by Conor Friedersdorf
A reader writes:
One of the things I find fascinating now that I'm part of it, is how unaware people are of the fields of Remote Sensing and Geography. You see photos like NASA's of Pakistan before and after the flood and don't really think about where those come from. It wasn't a happy coincidence that provided those photos, or NASA having to turn on the camera deliberately and flying the Satellite over to take a look at things. The fact is there's a fairly substantial number of satellites whipping around the Earth right now continuously taking pictures of it and cataloging the data. The larger problem for the NASA people would be sorting through all that data to find a pictures relatively free of clouds and hopefully from the same sensor at comparable angles, and figuring out how to adjust it into a normal color image. Basically all the effort involved would be going towards making it look good. I just find this interesting in the context of all the paranoia that you hear being tossed around these days, because we quite literally are watching you.
Some mitigating factors: getting a pixel size of about 3 feet square along the ground is considered pretty high resolution in the scientific field, and the pixels can in fact get a big as a square half a mile to a side. We are really much more interested in precisely measuring the spectrum of light being reflected rather than trying to write computer algorithms that can tell us if we're looking at a tree or a house based off it's shape, or worse, actually having to look at the pictures ourselves, so it's not like we're going to catch you nude sunbathing in your back yard. What we might be able to do is tell you whether you've been watering your yard enough though if we put in the time and effort to actually find your house.
You might wonder why we're doing this. My personal answer, and there may be others in my field who disagree, is that when it comes to any of the sciences involving things on the scale of the Earth, it's impossible really to do experiments in the classical sense. If you think you've found a link for instance between say Valley Fever and unused lots that would kick up a bunch of dust and the associated fungus spores in it, then without us you're limited to either tracking down individual cases, and doing the detective work and digging through files to figure out what the local environment was like before the person came down with it, or waiting until new data comes in and
going into the field to assess the situation first hand. While there is work involved in identifying unused dusty lots from space, being able to classify things well using computer algorithms is an
established if also still growing field. We might not be able to be 100% precise, but compensate by having a much richer catalog of data than is possible in any other way., spanning the globe and 30 years of history. So when people have questions about valley fever, or how land use in a city has changed, or if the local vegetation is changing either in what plants are there or some aspect of their health, or if there's really less water in the lake, or where could we expect fires this year or if things are in fact heating up and where, or even how bad really is the flooding in Pakistan, they have data to try and build the maps they need, and try to find links between these world
For some reason though despite being of use to a number of other fields and answering a lot of questions, we ourselves are sort of a hidden field. If you ask the average person what's going in the
or tell us something about how we got where we are. What is a Geographer going to do? Find that the Nile River is actually 20 feet of the left of where we actually thought it was?
And yet you can make maps of so many things, and gain so much information about the world through them, and asking the right questions. Remote sensing is just a small part of it. I read an
interesting paper the other day of someone trying to make a map of Law Enforcement. After all, you can't arrest everyone for every minor infraction. So where are the police, and what are they doing, and how does this relate to other aspects of the city? Do they mostly patrol safer areas, or high risk areas? If they're in the high risk areas are they making arrests that correspond to the crime statistics in the area, or are they doing the "Broken Window" thing there?