This forum is shutting down! Please post new discussions at community.naturalcapitalproject.org

Coastal Vulnerability - Should Sea Level Rise Impact on the Population?

Hello,

I have run the coastal vulnerability model twice for the same area with a sea level rise shape file included and excluded.  Due to uneven subsidence across the city there is quite a bit of spatial variation with respect to projected sea level change.  While the vulnerability index changes between scenarios, the population affected remains exactly the same.   Is this to be expected or I should I re-examine my model set-up?

I saw there was a similar debate about this in March, but it didn't seem that a conclusion was reached.

Thanks,

Divindy

Comments

  • Hi Divindy, 

    Great question.  'Population affected' would be a misnomer for what the population output is currently mapping.  What the population raster shows you is the number of people living within a user-defined radius of each shoreline segment that the model computes the hazard index for.  So, it's a very simple map summarizing numbers of people and will not change with scenarios (unless you were to change the population raster).  In general people link the hazard index (coastal exposure index) to the social component after the fact, often outside of InVEST.  

    For instance, from your sea level rise scenarios, you might calculate a distribution of high, medium and low vulnerability using the full distribution of coastal exposure values from those runs.  And then you might ask how many people are living along highly exposed shoreline (the top quantile of exposure indices) currently, and then how does that change in the future with SLR?  You have to make that link between shoreline exposed to coastal hazards and people, infrastructure, property values, etc.  

    A very well documented example of this workflow is presented in the Arkema et al. 2013 Nature Climate Change paper (see main paper and supplemental).  If you have further questions I'm happy to answer.

    Also, I would add that if you have more local data, census data or even a recent Landscan population layer that would be an improvement over the population layer included with the InVEST base data.

    Thanks,

    Jess
  • Hi Jess,

    Thanks for clarifying that and for the brilliant suggestion to follow the methodology in the Arkema paper! I had been thinking about doing something similar, but that now gives me something credible to cite as a methodology.  You have made my day!

    Do you know if there that in a similar to manner to how InVEST will calculate the population within the coastal boundary - if there is a way to calculate the quantity of other economic assets - e.g. from the local disaster department I have data on the economic cost if roads, fish ponds, rice farms, houses etc are damaged?  I have separate shape files for each of these layers for a current scenario and projected in the future. 

    Thanks!

    Divindy


  • Hi Jess,

    Sorry, one more question.  Am I right in my reading of the Arkema methodology that the impacts were on a binary scale (e.g if they were above the baseline upper quartile level they were counted as 100% damage or population affected) or was any sort of gradient considered?

    Thanks,

    Divindy

  • Jess_SilverJess_Silver Member
    edited August 2016
    Regarding your first question, you should be able to develop a simple methodology to link anything you'd like to the hazard index.  The main question you need to decide is how far inland do I consider something 'vulnerable' to a coastal hazard.  For instance, you might say in your analysis that you considered everything that was 1km from the shore as potentially vulnerable to coastal hazards.  If you had a great DEM you could do something refined by elevation.  

    Then you need to develop a (simple) workflow in your GIS to estimated the road length, or area of fish ponds/rice farms, etc. that may be highly vulnerable to hazards.  And then you can link this to economic values (I assume the damage values are by area or road length, or something like that).

    The Arkema did the valuation by assuming 100% of the population at risk would be affected, yes.  But, people link the social and exposure risk in more nuanced ways in lots of other studies.  Look for example at some recent work we did in collaboration with some folks in Maryland, USA (The Nature Conservancy, 2016. Maryland Coastal Resiliency Assessment. M.R. Canick, N. Carlozo and D. Foster. Bethesda, MD).  Other papers that might be interesting to you:  Economic Vulnerability to Sea-Level Rise along the Northern U.S. Gulf Coast, Thatcher et al. 2013, Assessment of coastal protection as an ecosystem service in Europe, Liquete et al. 2013.  Anyway, there are many analyses that combine social vulnerability with exposure in the coastal resilience/vulnerability literature ... what InVEST brings specifically is an assessment of where coastal and marine habitats might be playing critical roles in providing natural defenses.

    Hope that's helpful,

    J.
    Post edited by Jess_Silver on
  • Thanks Jess - I think I am there with things now (hopefully!).

    Best

    Divindy

Sign In or Register to comment.