2.3 Methodology and Data
2.3.3 Social damage calculations
The methodology used to calculate the number of people killed is very similar to the one used to calculate the economic damage. Small statistical sectors were drawn based on similar characteristics such as building density and land use; inhabitants were assumed to be spread homogeneously across the number of houses. As such, the number of inhabitants per sector was calculated based on the number of houses per statistical sector (ODPEM, 2011) and the number of people per household in 2001, which was an average of 3 for St. Mary (WRA, 2002). The spread of the inhabitants per statistical sector based on the number of houses per sector was statistically tested and has proven to be significant with a significance level of 5%. The number of casualties is expressed by Equation (2) from Chapter 1 (Vrisou van Eck et al., 1999). Since there was no available rise velocity information for the flood event of 2001, the drown factor based on the rise velocity was set at 100%, which corresponds to a rise velocity higher than 3 m per hour. This is the safest presumption, but could lead to an overestimation of the number of casualties.
50
22..33..22..33 RRooaadd ddaam maaggee
A road dataset of the area was available for this research, giving a class and corresponding width to each road (ODPEM, 2013a). Although this information allowed a more precise damage calculation, a comparison with satellite imagery showed big disparities between the available dataset and the reality. Many unpaved roads were missing and the simplification of the data, due to its larger scale, caused the location of other roads to differ from reality. Additionally, damage to roads showed to be significantly smaller than building damage. Therefore, the priority was given to adding all unpaved roads, losing the class information, in order to enhance the accuracy of the building damage spread.
Most roads and bridges in Jamaica are not elevated. Consequently, they suffer considerable damage during a flood. The depth-damage function for roads is the same as in LATIS because roads are constructed in the same manner in both regions and will thus suffer the same proportion of damage. The damage factor 𝑓𝑓 for roads is expressed by the following function (Vanneuville et al., 2003c):
𝑓𝑓 = 𝑚𝑚𝑚𝑚𝑚𝑚 (0.28 × 𝑑𝑑; 0.18 × 𝑑𝑑 + 0.1 ; 1) (6)
where 𝑑𝑑 represents the water depth in meters.
22..33..33 SSoocciiaall ddaam maaggee ccaallccuullaattiioonnss
The methodology used to calculate the number of people killed is very similar to the one used to calculate the economic damage. Small statistical sectors were drawn based on similar characteristics such as building density and land use; inhabitants were assumed to be spread homogeneously across the number of houses. As such, the number of inhabitants per sector was calculated based on the number of houses per statistical sector (ODPEM, 2011) and the number of people per household in 2001, which was an average of 3 for St. Mary (WRA, 2002). The spread of the inhabitants per statistical sector based on the number of houses per sector was statistically tested and has proven to be significant with a significance level of 5%. The number of casualties is expressed by Equation (2) from Chapter 1 (Vrisou van Eck et al., 1999). Since there was no available rise velocity information for the flood event of 2001, the drown factor 𝑓𝑓
𝑤𝑤based on the rise velocity was set at 100%, which corresponds to a rise velocity higher than 3 m per hour. This is the safest presumption, but could lead to an overestimation of the number of casualties.
The drown factor based on flood depth is expressed by the following equation:
𝑓𝑓𝑑𝑑= 𝑒𝑒𝑒𝑒𝑒𝑒 (1.16 × 𝑑𝑑 − 7.3) (7)
45 The drown factor based on flood depth is expressed by the following equation:
(7) where
50
22..33..22..33 RRooaadd ddaam maaggee
A road dataset of the area was available for this research, giving a class and corresponding width to each road (ODPEM, 2013a). Although this information allowed a more precise damage calculation, a comparison with satellite imagery showed big disparities between the available dataset and the reality. Many unpaved roads were missing and the simplification of the data, due to its larger scale, caused the location of other roads to differ from reality. Additionally, damage to roads showed to be significantly smaller than building damage. Therefore, the priority was given to adding all unpaved roads, losing the class information, in order to enhance the accuracy of the building damage spread.
Most roads and bridges in Jamaica are not elevated. Consequently, they suffer considerable damage during a flood. The depth-damage function for roads is the same as in LATIS because roads are constructed in the same manner in both regions and will thus suffer the same proportion of damage. The damage factor 𝑓𝑓 for roads is expressed by the following function (Vanneuville et al., 2003c):
𝑓𝑓 = 𝑚𝑚𝑚𝑚𝑚𝑚 (0.28 × 𝑑𝑑; 0.18 × 𝑑𝑑 + 0.1 ; 1) (6)
where 𝑑𝑑 represents the water depth in meters.
22..33..33 SSoocciiaall ddaam maaggee ccaallccuullaattiioonnss
The methodology used to calculate the number of people killed is very similar to the one used to calculate the economic damage. Small statistical sectors were drawn based on similar characteristics such as building density and land use; inhabitants were assumed to be spread homogeneously across the number of houses. As such, the number of inhabitants per sector was calculated based on the number of houses per statistical sector (ODPEM, 2011) and the number of people per household in 2001, which was an average of 3 for St. Mary (WRA, 2002). The spread of the inhabitants per statistical sector based on the number of houses per sector was statistically tested and has proven to be significant with a significance level of 5%. The number of casualties is expressed by Equation (2) from Chapter 1 (Vrisou van Eck et al., 1999). Since there was no available rise velocity information for the flood event of 2001, the drown factor 𝑓𝑓
𝑤𝑤based on the rise velocity was set at 100%, which corresponds to a rise velocity higher than 3 m per hour. This is the safest presumption, but could lead to an overestimation of the number of casualties.
The drown factor based on flood depth is expressed by the following equation:
𝑓𝑓𝑑𝑑= 𝑒𝑒𝑒𝑒𝑒𝑒 (1.16 × 𝑑𝑑 − 7.3) (7)
represents the water depth in meters (Vrisou van Eck et al., 1999). People are assumed to be inside their homes during the flood. Therefore, the doorstep level of 0.5 m has to be overcome before the house is considered as inundated (Deckers et al., 2009; ODPEM, 2013a; Vanneuville et al., 2003a).
2.3.4
Risk calculationsFor the case study of Annotto Bay, a social and an economic damage map were created based on a flood map of 2001 and not on a flood hazard map with a certain AEP.
Therefore, the corresponding flood risk maps could not be created in this study. In the future, however, a rainfall-runoff model could be created for the study area and flood hazard maps with AEPs could be generated. For each flood hazard map with a specific AEP, a damage map can be created using the same methodology, visualized in Figure 10. All damage maps can then be combined in order to generate one flood risk map, showing the risk of damage per year. The risk can be calculated by using the following equation (Deckers et al., 2009):
(8) with
51 where 𝑑𝑑 represents the water depth in meters (Vrisou van Eck et al., 1999). People are assumed to be inside their homes during the flood. Therefore, the doorstep level of 0.5 m has to be overcome before the house is considered as inundated (Deckers et al., 2009;
ODPEM, 2013a; Vanneuville et al., 2003a).
22..33..44 RRiisskk ccaallccuullaattiioonnss
For the case study of Annotto Bay, a social and an economic damage map were created based on a flood map of 2001 and not on a flood hazard map with a certain AEP. Therefore, the corresponding flood risk maps could not be created in this study. In the future, however, a rainfall-runoff model could be created for the study area and flood hazard maps with AEPs could be generated. For each flood hazard map with a specific AEP, a damage map can be created using the same methodology, visualized in Figure 10. All damage maps can then be combined in order to generate one flood risk map, showing the risk of damage per year. The risk can be calculated by using the following equation (Deckers et al., 2009):
𝑅𝑅 = ∑𝑖𝑖=11𝑖𝑖(𝐷𝐷𝑖𝑖− 𝐷𝐷𝑖𝑖−1) (8)
with 𝑅𝑅 as the risk, 𝐷𝐷
𝑖𝑖as the damages related to a flood with an AEP of 1 𝑖𝑖 ⁄ .
Since the creation and validation of flood hazard maps with AEPs is time-consuming, in practice only a few would be created. The flood damage between two AEPs will be linearly interpolated in order to simplify the equation (Vanneuville et al., 2003a). This simplification is done based on the choice of AEPs after the creation of the rainfall-runoff models.
22..44 RReessuullttss
22..44..11 EEccoonnoom miicc ddaam maaggee m maapp
Before generating the final economic damage map for Annotto Bay, the three types of damage are visualized separately in Figure 13. The cost of each type of damage and the damaged area is given in Table 7. It is immediately clear that the building damage cost is substantially higher than the cost of the other two types, while the crop damage has the largest spread. The total damaged area is not equal to the sum of the areas per damage type, as some polygons contain more than one type. For example, rural areas with housing will have building damage as well as crop damage.
as the risk,
51 where 𝑑𝑑 represents the water depth in meters (Vrisou van Eck et al., 1999). People are assumed to be inside their homes during the flood. Therefore, the doorstep level of 0.5 m has to be overcome before the house is considered as inundated (Deckers et al., 2009;
ODPEM, 2013a; Vanneuville et al., 2003a).
22..33..44 RRiisskk ccaallccuullaattiioonnss
For the case study of Annotto Bay, a social and an economic damage map were created based on a flood map of 2001 and not on a flood hazard map with a certain AEP. Therefore, the corresponding flood risk maps could not be created in this study. In the future, however, a rainfall-runoff model could be created for the study area and flood hazard maps with AEPs could be generated. For each flood hazard map with a specific AEP, a damage map can be created using the same methodology, visualized in Figure 10. All damage maps can then be combined in order to generate one flood risk map, showing the risk of damage per year. The risk can be calculated by using the following equation (Deckers et al., 2009):
𝑅𝑅 = ∑𝑖𝑖=11𝑖𝑖(𝐷𝐷𝑖𝑖− 𝐷𝐷𝑖𝑖−1) (8)
with 𝑅𝑅 as the risk, 𝐷𝐷
𝑖𝑖as the damages related to a flood with an AEP of 1 𝑖𝑖 ⁄ .
Since the creation and validation of flood hazard maps with AEPs is time-consuming, in practice only a few would be created. The flood damage between two AEPs will be linearly interpolated in order to simplify the equation (Vanneuville et al., 2003a). This simplification is done based on the choice of AEPs after the creation of the rainfall-runoff models.
22..44 RReessuullttss
22..44..11 EEccoonnoom miicc ddaam maaggee m maapp
Before generating the final economic damage map for Annotto Bay, the three types of damage are visualized separately in Figure 13. The cost of each type of damage and the damaged area is given in Table 7. It is immediately clear that the building damage cost is substantially higher than the cost of the other two types, while the crop damage has the largest spread. The total damaged area is not equal to the sum of the areas per damage type, as some polygons contain more than one type. For example, rural areas with housing will have building damage as well as crop damage.
as the damages related to a flood with an AEP of .
Since the creation and validation of flood hazard maps with AEPs is time-consuming, in practice only a few would be created. The flood damage between two AEPs will be linearly interpolated in order to simplify the equation (Vanneuville et al., 2003a). This simplification is done based on the choice of AEPs after the creation of the rainfall-runoff models.
2.4 Results
2.4.1
Economic damage mapBefore generating the final economic damage map for Annotto Bay, the three types of damage are visualized separately in Figure 13. The cost of each type of damage and the damaged area is given in Table 7. It is immediately clear that the building damage cost is substantially higher than the cost of the other two types, while the crop damage has the largest spread. The total damaged area is not equal to the sum of the areas per damage type, as some polygons contain more than one type. For example, rural areas with housing will have building damage as well as crop damage.
50
22..33..22..33 RRooaadd ddaam maaggee
A road dataset of the area was available for this research, giving a class and corresponding width to each road (ODPEM, 2013a). Although this information allowed a more precise damage calculation, a comparison with satellite imagery showed big disparities between the available dataset and the reality. Many unpaved roads were missing and the simplification of the data, due to its larger scale, caused the location of other roads to differ from reality. Additionally, damage to roads showed to be significantly smaller than building damage. Therefore, the priority was given to adding all unpaved roads, losing the class information, in order to enhance the accuracy of the building damage spread.
Most roads and bridges in Jamaica are not elevated. Consequently, they suffer considerable damage during a flood. The depth-damage function for roads is the same as in LATIS because roads are constructed in the same manner in both regions and will thus suffer the same proportion of damage. The damage factor 𝑓𝑓 for roads is expressed by the following function (Vanneuville et al., 2003c):
𝑓𝑓 = 𝑚𝑚𝑚𝑚𝑚𝑚 (0.28 × 𝑑𝑑; 0.18 × 𝑑𝑑 + 0.1 ; 1) (6)
where 𝑑𝑑 represents the water depth in meters.
22..33..33 SSoocciiaall ddaam maaggee ccaallccuullaattiioonnss
The methodology used to calculate the number of people killed is very similar to the one used to calculate the economic damage. Small statistical sectors were drawn based on similar characteristics such as building density and land use; inhabitants were assumed to be spread homogeneously across the number of houses. As such, the number of inhabitants per sector was calculated based on the number of houses per statistical sector (ODPEM, 2011) and the number of people per household in 2001, which was an average of 3 for St. Mary (WRA, 2002). The spread of the inhabitants per statistical sector based on the number of houses per sector was statistically tested and has proven to be significant with a significance level of 5%. The number of casualties is expressed by Equation (2) from Chapter 1 (Vrisou van Eck et al., 1999). Since there was no available rise velocity information for the flood event of 2001, the drown factor 𝑓𝑓
𝑤𝑤based on the rise velocity was set at 100%, which corresponds to a rise velocity higher than 3 m per hour. This is the safest presumption, but could lead to an overestimation of the number of casualties.
The drown factor based on flood depth is expressed by the following equation:
𝑓𝑓𝑑𝑑= 𝑒𝑒𝑒𝑒𝑒𝑒 (1.16 × 𝑑𝑑 − 7.3) (7)
51 where 𝑑𝑑 represents the water depth in meters (Vrisou van Eck et al., 1999). People are assumed to be inside their homes during the flood. Therefore, the doorstep level of 0.5 m has to be overcome before the house is considered as inundated (Deckers et al., 2009;
ODPEM, 2013a; Vanneuville et al., 2003a).
22..33..44 RRiisskk ccaallccuullaattiioonnss
For the case study of Annotto Bay, a social and an economic damage map were created based on a flood map of 2001 and not on a flood hazard map with a certain AEP. Therefore, the corresponding flood risk maps could not be created in this study. In the future, however, a rainfall-runoff model could be created for the study area and flood hazard maps with AEPs could be generated. For each flood hazard map with a specific AEP, a damage map can be created using the same methodology, visualized in Figure 10. All damage maps can then be combined in order to generate one flood risk map, showing the risk of damage per year. The risk can be calculated by using the following equation (Deckers et al., 2009):
𝑅𝑅 = ∑𝑖𝑖=11𝑖𝑖(𝐷𝐷𝑖𝑖− 𝐷𝐷𝑖𝑖−1) (8)
with 𝑅𝑅 as the risk, 𝐷𝐷
𝑖𝑖as the damages related to a flood with an AEP of 1 𝑖𝑖 ⁄ .
Since the creation and validation of flood hazard maps with AEPs is time-consuming, in practice only a few would be created. The flood damage between two AEPs will be linearly interpolated in order to simplify the equation (Vanneuville et al., 2003a). This simplification is done based on the choice of AEPs after the creation of the rainfall-runoff models.
22..44 RReessuullttss
22..44..11 EEccoonnoom miicc ddaam maaggee m maapp
Before generating the final economic damage map for Annotto Bay, the three types of
damage are visualized separately in Figure 13. The cost of each type of damage and the
damaged area is given in Table 7. It is immediately clear that the building damage cost is
substantially higher than the cost of the other two types, while the crop damage has the
largest spread. The total damaged area is not equal to the sum of the areas per damage
type, as some polygons contain more than one type. For example, rural areas with housing
will have building damage as well as crop damage.
THE URBAN CASE STUDY: ANNOTTO BAY, JAMAICA
Figure 13 Economic damage maps for the 2001 flood in Annotto Bay, Jamaica, per damage type (top left:
building damage, to right: road damage, bottom: crop damage).
Table 7 Overview of the damage cost and damaged area per damage type.
DAMAGE COST (USD) DAMAGED AREA (m²)
Building damage 7 080 000 520 000
Road damage 50 000 140 000
Crop damage 370 000 2 050 000
TOTAL DAMAGE 7 500 000 2 460 000
As a result of the use of the two buffers for building locations, the total damaged building area is limited. Without buffers, the damaged area for buildings would be 1 860 000 m². However, when only looking at the average building area of the houses that were damaged, the building area of the damaged houses is only 130 000 m².
Hence, the buffers still cause an overestimation of the damaged building area, but the result is much closer to the reality.
In a next step, the final economic damage map was generated, by adding up the three separate damage types. This map is shown in Figure 14. Due to the high damage values of buildings, the distinction in crop damage has disappeared. However, it is clear the high risk areas are located in the low-lying urban areas and along the main road, parallel to the coastline. The wetlands do not have any damage, except for the buffers along the roads, since a few houses are located there.
Figure 14 Economic damage map for the 2001 flood in Annotto Bay.
With a maximum damage cost of 175 USD m-², this flooding has not completely destroyed any houses, due to the low flow velocity of the water (< 3 m s-1)(ODPEM, 2013a).
Considering that the replacement value of buildings is on average 836 USD m-², the estimated damage is almost five times lower than the value for complete destruction.
2.4.2
Social damage mapThe number of calculated casualties per km² is displayed in Figure 15. The number is considerably higher in the built-up area, because of the higher population density in this area. In Table 8, an overview is given of the absolute and relative estimated numbers. The total number of people killed during the flood is extremely low, almost zero.
THE URBAN CASE STUDY: ANNOTTO BAY, JAMAICA
Figure 15 Social damage maps for Annotto Bay, Jamaica.
Table 8 Overview of the estimated number of casualties for Annotto Bay, Jamaica.
ESTIMATED VALUES Total number of casualties 1.74
Average number of casualties per km² 0.03 Maximum number of casualties per km² 0.34
2.5 Discussion
Although several cost and damage assessments have been done for the entire
Although several cost and damage assessments have been done for the entire