Monitoring, Delineating and Analysing Shorelines using Remote Sensing
- Arpit Shah
- Jun 18, 2024
- 13 min read
Updated: Dec 6
1. INTRODUCTION
Where Land meets the Sea...
The phrase carries a poetic quality—one that evoked, for me, the memory of a pristine beach. Perhaps it did for you as well.
But where, in an exact sense, does land meet the sea?
The physical interface is called the Shoreline and its position and shape remain in continuous flux—shaped by natural forces such as waves, winds and currents, as well as anthropogenic factors including coastal infrastructure, sand mining, and other interventions.
A useful way to categorize the natural drivers is by their temporal behaviour. Tides operate daily and cyclically; a storm surge is sporadic; and marine transgression—for example, sea-level rise from melting glaciers—becomes evident over much longer time horizons. Together, these processes make the coast inherently dynamic.
Below is a basic illustration of the anatomy of a coastal environment:
“Some 37 per cent of the world's population lives within 100 kilometers of the coast, at a population density twice the global average.” - United Nations Environment Programme.
With such high population density—combined with growing threats from Coastal Erosion (loss of sediments to the sea) and Marine Transgression (sea-level rise due to climate change)—regular and accurate monitoring of Shoreline position and its evolution becomes indispensable.
India, with over 7,000 km of coastline (18th longest in the world), conducts extensive research in this space through the National Centre for Coastal Research, Ministry of Earth Sciences.
Below is a key insight from its 1990–2018 Shoreline Change Survey:
33.6% of the Indian coastline is vulnerable to erosion, 26.9% is under accretion, and 39.6% remains stable —Press Information Bureau (December 2023)
Beyond the India-specific data, there is an important conceptual distinction that readers should note:
The terms 'Shoreline' and 'Coastline' are often used interchangeably. Are they one and the same? No, they aren't identical.
It took me a while to figure this out, but here’s what I’ve gathered: although the exact definition may vary from country to country, the word coastline has two distinct interpretations — one geopolitical and the other geological.
The geopolitical coastline refers to the predetermined administrative boundary of a nation or region that lies adjacent to a water body (not to be confused with a country’s maritime boundary). The Indian Ministry’s use of the word “coastline” in the media release above is in this geopolitical sense.
My understanding is that the precise location of this geopolitical coastline corresponds roughly to the boundary of the Littoral Zone — essentially the edge of the Nearshore or the end of the continental plate — marked as the Low-tide breaker line in Figure 1.
The geological interpretation of coastline is distinctly different. Refer to the depiction below:
An easy way to grasp what is being illustrated is this: while the shoreline is where the sea meets the shore, the coastline, geologically speaking, is where the shore meets the land.
If you're wondering how to distinguish shore from land, the difference is fairly straightforward: a shore is composed of sand—either formed from the long-term weathering of rocky structures by waves and winds, or accumulated as sediments deposited by the sea through currents and tides (The difference between these two sediment sources is explained lucidly here).
What should strike you, however, is that in neither interpretation—geopolitical or geological—does the definition of coastline align with that of shoreline. This distinction matters because, although the terms are often used interchangeably in casual conversation, nearly all scientific references I’ve come across that pertain to the precise location of where land meets the sea (technically described as the physical interface of land and water — Dolan et al., 1980) consistently use the term shoreline, not coastline. (Re-read the media release above if you’d like 😉.)
Hence, this post concerns itself with Shoreline Monitoring, Delineation, and Analysis.
HYPERLINKS TO SECTIONS
1. Introduction
2. Scope of this Study
All the seven workflows are demonstrated in this video compilation-
2. SCOPE OF THIS STUDY
As it turns out, there are multiple ways to detect or estimate the position of a shoreline—photographs, drone imagery, beach surveys, vehicle-mounted GPS, remote sensing, and others. Selecting an appropriate method depends on several factors: the desired spatial and temporal resolution, data frequency, accuracy requirements, and the cost of acquiring and processing the data.
In this post, I present video demonstrations of seven satellite-based remote sensing workflows that can be—and are being—used to monitor shorelines and study their evolution through time. These techniques span a wide spectrum: from rudimentary visual monitoring using Google Earth to more advanced extraction methods such as Radar Reflectance Thresholding, NDVI+Tasseled Cap transformation, SAET and EPR4Q. While this list is not exhaustive, it is deliberately diverse in terms of methodology.
For the demonstrations, I have used two major sources of satellite imagery: Landsat (NASA, USA) and Sentinel (ESA, Europe). Both offer spatial and temporal resolutions suitable for Earth Observation and are widely used in research across numerous disciplines.
Beyond knowledge-sharing with enthusiasts, I also hope this work encourages the creation of more open-source geospatial intelligence (OSINT-Geo). Given the scale and complexity of climate-related threats today, it is imperative that we have aware, informed, and trained individuals who can monitor environmental changes continuously and preempt potential disasters.
2.1 Viewing Shoreline Evolution Using Google Earth
Study Area: Someswar - Batapady Beach (South-West India)
This first technique involves using Google Earth - Alphabet Inc.'s satellite imagery rendering platform, to visually monitor shoreline evolution over time through its (optical) Historical Imagery tool (Alphabet also offers a far more advanced imagery-processing environment—Google Earth Engine—though I have not demonstrated it in this post). The built-in Measurement tool allows users to estimate erosion or accretion rates as well.
While elementary, this method provides a quick, macro-level perspective on the state and evolution of a shoreline. For instance, within just a few minutes, I was able to identify which stretches of the Someswar–Batapady coast along the Arabian Sea had been most severely affected by erosion.
TIMESTAMPS
00:00 - Video #.1 Headline
00:04 - 1.1 Exploring the Area of Interest on Google Maps
02:30 - 1.2 Visually monitoring the Shoreline using Google Earth Pro
2.2 Viewing Shoreline Evolution Using EO Browser
Study Area: Marina Beach, Chennai - Tamil Nadu (South-East India)
EO Browser offers several advantages over Google Earth for shoreline-monitoring purposes:
It provides a continuous stream of historical satellite imagery from multiple missions.
It supports on-the-fly image processing in the cloud, with both preset and custom band combinations.
It includes analytical tools such as histograms and time-lapse video generation, which allow for more accurate thresholding and enhanced visualization.
These capabilities enabled me to clearly identify where and when the Adyar river’s outlet into the Bay of Bengal was periodically blocked by accretion from Marina Beach—an important observation, as such blockages can inhibit tidal flushing and exacerbate flooding and water-quality issues.
TIMESTAMPS
00:04 - 2.1 Exploring Area of Interest and Band Combinations on EO Browser
02:48 - 2.2 Creating a Time-Lapse Video of the Area of Interest using EO Browser
2.3 Extracting Shoreline by Thresholding Radar Reflectances
Study Area: Visakhapatnam/Vizag - Andhra Pradesh (South-East India)
While the previous two workflows allowed me to visually monitor the shoreline from rendered satellite imagery and observe its evolution through time, this is the first workflow where I demonstrate the actual extraction and delineation of a shoreline from a downloaded raw satellite imagery dataset. The study area—Visakhapatnam, a bustling coastal tourism hub—hosts several beaches that have experienced significant erosion in recent years, adversely affecting visitor footfall.
Radar Satellite Imagery (SAR) offers a couple of notable advantages over Multispectral Satellite Imagery. Unlike optical sensors, SAR has an active illumination source—a microwave emitter—which enables the satellite to:
capture readings even in zero-light conditions, including night-time, and
penetrate atmospheric constituents such as clouds and aerosols, thanks to its longer wavelengths, allowing for more reliable assessment of surface features from the returned signal (backscatter)
TIMESTAMPS:
00:00 - Video #.3 Headline
00:04 - 3.1 Exploring Area of Interest using Google Maps
03:08 - 3.2 Downloading SAR (Synthetic Aperture Radar) Imagery using Copernicus Browser
08:48 - 3.3 Post-Processing the Raw Radar Imagery using SNAP Software
25:25 - 3.4 Post-Processing the SNAP-processed Radar Imagery using ArcGIS Pro
This method of delineating shorelines comes with a couple of limitations:
the spatial resolution of SAR Imagery is not high enough for precise shoreline extraction, and
An outdated Global Digital Elevation Model (DEM) makes it challenging to reliably isolate water pixels.
Nonetheless, this approach can be paired or cross-validated with methods that use multispectral imagery, which often yields more accurate shoreline delineation.
2.4 Extracting Shoreline by using Water Radiometric Index
Study Area: Anjuna - Goa (South-West India)
Out of the 41 beaches surveyed by NCSCM in the popular beach state of Goa, 21—including my Study Area - Anjuna Beach—were found to be experiencing sand erosion.
The way different components of solar radiation—visible light, infrared and ultraviolet wavelengths of the electromagnetic spectrum—interact with surface features is fundamental to distinguishing one feature from another in Multispectral Remote Sensing. The same principle applies to shoreline delineation. Water exhibits low reflectance in the visible spectrum and zero reflectance in the Near Infrared (NIR), which contrasts sharply with the reflective properties of soil and vegetation (both of which reflect NIR in much larger quantities). Because of this spectral behavior, a water-retrieval index such as the Modified Normalized Difference Water Index (MNDWI) can be used to separate land from water, making it possible to extract shorelines from multispectral imagery.
Since multispectral imagery depends on a passive illumination source—sunlight, it is essential to select cloud-free satellite datasets to ensure accurate shoreline extraction.
TIMESTAMPS
00:00 - Video #.4 Headline
00:04 - 4.1 Exploring Area of Interest and Band Combinations on EO Browser
06:48 - 4.2 Downloading Optical Imagery from Copernicus Browser
10:13 - 4.3 Post-Processing the Raw Optical Imagery using SNAP Software
24:15 - 4.4 Post-Processing the SNAP-processed Optical Imagery on ArcGIS Pro
35:59 - 4.5 Determining Land-Sea Threshold using EO Browser
2.5 Extracting Shoreline by using Vegetation Index + Tasseled Cap transformation technique
Study Area: Anjuna - Goa (South-West India)
Just as the previous workflow identified water features first and then traced the edge where they met non-water features to isolate the shoreline, this workflow applies a similar logic using a contrasting method. Here, a vegetation index—Normalized Difference Vegetation Index (NDVI)—is used to identify non-water features, and the boundary where these features meet water is then extracted as the shoreline.
I have deliberately kept the study area and the imagery acquisition dates the same as in the previous workflow. This allows for a direct comparison between shorelines derived using the Water Index and those derived using the Vegetation Index, enabling meaningful insights. For this workflow, however, I have used a different source of multispectral satellite imagery: Landsat 8.
TIMESTAMPS
00:00 - Video #.5 Headline
00:04 - 5.1 Landsat 8 vs Sentinel-2 Optical Imagery
01:44 - 5.2 Downloading Landsat 8 Imagery from USGS Earth Explorer
07:20 - 5.3 Deploying NDVI+Tasseled Cap technique using Landsat toolbox on ArcGIS Pro
27:28 - 5.4 Visualizing Landsat 8-derived shoreline on Google Earth and comparing it with the Sentinel 2-derived shoreline
2.6 Automatic Extraction of Shoreline using SAET algorithm and performing Change Analysis
Study Area: Satabhaya - Odisha (East India)
If you've seen the previous workflow demonstrations, you'll have noticed how tedious shoreline extraction can become, especially when the coast is irregular in shape. The landscape around a shoreline is constantly shifting—a complex interplay of tides, waves, floods, winds and sediment dynamics—all of which make it difficult to isolate a clean, contiguous line of edge pixels that can reliably represent the shoreline.
The Shoreline Analysis and Extraction tool (SAET) - developed by the European Coastal Flood Awareness System (ECFAS) and released only recently (July 2023), is a welcome advancement in this regard. It automates the search, download and processing of multispectral imagery, and the shoreline it extracts is contiguous across the study area—as it ideally should be. My study area, Satabhaya and nearby regions in Odisha, is a challenging coastal terrain: highly eroded, rugged, and interspersed with flooded patches and tiny water bodies. Yet SAET handled it remarkably well. I was able to process Landsat 8 imagery and derive a clean, continuous shoreline within minutes. While the initial setup of SAET can be a bit involved, using the tool thereafter is remarkably seamless.
This workflow is also the first in which I demonstrate Shoreline Change Analysis. Using QGIS, I computed erosion and accretion rates between two satellite datasets by drawing a landward baseline parallel to the shoreline, dividing it into equally spaced sectors (or hubs), and measuring the average distance between each sector and the corresponding shoreline point.
Have a look at the demonstration below:
TIMESTAMPS
00:00 - Video #.6 Headline
00:04 - 6.1 Getting to know and preparing the SAET tool
10:54 - 6.2 Downloading Landsat 8 Optical Imagery for Area of Interest using SAET tool
16:02 - 6.3 Extracting Shoreline from Downloaded Landsat 8 Imagery using SAET
19:26 - 6.4 Observing the extracted shoreline on ArcGIS Pro
20:41 - 6.5 Using Google Earth Pro to validate the accuracy of the extracted shoreline
23:39 - 6.6 Using the Raw Landsat 8 Imagery itself to validate the accuracy of the extracted shoreline 28:16 - 6.7 Quantitative Analysis of Temporal Shoreline Movement using QGIS
49:51 - 6.8 Symbolizing Results on ArcGIS Pro
2.7 Granular Shoreline Change Analysis and Forecasting using EPR4Q Model
Study Area: Satabhaya - Odisha (East India) and Juhu Beach, Mumbai - Maharashtra (West India)
The End Point Rate Tool for QGIS (EPR4Q) was developed by Dr. Lucas Terres de Lima and his research colleagues at CESAM, University of Aveiro in Portugal. It has been technically validated and compares favorably with DSAS and AMBUR—two widely used, research-grade methods for shoreline change analysis. As Dr. Lima informed me, he did not get the opportunity to continue developing the model after his research tenure, which is why EPR4Q does not function reliably on the latest versions of QGIS. This isn’t a cause for concern, though, as I’ve covered the necessary preparatory steps for running it smoothly in my video demonstration.
While I had already performed quantitative shoreline analysis for Satabhaya using SAET in the previous workflow, I was genuinely impressed by the granularity of EPR4Q’s output over the same area. The tool casts numerous evenly spaced transects across the study region and computes the distance between each transect and the shoreline relative to a baseline—resulting in a highly detailed understanding of shoreline movement.
Like other advanced shoreline change analysis tools, however, EPR4Q struggles with embayed or curved coasts. Fortunately, there are effective workarounds. I demonstrate these in the latter half of the video using Mumbai’s Juhu Beach as the study area—a task that I grappled with for several days before eventually reaching out to Dr. Lima, who graciously shared valuable guidance.
Go give EPR4Q a try!
TIMESTAMPS
00:00 - Video #.7 Headline
00:04 - 7.1 Rewinding the previous six workflows
05:31 - 7.2 Overview of some of the Advanced Tools for Quantitative Analysis of Temporal Shoreline Movement
08:10 - 7.3 Introducing EPR4Q tool for Temporal and Predictive Shoreline Analysis and Prerequisites to run the tool without errors
18:31 - 7.4 Running the EPR4Q model on QGIS and Interpreting the results over the Area of Interest on ArcGIS Pro
36:18 - 7.5 Challenges faced while running EPR4Q tool over Embayed Shorelines and how to address it
3. Concluding Remarks
What eventually evolved into an elaborate video series on Shoreline Detection and Monitoring actually began as a learning endeavor a couple of years ago. I had set out to write this post back then, but paused the assignment due to a lack of meaningful output, an unclear storyline, other projects taking priority, and a host of smaller reasons. Only about four months ago (~February 2024), after weeks of exploring learning resources online, did I finally feel confident enough to weave a compelling narrative on the topic of shoreline monitoring, delineation and analysis.
That said, I do regret not being able to make the erosion/accretion analysis across multiple timelines directly comparable. To achieve that, I would have had to factor in tidal conditions—i.e., select imagery captured during the same or similar tidal phase across different datasets. This is an aspect I did not account for in my demonstrations; instead, I simply selected satellite imagery that was six months or a year apart.
I also realize it is taking me longer (six months!) to produce new content, but I find the output richer and far more in-depth, qualitatively speaking. And it dawned on me only recently that I’ve added YET ANOTHER post related to Water research 😮. This was not intentional—my aim was simply to add a Remote Sensing post since it had been a while since my last OSINT project in this area—but perhaps it was destiny that I ended up interfacing with Water, just as a shoreline inevitably does 😊.
I feel happy and fulfilled that my work is being accessed and appreciated by students, researchers and enthusiasts around the world. I do try to respond to all queries—whether over email or YouTube—as soon as possible. Climate Change and Earth Observation are macro-level Operations problems that are close to my heart, and I would be delighted to collaborate with individuals and institutions working to improve human–nature interactions before it is too late.
ABOUT US - OPERATIONS MAPPING SOLUTIONS FOR ORGANIZATIONS
Intelloc Mapping Services, Kolkata | Mapmyops.com offers a suite of Mapping and Analytics solutions that seamlessly integrate with Operations Planning, Design, and Audit workflows. Our capabilities include — but are not limited to — Drone Services, Location Analytics & GIS Applications, Satellite Imagery Analytics, Supply Chain Network Design, Subsurface Mapping and Wastewater Treatment. Projects are executed pan-India, delivering actionable insights and operational efficiency across sectors.
My firm's services can be split into two categories - Geographic Mapping and Operations Mapping. Our range of offerings are listed in the infographic below-
A majority of our Mapping for Operations-themed workflows (50+) can be accessed from this website's landing page. We respond well to documented queries/requirements. Demonstrations/PoC can be facilitated, on a paid-basis. Looking forward to being of service.
Regards,
Article & Video Credits: RUS Copernicus, NASA USGS, GeographyRealm, Esri ArcGIS Pro, QGIS, ESA SNAP, ECFAS, Dr. Lucas Terres De Lima besides several other individual researchers, organizations and institutes who have contributed towards development of Shoreline monitoring., delineation and change analysis.
