Heading to higher ground: Uncertainty in elevation data and sea level rise can under predict affected populations

One of the most visible threats to Oceania is sea level rise. Rightly so, as the Pacific Community (SPC) has estimated that the vast majority of people living in Oceania live within 1 kilometer of the coastline.

Screenshot of interactive map created by SPC to visualize populations within 1km of coastline

What can seem as a simple calculation of determining what percentage of a population lives within a certain distance of the coast (or within a certain range of sea level rise) is actually filled with large amounts of uncertainty. Ignoring the difficulty in collecting accurate and current census data, predicting sea level rise and how this interacts with coastal topography and local morphology is a multifaceted undertaking that requires large amounts of data to accurately calculate. Large blanket assumptions are often used to avoid addressing these complex relationships most commonly leading to under predicting the potential threat.

Historical datasets: Their uses and assumptions

In 2000, the United States conducted the Shuttle Resource Topography Mission (SRTM) to measure the elevation of the globe. Prior to 2015, this global elevation dataset (the most widely used and the only of its kind) had a resolution of 90 meters, meaning every 90 square meters was given a single elevation value. In 2015, the United States released an updated global SRTM dataset with a resolution closer to 30 meters.  This was a major improvement for research purposes and is still the go-to source for free global elevation data.

Difference between the 90-meter and 30-meter Shuttle Resource Topography Mission (SRTM) Data

However, numerous studies have been conducted to quantify the error in these elevation datasets, but more importantly is understanding the difference in vulnerability when comparing SRTM data with that of a higher resolution. When compared to 3-meter resolution data in the United States, SRTM data underestimates population exposure by 60%. Some global studies predicts that SRTM data underestimates global exposure by a factor of three. This discrepancy in predictions come from a variety of sources, and it is important to understand because these inconsistencies increase outside of the US, especially in Oceania.

Firstly, coastlines are dynamic features that are manipulated by anthropogenic and natural processes (i.e. dredging, construction, erosion, etc). Secondly, the intertidal zone, or the area that is between the range of high and low tides, is constantly changing making it impossible to identify the location at which land becomes ocean. These two factors compound on each other making it extremely difficult (timely and expensive) to map coastal topography at a higher resolution, especially in remote regions such as Oceania.

Sea level rise forecasts: More than just a water level

Sea level rise estimates for the next century range from 0.5 to 2 meters, depending on the study and emissions scenario. Selecting a height within this range allows for researchers and policymakers to quickly produce maps of the probable future norm indicating areas that will regularly be inundated based on this relative change. However, local and regional sea level rise depends on a variety of factors include local topography/bathymetry, wind speed/direction, ambient temperature, and ocean currents. It is because of this that sea level rise will not be uniform across the globe and it explains why Oceania is more affected by sea level rise compared to other locations.  For example, an island that is in the path of a major ocean current will see higher sea level rise than a place that isn’t as the currents force water further inland. This is the same for tropical areas that receive high winds.

Not only are Oceania countries less equipped to respond to sea level rise compared to industrialized nations (i.e. New York, New York compared to Funafuti, Tuvalu) but they will also experience a greater change in sea level. This is why using a global average to represent sea level rise is a dangerous assumption that can contribute to under representation in the identification of at-risk areas.

Global variability in sea level rise since 1993, showing a greater relative increase in Oceania/South Pacific (NOAA)

Furthermore, sea level rise is not limited to a simple increase in where the ocean is. Studies show that sea level rise exacerbates the effects of storm surges and tides. As sea levels rise, tides and storm surges move further inland, causing more erosive forces in areas that were previously protected and further allows the ocean to encroach inland. The below graphics illustrate this compounding process known as shoreline retreat.

shoreline retreat 1

Mean sea level (MSL) combined with sea level rise (SLR) contribute to tidal encroachment inland: Meaning sea level rise is more than just water levels rising but also physically removing land

Future Directions: Sea-level rise is only one piece of the flooding problem

Identifying vulnerable communities in Oceania in relationship to sea-level rise requires extensive regional and local level knowledge including where people live in relationship to the coast and the local morphology of the coastline. Especially in the case for these island nations, simply stating a change in height is an oversimplification that takes away from the seriousness of their vulnerability over the next couple of decades. For some Oceania countries, sea-level rise will completely cover entire nations and for others it will drastically reduce their already limited land area. On top of that, saltwater intrusion into fresh water sources will further weaken countries’ ability to provide for its citizens.

There have been attempts to improve the elevation and flooding data commonly used by researchers, as in the popular CoastalDEM produced by Climate Central, which uses machine learning to improve the accuracy of SRTM elevation predictions. However, this global model, along with others, still is unable to incorporate coastal defenses (i.e. sea walls or levees), shoreline retreat (as well as other erosion processes), and the effects of compound flooding. Compound flooding is the co-occurrence of high sea-levels with upstream flooding, predominantly occurring in urban areas, which is where 68% of the population in Oceania reside. These models have value for being able to quickly forecast global changes in sea-level at a global scale. Nevertheless, more intensive local and regional examinations are essential to accurately identify communities that are at risk of sea-level rise.

Matthew Preisser is a second year dual degree student at the LBJ School and Cockrell School of Engineering (Master's of Public Affairs and Environmental and Water Resources Engineering). He graduated from Auburn University with a degree in Biosystems Engineering and minors in German and Sustainability in 2018. His research interests include the application of technology and applied sciences to benefit at-risk communities impacted by climate change and extreme weather events. Matthew is currently a National Science Foundation Graduate Research Fellow where he is researching the incorporation of socioeconomic factors into predictive flood modeling. He is also a Brumley Next Generation Fellow through the Robert Strauss Center.

Tagged with: , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *