMI VULNERABILITY AND RISK ANALYSIS 1.1 Introduction The tsunami event occurred on 26th December 2004 in South Eastern Asia caused more than casualties along the coasts of Indonesia, Thailand, Malaysia, India, Sri Lanka and other countries facing the Indian Ocean. Tsunami was initiated by an extremely high magnitude earthquake (9.3 on the Richter scale) localized a few kilometers eastward of the Sumatra coasts, along the subductive system characterized by the Sunda Arc. Because of the presence of this active tectonic margin a new Tsunami event can not be excluded in the future. TSUNAMI VULNERABILITY AND RISK ANALYSIS 1

10 In order to better face any future occurrence of Tsunamis, the SAVE module of the CRATER project focuses on the risk analysis method. In the last years a number of International and National associations, such as UNDP (United Nations Development Program), NOAA (National Oceanic and Atmospheric Administration, U.S.) and FEMA (Federal Emergency Management Agency, U.S.) suggested guidelines for extreme natural events risk analysis. CRATER project started from these guidelines and from data gathered after the Tsunami of 26 th December One of the main aims of the CRATER project was to create a complete tool for Tsunami risk analysis in coastal areas. A risk to a natural event is defined as the mathematical product between vulnerability and hazard; it refers to the expected loss from a given hazard to a given element at risk. Vulnerability is defined as the potential for damage while hazard, for a Tsunami event, is defined as the wave height. Risk management is a two parts process involving risk assessment and risk evaluation. 2 RISK ASSESSMENT AND EVALUATION

11 Risk assessment is mainly a scientific and quantitative exercise out coming from analysis of field and-or experimental data (e.g. modeled tsunami wave height) and from an overall understanding of the nature of the hazard and of vulnerable parameters (UNDP, 1994). Risk evaluation joins perceived risk to a more enlarged qualitative analysis including, for example, cost benefit trade off and socio economic impact. Main vulnerable parameters -in the case of a Tsunami event- are: population, built environment, infrastructures, ecosystem and environment. Understanding vulnerability and hazard level, identification of possible mitigation measures, of the socio-economic impact caused by the event and the cost-benefit trade off they all give information needed to evaluate the level of risk. Decision makers and end users such as local authorities, NGO, disaster and prevention officers will be able to decide the level of risk and plan which protection level is needed to put in place proposed mitigation measures. TSUNAMI VULNERABILITY AND RISK ANALYSIS 3

12 4 RISK ASSESSMENT AND EVALUATION

13 2. VULNERABILITY LEVEL Vulnerability parameters identifies for the study are: - population - built environment - socio-economic aspects - environment For each of these parameters a list of impact elements have been made. Impact elements are those characteristics of the parameter considered that could be mostly affected by the tsunami wave. Summation of the impact elements defines the vulnerability of a chosen parameter. VULNERABILITY LEVEL 5

14 For example, in the case of built environment the impacts elements are: - building material m - description of ground floor g - number of stories s - design d - foundations f For every building we assigned a numeric value to each impact elements. These values are integer numbers ranging between 1 and 5, where 5 stands for the maximum contribution given by the impact element to the total building vulnerability. A building made of concrete will have a low value of m, for example m=1, because it is supposed to be more resistant to the Tsunami wave. If the building material is wood the value of m will be higher, for example m=5. As for the impacts elements, a level of vulnerability ranging between 1 and 5 has been chosen for each vulnerable parameter. 6 RISK ASSESSMENT AND EVALUATION

15 Vulnerability level is given by: V(a, A) = S i (w i e i ) for i = 1, n (1) Where: V (a, A) = vulnerability level of the element a (e.g. a given building), belonging to the vulnerability parameter A (e.g. the built environment). w i = weighting coefficient e i = vectorial value estimated for the impact element n= total number of impact elements related to the parameter A. VULNERABILITY LEVEL 7

16 2.1 Multi-criteria analysis Multi-criteria analysis is a decision making process. CRATER team experts have optimized this technique and applied it for the first time to a vulnerability problem. Outcome of the multi-criteria analysis are the weighting coefficients w i of the relation (1). The practical example given below, in which multi-criteria analysis is applied to the built environment vulnerable parameter, would help the understanding of this process. First step: to identify the impact elements that need to be weighted: Impact elements: Building material (m) Description of ground floor (g) Number of stories (s) 8 RISK ASSESSMENT AND EVALUATION

17 Design (d) Foundations (f) Second step: identify the weighting criteria: Weighting Criteria: Structural damage Damage given to flooding Weighting criteria are identifying the type of damage the parameter would be subject to. VULNERABILITY LEVEL 9

18 Third step is a pair wise comparison between weighting criteria. Structural damage Damage given flooding to Fictitious factor Total Weight (= Total/3) Structural damage \ Damage given to flooding 0 \ Fictitious factor 0 0 \ 0 0 In the matrix above the weighting criteria have been compared with pair wise matches. Introduction of a so called fictitious factor is needed for obvious calculation reasons. From the above table is clear that in this case a structural damage has more weight than the damage given to flooding. 10 RISK ASSESSMENT AND EVALUATION

19 Impact elements are valued among themselves in respect to a single weighting criterion. This should give a ranking level of impact element in respect to a given weighting criterion. First table below is relative to the structural damage criterion, second one to the damage given to flooding. Structural damage m G s f D Fictitious factor Total Relative weight = Total/ 15 m \ G 1 \ S 0 0 \ F \ D \ Fictitious factor \ 0 0 VULNERABILITY LEVEL 11

20 Damage given to flooding M G s Fictitious factor Total Relative weight = Total/ 6 M \ G 1 \ S 1 1 \ Fictitious factor \ 0 0 Elements d and t have not been inserted being these elements not related to flooding. 12 RISK ASSESSMENT AND EVALUATION

21 Summarizing the above results, we obtain: Weight criterion of Relative weight m Relative weight g Relative weight s Relative weight d Relative weight f Structural damage Damage given flooding to VULNERABILITY LEVEL 13

22 Total weight calculation: Total weight m Total weight g Total weight s Total weight d Total weight f Structural damage Damage given flooding to TOTAL NB. Total weight of each impact element, in respect to both criteria, is calculated as product between the criterion weight and the relative weight of the impact element considered. Final calculation of building vulnerability will be then given by: V(a, built environment) = m g s d f 14 RISK ASSESSMENT AND EVALUATION

23 Single values of impact elements m, g, s, d, f are ranging between 1 and 5 as previously stated. This method could be applied to any of the other vulnerability parameters. Once vulnerability level is calculated it will be possible to insert these values in a vulnerability map. Next paragraph describes how to create a built environment vulnerability map using GIS software. The method shown can be applied also to all others vulnerable parameters. VULNERABILITY LEVEL 15

24 2.2 TUTORIAL: Creating a built environment vulnerability map In this paragraph we present a list of all the steps required to create a built environment vulnerability map of the study area using the ArcGis ArcMap 9.0 software by ESRI. The aim is to assign to each building in the pilot area a vulnerability level value, as an integer number between 1 and 5. The CD-Rom included with the manual contains all the files needed to perform the case study simulation. Starting Data: 1. A geo-referenced shape file representing all the buildings of the study area as polygons. We will call it buildings. This shape file for the pilot area is located in Buildings_Vulnerability_tutorial/Shapefile. 16 RISK ASSESSMENT AND EVALUATION

25 2. The attribute table of the buildings shape file must contain one field for each impact element related to the built environment. These impact elements are: a) Building material field name: Build_mat b) Description of ground floor field name: g_floor c) Foundations field name: foundat d) Design field name: design e) Number of stories field name: BL_NSTOREY You should have data for any impact element for each building. The five fields containing the impact elements must be filled with all these data. For example, if a building is made of concrete, the user will have to write concrete in the field Build_mat, at the line related to that building. VULNERABILITY LEVEL 17

26 In the case that you miss information about any of the impact elements of a particular building you just have to write unknown in the relative field, at the line related to that building. Steps required: 1. Open ArcGis ArcMap 9.0. Import all shape files in the Building_vulnerability_tutorial/Shapefile folder and the aerial photo located in the Aerial_Photo folder. 2. Create 5 new fields (short integer) in the attribute table of buildings shape file, one for each impact element, and call them: m for the building material g for description of ground floor s for the number of stories 18 RISK ASSESSMENT AND EVALUATION

27 d for the design f for the foundations Each of these fields must be filled with the impact elements value that will be introduced at step 3. To create a new field: a) Open the buildings attribute table and click on Options, then choose Add field (Figure 1). VULNERABILITY LEVEL 19

28 Figure 1 20 RISK ASSESSMENT AND EVALUATION

29 b) In the window that appeared type m in the Name box and select Short Integer in the Type box, then click Ok (Figure 2). Figure 2 VULNERABILITY LEVEL 21

30 c) Do the same for g, s, d, f, obtaining (Figure 3): Figure 3 22 RISK ASSESSMENT AND EVALUATION

31 3. Choose an integer numeric value between 1 and 5 for each impact element, where 5 stands for maximum vulnerability and 1 for minimum vulnerability. For example, the impact element value for building material (that is m ) will be 5 if the structure is made of wood (high vulnerability), 3 if it is made of bricks (mean vulnerability) and 1 if it is made of reinforced concrete (low vulnerability). Suggested impact element s values are listed in Table 1. VULNERABILITY LEVEL 23

32 IMPACT ELEMENT Suggested impact element s value: =1 =2 =3 =4 =5 m Reinforced concrete Bricks or wood + concrete Wood g Open plan without movable objects Open plan with movable objects No open plan s 5 stories 4 stories 3 stories 2 stories 1 story f deep pile foundations mean foundations surface spread foundations d long building side perpendicular to the coast line long building side oblique to the coast line long building side parallel to le coast line Table 1 24 RISK ASSESSMENT AND EVALUATION

33 4. Put the numeric value chosen for each impact element in the relative field ( m, g, s, d, f ) following these instructions: a) Open the attribute table of the buildings layer b) Click on Options and then on Select by attributes c) Double click on the field build_mat d) Click on the symbol = e) Click on the button Get Unique Values. A list of all the buildings materials will appear in the box on the right. f) Double click on the first material of the list, for example concrete (Figure 4) VULNERABILITY LEVEL 25

34 Figure 4 g) Click on Apply. All buildings made of concrete are now selected. 26 RISK ASSESSMENT AND EVALUATION

35 h) Close the attribute table i) Click on the editor button and choose start editing j) Select the buildings layer in the Target box k) Click on the attribute button, on the left of the Target box (Figure 5). A new window called attributes will appear (Figure 6) VULNERABILITY LEVEL 27

36 Figure 5 28 RISK ASSESSMENT AND EVALUATION

37 Figure 6 VULNERABILITY LEVEL 29

38 l) Double click on buildings group in the box on the left m) In the box on the right, click on the m field in the Property column. The line of the m field will be selected. n) Click in the m line you have selected, under the Value column o) Type the numeric value you want to assign to the impact element m, for all buildings made of concrete (for example if m=1 type 1)(Figure 7). 30 RISK ASSESSMENT AND EVALUATION

39 Figure 7 VULNERABILITY LEVEL 31

40 p) Press enter. Now all buildings made of concrete have an impact element value equivalent to 1 (m=1). q) Save your edits (Click on editor button and choose Save Edits ) r) Repeat all from line a), choosing the subsequent building material at line f). When all building materials have been considered it could remain only some buildings with unkown material: assign them the maximum impact element s value, that is 5. s) Repeat all changing impact element at line c). Finally you will obtain an attribute table looking like the one in Figure RISK ASSESSMENT AND EVALUATION

41 Figure 8 VULNERABILITY LEVEL 33

42 5. Calculate the vulnerability level of every building. a) Stop editing clicking on Editor and select Stop editing. b) Open the buildings attribute table and click on Options, then choose Add field c) In the window that appears type BV1 in the Name box and select Float in the Type box. Then put both precision and scale equal to 2 (Figure 9). Click Ok. 34 RISK ASSESSMENT AND EVALUATION

43 Figure 9 VULNERABILITY LEVEL 35

44 d) Restart editing e) In the attribute table select the BV1 field clicking on the name of the field. f) Right click on the field name BV1 in the attribute table and select Calculate values (Figure 10) 36 RISK ASSESSMENT AND EVALUATION

45 Figure 10 VULNERABILITY LEVEL 37

46 g) Insert the relation for the computation of the building vulnerability level in the box under BV1=. The relation is: BV1= m g s d f Click on m,g,s,d,f in the Fields box and on the mathematical operators buttons to insert them in the relation above (Figure 11). This relation comes from the general vulnerability level relation, in which weighting factors have been calculated using multi-criteria analysis. 38 RISK ASSESSMENT AND EVALUATION

47 Figure 11 VULNERABILITY LEVEL 39

48 f) Click Ok. In the field BV1 you have calculated the vulnerability level of each building as a floating number (Figure 12). In order to obtain integer numbers you must approximate them all. 40 RISK ASSESSMENT AND EVALUATION

49 Figure 12 VULNERABILITY LEVEL 41

50 g) Close the attribute table. Save your edits and stop editing h) Open the attribute table. Add a new field called BV_MAX (short integer). i) Select by attributes ( Options - Select by Attributes ) all buildings with a BV1 value included in the [0;1] interval (BV1>0 and BV10 AND "BV1" 1 and BV12 and BV13 and BV14 and BV1 15 K E RISK ASSESSMENT AND EVALUATION

61 S DN depends on the time arrival of the tsunami wave (day or night), and: S DN = BV/5 if the Tsunami happens during the night (all the people are inside buildings) S DN = (1/2 + BV/10) if the Tsunami happens during the day (people are for a half inside buildings and for a half outside) If the chosen reference area is bigger than a building (for example a village) the BV value is the mean of all buildings inside the reference unit. S H is a switch taking into account the high tourist season, and: S H = 1 if the Tsunami occurs in high tourist season S H = 0 if the Tsunami occurs in low tourist season VULNERABILITY LEVEL 53

62 S L is a switch taking into account the low tourist season, and: S L = 1 if the Tsunami occurs in low tourist season S L = 0 if the Tsunami occurs in high tourist season PV H (population vulnerability in High season) is the population vulnerability level in the reference unit, calculated for high season using relation (1) and the population impact elements listed above. PV H = [w 1 (density) H + w 2 (gender) H + w 3 (children/seniors/invalids) H + w 4 (mean income) H ] Weights must be calculated with multi-criteria analysis. To calculate weights, chose as first the population weighting criteria and follow the example given for buildings vulnerability. PV L (population vulnerability in Low season) is the population vulnerability level in the reference unit, calculated for low season. 54 RISK ASSESSMENT AND EVALUATION

63 PV H = [w 1 (density) L + w 2 (gender) L + w 3 (children/seniors/invalids) L + w 4 (mean income) L ] 2.4 Creating a Socio-Economic Aspects vulnerability map Expressed in the simplest terms, a disaster affects assets (direct damages) and the flow for the production of goods and services (indirect losses) (ECLAC, 2003). In this way socio-economic aspects can suffer both direct and indirect damages. Only vulnerability to direct damages can be plotted on maps, because indirect damages are not linked to any particular location on the territory. Indirect damages must be considered from a qualitative-descriptive point of view. A direct damages vulnerability map can be created following the method presented above. VULNERABILITY LEVEL 55

64 For buildings the only socio-economic impact element is the buildings use. The relation (1) is reduced to: SEV (Socio-Economic Vulnerability) = (e 1 x BV)/5 where e 1 is the value given to the impact element building use. Suggested values for e 1 are shown in the following table: 56 RISK ASSESSMENT AND EVALUATION

65 Socio-economic vulnerability of buildings Building use e 1 PUBLIC HEALTH 5 EDUCATION 3 DRINKING WATER AND SANITATION 4 TRANSPORTS 3 ENERGY 3 INDUSTRY AND COMMERCIAL 2 AGRICOLTURE AND LIVESTOCK 3 TOURISM 4 AUTHORITIES 5 RELIGIOUS/HISTORICAL 2 LIVING HOUSE 1 VULNERABILITY LEVEL 57

66 For land use the relation (1) becomes: SEV = (e 1 ) Suggested e 1 values for land use are shown in the following table: 58 RISK ASSESSMENT AND EVALUATION

67 Socio-economic vulnerability of land use Building use e 1 TRANSPORTS 3 RIVER AND CHANNELS 1 LAKES AND WET LANDS 2 NOT CULTIVATED LANDS 1 AGRICOLTURE: ANNUAL CROPS 2 AGRICOLTURE: PERMANENT PLANTATIONS 3 BEACH 2 FISHING POOLS 3 VULNERABILITY LEVEL 59

68 2.5 Creating an Environment vulnerability map For a more specific analysis we have divided the Environment into four sub-parameters: 1. Surface water 2. Ground water 3. Coastal zone 4. Internal areas 60 RISK ASSESSMENT AND EVALUATION

69 Impacts elements are: 1. Surface water a) the river capacity b) rainfall/evaporation 2. Ground water a) presence of low lying areas, eroded pockets, restricted drainage canals b) number of wells c) tide effect on rivers mouths VULNERABILITY LEVEL 61

70 d) rainfall/evaporation e) soil permeability f) depth of the aquifer g) the hydraulic conductivity of the aquifer 3. Coastal zone a) Coastal features (sandy beach, sandy beach with dunes, coast with rocks) b) Artificial coastal defenses (walls behind the beach) c) Topography (flat area= high vulnerability) d) Wetlands extent, protected wetlands 62 RISK ASSESSMENT AND EVALUATION

71 e) Submerged and intertidal zones features (coral reef extent and depth, mangroves forests extent, Posydonia praires extent) VULNERABILITY LEVEL 63

72 4. Internal areas a) protected areas extent (high vulnerability) b) extent of zones of ecological interest (mean vulnerability) c) extent of agricultural areas (low vulnerability) Every sub-parameter must be divided into a number of reference units, for example following the administrative boundaries. Reference unit must be imported in the GIS project as a geo-referenced polygons shape file. For each reference unit you have to calculate the vulnerability level, using relation (1). 64 RISK ASSESSMENT AND EVALUATION

73 3. HAZARD MAPPING In the case of a Tsunami risk, hazard has been defined as the maximum height of the water column reached in each point of the study area during the flooding of the 26 December To obtain these values, CRATER experts simulated the inundation and the consequential flooding using two different numerical models. The first, starting from the record of the tide level gauges, transferred the waves from offshore to the coast line. The second one, starting from the waves heights on the coast line, simulated the flooding of the study area, giving as output the maximum water height reached in each point inland. In order to give a prompt vision of how many floors of each building got inundated, assuming the height of each floor equal to 3 meters, the water height values have been divided in 4 intervals: - From zero to 3 meters H=1 - From 3 to 6 meters H=2 HAZARD MAPPING 65

74 - From 6 to 9 meters H=3 - More than 9 meters H=4 Where H stands for the hazard level. H will be used in the computation of the risk level. Using a GIS software, these four intervals have been plotted on an hazard map, identifying 4 different areas (Figure 17). 66 RISK ASSESSMENT AND EVALUATION

75 Figure 17. Hazard map of the Khao Lak study area, Thailand. HAZARD MAPPING 67

76 Users can obtain an approximated evaluation of the hazard map imaging that the study area will be flooded until the topography level curve having the same height of the Tsunami wave on the coast line. In this way users can calculate the maximum water level reached in each position of the study area as the difference between the Tsunami wave height on the coastline and the ground topographic elevation in that point inland. 68 RISK ASSESSMENT AND EVALUATION

77 4. RISK LEVEL As previously stated, the numeric value of risk can be calculated as the product between vulnerability and hazard level. Since vulnerability level ranges from 1 to 5 and hazard level ranges from 1 to 4, risk level of each vulnerable element will be given by: R=VxH/4 R must be an integer number ranging from 1 to 5, where 5 stands for the maximum risk level. Once risk level has been calculated it will be possible to plot it on a risk map, using a GIS software. The following paragraph shows how to create a risk map for the built environment vulnerable parameter. The same method allows to calculate the risk level for all others vulnerable parameters. RISK LEVEL 69

78 4.1 TUTORIAL: Creating a built environment risk map Now we are going to list all the steps required to create a built environment risk map of the study area, using the ArcGis ArcMap 9.0 software by ESRI. This tutorial starts from the results given above. The main aim is to assign to each building a risk level value as an integer number between 1 (minimum risk) and 5 (maximum risk). The CD-Rom included with the manual contains all the files needed to perform this tutorial. Starting data: The risk level of each building is given by the product between vulnerability level and hazard level, so data required for creating a buildings risk map are: The buildings shape file obtained resulting from the building vulnerability tutorial. In the attribute table of this shape file there is a field called BV_MAX reporting the vulnerability 70 RISK ASSESSMENT AND EVALUATION

79 level of each building of the pilot area. This shape file is located in the Building_risk_tutorial /Shapefile folder, with the name of buildings. The Hazard map of the pilot area. Hazard map is a polygons shape file called hazard map located in the Building_risk_tutorial /Shapefile folder. Every polygon represents an area with constant maximum water level reached during the Tsunami of December 26, 2004.The buildings attribute table reports in the field HAZARD the hazard level value for each polygon. Steps required: 1. Open ArcGis ArcMap 9.0. Import all shape files present in the Building_risk_tutorial/Shapefile folder and the aerial photo located in the Aerial_Photo folder of the CD Rom. 2. Open the Arc Toolbox. RISK LEVEL 71

80 3. Double click on Analysis Tools 4. Double click on Overlay 5. Double click on Intersect. The Intersect window appears. 6. In the Input Features box select the buildings layer and the hazard_map layer. 7. In the Output Feature Class box select the folder in which you want the output shape file of the intersection to be saved (Figure 18) 72 RISK ASSESSMENT AND EVALUATION

81 Figure 18 RISK LEVEL 73

82 8. Click OK. A new polygons shape file called buildings_intersect appears in your layer s list (Figure 19). This shape file is the intersection between buildings shape file and hazard_map shape file. His attributes table contains both buildings fields and hazard_map fields (Figure 20). 74 RISK ASSESSMENT AND EVALUATION

83 Figure 19 RISK LEVEL 75

84 Figure 20. The buildings_intersect attribute table contains both buildings fields and hazard_map fields 76 RISK ASSESSMENT AND EVALUATION

85 9. Open the buildings_intersect attribute table. Add a new field and call it R1 (Type: Float, Field properties: Precision =2, Scale=2) ( Figure 21). Click OK RISK LEVEL 77

86 Figure Start editing 11. In the buildings_intersect attribute table select the R1 field clicking on the name of the field. 12. Right click on the field name R1 in the attribute table and select Calculate values (Figure 22) 78 RISK ASSESSMENT AND EVALUATION

87 Figure 22 RISK LEVEL 79

88 13. Insert the relation for the computation of the building risk level in the box under R1=. The relation is: R1 = BV_MAX * HAZARD / 4 Click on BV_MAX and on HAZARD in the Fields box and on the mathematical operators buttons to insert them in the relation above (Figure 23). 80 RISK ASSESSMENT AND EVALUATION

89 Figure 23 RISK LEVEL 81

90 14. Click Ok. In the field R1 you have just calculated the risk level of each building, but not as integer numbers (Figure 24). In order to obtain integer numbers you must approximate them all. Figure RISK ASSESSMENT AND EVALUATION

91 15. Close the attribute table. Save your edits and stop editing 16. Open the buildings_intersect attribute table. Add a new field and call it RISK_MAX (short integer). 17. Select by attributes ( Options - Select by Attributes ) all buildings with a R1 value included in the ]0;1] interval (R1>0 and R10 AND "R1" 1 and BV12 and BV13 and BV14 and BV1