I've read with interest the discussion on remote sensing and the urban
environment. A couple of comments from the operational side.
1. Windscreen surveys ("lateral remote sensing"?) are often used to collect
information about physcial vulnerability and this information then used to
define social vulnerability. At times, direct contact with the resident
population is not possible, or advised. The issue of correlating what you
see with what it means is not just one of distance, although collecting
video at 10 meters can be more risky than collecting it at 10 km.
2. Contact with people is critical to accurately assess vulnerability.
Although probably any intro remote sensing course contains a section on
ground truthing, many remote sensing efforts I have come across don't pay
much attention to this basic rule.
3. Creating the correlations between remotely sensed and ground data is
difficult and often victim of an overstatng of what rs can tell you. You
almost need a doubting Thomas in every project to seriously question the
results to keep the project managers honest. Of course, this is where a PhD
is different.
4. As in any data collection, the process and analysis should be driven by
what you need to know, not what is available as information. To reuse the
tired phrase, the process should driven by demand not supply. One may need
to spend more time with the users (client) to clearly understand what they
*need*, not what they think they want.
5. Any vulnerability assessment should tie back to those being assessed.
Why not have the population being assessed look at the pictures and decide
their own vulnerabilities? Sure, the experts can have an input, but this
approach(for which there is practical experience) makes remote sensing an
inclusive, rather than an extractive, process.
Hope these comments are of interest.
C. Kelly
([log in to unmask])
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|