214
Many of us carry around a remote sensing device in our pocket, a camera. Taking
pictures is the basis of remote sensing and can give us a lot of information, like location, date and time, and what is in the image. Now put a much more sophisticated
camera on a satellite and have it orbit the Earth, and that is what I work with – pictures from space. For the most part, being able to recognize individual objects in the
picture still requires some aspect of visual interpretation, although there have been
lots of advancements in the field of machine learning to tease out specific types of
objects or when something is different between two pictures of the same thing.
These pictures from space can tell us a lot about the world around us. They help
show us what happens after disasters. They can tell us the extent of flooding during
storms, how much the Earth has shifted after an earthquake, and they can help us
estimate the damage to infrastructure. We can also see where urban development is
happening, where forests are being burned, where algal blooms occur, where methane is being released, where extreme rains happen, and the list goes on. The images
are collected through specialized sensors that can measure the “light” and energy
we can and cannot see with the naked eye. Some sensors measure visible light while
others measure ultraviolet light, near-infrared, thermal infrared, radio, and sound
waves. Using these types of sensors on satellite, planes, drones, and ground instruments, we can figure out what, where, when, why, and how the world around us is
changing. However, the observations that we make using these pictures from space
are only models and some models can be better than others. In order to make sure
our models accurately represent the landscape, we need to calibrate and validate
these models with data that we collect on the ground. By visiting sites and making
observations on the ground we can assure ourselves and people making decisions
using this information that the models generated using satellite imagery are accurate.
The following is a set of my own experiences traveling to remote locations
around the world to test and observe how accurate our models generated using satellite images are. By doing this we can have a better idea on what types of coastal
changes are taking place regionally and globally. In particular, I tell the stories of
the where we worked, the remote sensing tools we used, how we go about collecting
data, and why these measurements are so important to you, me, and the communities that live along the coast.
Gabon
Twelve of us, huddled under a concrete awning dressed in our rain gear are trying to
escape the incessant tropical rain. We are at a dock in Libreville, Gabon waiting for
our boat to show up. The group is made up of scientists from the US Forest Service,
NASA’s Goddard Space Flight Center and NASA’s Jet Propulsion Laboratory, and
Duke University. With us is also a group of professors and students from the
Environmental Sciences Department at Omar Bongo University and rangers from
Gabon’s National Park Service. Our plan is to cross the estuary south to Pongara
National Park (Fig. 13.1), where we are expecting to find some of the tallest
D. Lagomasino
Many of us carry around a remote sensing device in our pocket, a camera. Taking
pictures is the basis of remote sensing and can give us a lot of information, like location, date and time, and what is in the image. Now put a much more sophisticated
camera on a satellite and have it orbit the Earth, and that is what I work with – pictures from space. For the most part, being able to recognize individual objects in the
picture still requires some aspect of visual interpretation, although there have been
lots of advancements in the field of machine learning to tease out specific types of
objects or when something is different between two pictures of the same thing.
These pictures from space can tell us a lot about the world around us. They help
show us what happens after disasters. They can tell us the extent of flooding during
storms, how much the Earth has shifted after an earthquake, and they can help us
estimate the damage to infrastructure. We can also see where urban development is
happening, where forests are being burned, where algal blooms occur, where methane is being released, where extreme rains happen, and the list goes on. The images
are collected through specialized sensors that can measure the “light” and energy
we can and cannot see with the naked eye. Some sensors measure visible light while
others measure ultraviolet light, near-infrared, thermal infrared, radio, and sound
waves. Using these types of sensors on satellite, planes, drones, and ground instruments, we can figure out what, where, when, why, and how the world around us is
changing. However, the observations that we make using these pictures from space
are only models and some models can be better than others. In order to make sure
our models accurately represent the landscape, we need to calibrate and validate
these models with data that we collect on the ground. By visiting sites and making
observations on the ground we can assure ourselves and people making decisions
using this information that the models generated using satellite imagery are accurate.
The following is a set of my own experiences traveling to remote locations
around the world to test and observe how accurate our models generated using satellite images are. By doing this we can have a better idea on what types of coastal
changes are taking place regionally and globally. In particular, I tell the stories of
the where we worked, the remote sensing tools we used, how we go about collecting
data, and why these measurements are so important to you, me, and the communities that live along the coast.
Gabon
Twelve of us, huddled under a concrete awning dressed in our rain gear are trying to
escape the incessant tropical rain. We are at a dock in Libreville, Gabon waiting for
our boat to show up. The group is made up of scientists from the US Forest Service,
NASA’s Goddard Space Flight Center and NASA’s Jet Propulsion Laboratory, and
Duke University. With us is also a group of professors and students from the
Environmental Sciences Department at Omar Bongo University and rangers from
Gabon’s National Park Service. Our plan is to cross the estuary south to Pongara
National Park (Fig. 13.1), where we are expecting to find some of the tallest
D. Lagomasino