scieee AI-readable full text Open interactive document viewer

On the use of logistic functions for coastal flood assessment

Teruel Cano, David

Published
2017-06-01
Publisher
Universitat Politècnica de Catalunya
Repository
upcommons.upc.edu
Language
en

Abstract

Coastal flooding results as the combination of wave conditions, mean water level and terrain characteristics. The associated damage can be expressed as a function of the flooding water column and it will rarely remain stationary in a sea level scenario. The objective of the study is to determine the importance of the dynamic (waves) and hydrostatic (sea level) component as a function of a damage level and to determine what combinations of hydrodynamic drivers may lead to unacceptable damage levels. The analysis has been performed through the use of logistic functions in the beach of Sant Pere Pescador (Girona province), covering an extension of 6,3 kilometres with an average width of 90m. The approach of the research consists of the creation of two multinomial regression models to be analyzed for 6 hydrodynamic scenarios derived from a coupled hydromorphodynamic model, one made in a node-by-node basis and the other one through polygons. The results show that a node-by-node model has nearly 10% of information loss due to the small variability and the small number of measurements. However, both models successfully predict similar expected prediction and probability to reach a flood categories for the proposed scenarios. Results from the multinomial regression clearly reveals sea level rise as the main contribution to flood increase, with the wave return period being less significant. The model also suggests that a drastic increase in the flooding water column, approximately between 0,6 and 1 m will occur for sea level rise bigger than 0,25m. According the IPCC predictions for the sea level rise in the Mediterranean Coast, this is expected to occur between a time range of 2040-2070 (for the RCP 8.5, the most dramatic climate scenario) or starting in 2045 (for RCP 2.6 and RCP 4.5, less severe scenarios). For what concerns the spatial extension of the flooding, the expected categories of flooding shows that the northern region of the Sant Pere Pescador beach is likely to suffer more severe flooding than the central and southern regions, both in intensity and spatial extent, reaching water depths larger than 0,6 - 1m for scenarios with sea level rise larger than 0,25m.

Full text

On the use of Logistic Functions for Coastal Flood Assessment Date: July 2017 Author: David Teruel Cano Annex 2 Maps Guidelines and references The following Annex contains a list of the most relevant maps generated during this project. Although numbers and data is necessary to obtain and predict the flooding of a certain region, it is equally important to understand how those predictions translate into the environment, and the interdependences that those values have with the surroundings. During the project, several maps have been designed and analysed with the use of QGIS. However, due to the big number of pictures generated, it was not possible to include them all in the memory. Therefore, in this Annex, the main maps are collected. They can be consulted while reading the project, since the information included results, in many cases, from the analysis of the following maps. The document is organised into the following sections: • Predictions of the Point-by-Point Regression model 5 maps that show the results of applying the logistic regression model for 5 predictions: Prediction Sea Level Rise (m) Return Period (year) 1 0 10 2 0 25 3 0 50 4 0,6 75 5 0,6 100 • Predictions of the Polygon-based Regression model This set of maps show the same logistic regression model but applied in a Polygon model. It shows the evolution in categories of water depth for Predictions 1, 3 and 5. • Probability of Reaching Category 2 and 3 for Prediction These maps show the probability of the polygons to reach category 2 and category 3 for the Prediction 5 scenario. This map is especially relevant to show how close would polygons be to increase from category 2 to category 3 or vice versa. 0 . Maps 2 • Flooded area for 6 Climate Scenarios The following maps show the area that would be flooded for the 6 climatic scenarios used to construct the logistic regression model. The figure shows the characteristics of the conditions producing the maps shown: Conditions SLR Return Period 2015 0 10 50 100 2100 0,6 10 50 100 • Assets at Risk This section includes a complete list of assets at risk for the 6 climate scenarios. The maps show the resulting intersection of the flooded area maps and the land uses map described in the project. The main assets analysed are: o Transportation Facilities o Vegetation and Land Cover o Water Resources • Risk Map The last map included in the Annex is the complete risk Map featuring the most vulnerable regions obtained from multiplying the probability of flood occurrence x the consequences. 500 0500 m Predictions of the Point-by-Point Regression model SCENARIO Prediction 1 [3928] 1 [1288] 2 [1822] 3 [522] 4 [0] Legend 1 On the use of logistic functions for Coastal Flood Assessment PREDICTION 1: Flood depth category for Sea Level Rise:0 and a Wave-return Period:10 500 0500 m Predictions of the Point-by-Point Regression model Prediction 2 [3928] 1 [2314] 2 [982] 3 [336] 4 [0] Legend 1 On the use of logistic functions for Coastal Flood Assessment PREDICTION 2: Flood depth category for Sea Level Rise: 0 and a Wave-return Period: 25 500 0500 m Predictions of the Point-by-Point Regression model Prediction 3 [3928] 1 [1288] 2 [1822] 3 [522] 4 [0] Legend 1 On the use of logistic functions for Coastal Flood Assessment PREDICTION 3: Flood depth category for Sea Level Rise: 0 and a Wave-return Period: 50 500 0500 m Predictions of the Point-by-Point Regression model Prediction 4 [3928] 1 [25] 2 [622] 3 [2867] 4 [118] Legend 1 On the use of logistic functions for Coastal Flood Assessment PREDICTION 4: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 75 500 0500 m Predictions of the Point-by-Point Regression model Prediction 5 [3928] 1 [15] 2 [592] 3 [2859] 4 [166] Legend 1 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Predictions of the Polygon-based Regression model Prediction 1 Category 1 Category 2 Category 3 Category 4 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 1: Flood depth category for Sea Level Rise: 0 and a Wave-return Period: 10 300 0300 m Predictions of the Polygon-based Regression model Prediction 1 Category 1 Category 2 Category 3 Category 4 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 1: Flood depth category for Sea Level Rise: 0 and a Wave-return Period: 10 300 0300 m Predictions of the Polygon-based Regression model Prediction 5 Category 1 Category 2 Category 3 Category 4 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 2 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 2 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 2 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 3 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 3 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Probability of Reaching Category 2 and 3 for Prediction 5 Probability of Category 3 0.00 - 0.25 0.25 - 0.50 0.50 - 0.75 0.75 - 1.00 Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Flood depth category for Sea Level Rise: 0,6m and a Wave-return Period: 100 300 0300 m Flooding Area for 6 scenarios SLR:0 + Hs_Tr:10 SLR:0 + Hs_Tr:50 SLR:0 + Hs_Tr:100 SLR:0,6 + Hs_Tr:10 SLR:0,6 + Hs_Tr:50 SLR:0,6 + Hs_Tr:100 Legend 4 On the use of logistic functions for Coastal Flood Assessment Year 2015: 0 Sea Level Rise and Year 2100: 0,6m Sea Level Rise 300 0300 m Assets at Risk - Transportation Transportation flooded Legend 5 On the use of logistic functions for Coastal Flood Assessment Year 2015: 0 Sea Level Rise + Wave return period: 10 Year 2015: 0 Sea Level Rise + Wave return period: 50 300 0300 m Assets at Risk - Transportation Transportation flooded Legend 5 On the use of logistic functions for Coastal Flood Assessment Year 2015: 0 Sea Level Rise + Wave return period: 100 Year 2100: 0,6m Sea Level Rise + Wave return period: 10 300 0300 m Assets at Risk - Water resources Water bodies affected Legend 5 On the use of logistic functions for Coastal Flood Assessment Year 2100: 0,6m Sea Level Rise + Wave return period: 50 Year 2100: 0,6m Sea Level Rise + Wave return period: 100 300 0300 m Risk analysis for Prediction 5 Risk Analysis Low Moderate High Very High Legend 2 On the use of logistic functions for Coastal Flood Assessment PREDICTION 5: Sea Level Rise: 0,6m and a Wave-return Period: 100