Genetic improvement of shoreline evolution forecasting models

Najar, Mahmoud Al; Almar, Rafaël; Bergsma, Erwin W. J.; Delvit, Jean-Marc; Wilson, Dennis G.

doi:10.1145/3520304.3534041

Cited by 3 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we use a mostly standard CGP representation with modifications to optimization for NSGA-II and more efficient mutations, and modifications to the function set in order to represent ShoreFor. As in (Al Najar et al, 2022), we employ the following mutation-level constraints: 1) We discard all mutated graphs with direct input-output connections. 2) We ensure that for the same set of random inputs, the outputs produced by parent graph and the mutated graph are different in order to minimze the chances of having behaviorally identical individuals in the population.…”

Section: Cartesian Genetic Programmingmentioning

confidence: 99%

Improving Shoreline Forecasting Models with Multi-Objective Genetic Programming

Najar¹,

Almar²,

Bergsma³

et al. 2023

Preprint

View full text Add to dashboard Cite

Given the current context of climate change and the increasing population densities at coastal zones around the globe, there is an increasing need to be able to predict the development of our coasts. Recent advances in artificial intelligence allow for automatic analysis of observational data. Symbolic Regression (SR) is a type of Machine Learning algorithm that aims to find interpretable symbolic expressions that can explain the relations in the data. In this work, we aim to study the problem of forecasting shoreline change using SR. We make use of Cartesian Genetic Programming (CGP) in order to encode and improve upon ShoreFor, a shoreline prediction model. Coupled with NSGA-II, the CGP individuals are evaluated and selected during evolution according to their predictive skills at five different coastal sites. This work presents a comparison between the CGP-evolved models and the base ShoreFor model. In addition to its ability to produce well-performing models, the work demonstrates the usefulness of CGP as a research tool to gain insight into the behaviors of shorelines at different points around the globe.

show abstract

Section: Cartesian Genetic Programmingmentioning

confidence: 99%

Improving Shoreline Forecasting Models with Multi-Objective Genetic Programming

Najar¹,

Almar²,

Bergsma³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Castelle et al (2022) explored time–space averaging of SDS (satellite‐derived shoreline) data along with wave hindcasting, longshore drift estimations, and climate indices. Najar et al (2022) used Cartesian genetic programming (CGP) to implement ShoreFor , a shoreline equilibrium model for forecasting wave‐driven shoreline changes. Pradeep et al (2022) used remote sensing data and DSAS tools to analyse coastal variation, and the findings were used further in a machine learning algorithm (K‐means clustering) to study the west coast of India.…”

Section: Introductionmentioning

confidence: 99%

Improved geospatial analysis of shoreline modification using a weighted‐average‐based novel formulation

Mohanty

Rajani

Majumdar

et al. 2023

Earth Surf Processes Landf

View full text Add to dashboard Cite

This study presents a novel weighted-average method (WAM) that captures the detailed nuances of shoreline change patterns over long and short periods by considering shorelines belonging to five different time frames. Using the WAM, the average shoreline modification for the longer period (136 years with intervals of $34 years) and for the shorter period (annual changes over a period of 5 years) have been calculated for the 9 km stretch of the Kollam coast in Kerala, India. Historically, Kollam has been a prominent seaport on the Malabar Coast of India and was extensively mapped due to its importance for political, strategic, and trade reasons. The coastal information available in these older maps is unique as there are no comparable data for the coastline of that period that can be used to assess changes along the coast. The WAM demonstrated the efficacy of the historical maps as pivotal reference points. It reliably captured the different shoreline activities (i.e., growing, stable, and receding regions) of natural and anthropogenic origins within the study area, and the findings are consistent with results obtained by using the linear regression rate (LRR) and the end point rate (EPR) methods. This work presents both qualitative and quantitative analyses to evaluate shoreline change using various statistical and mathematical methods. The findings should serve as baseline information for coastal management, notably mitigating coastal hazards-whether of natural or anthropogenic originwhich could result in the enhancement of ecosystem services for coastal communities.

show abstract