Remote sensing is the process of collecting data without physically interacting with the feature being analyzed. Most remotely sensed data consists of data gathered from satellites or aerial images. Gathering data through physical contact is known as in-situ measurement. Examples of in-situ measurements include gathering water samples and testing soil temperature. The physical contact necessary when collecting in-situ measurements can affect the data collected and widespread in-situ data collection is often too costly and time consuming to be practical for large studies. Remote sensing platforms can collect data across a wide area, lower the manpower necessary, and reduce the need for in-situ measurements.
Brief History of Remote Sensing
The invention of the camera was the first step towards remote sensing. Early photographs were produced using plates of photosensitive materials that are exposed to sunlight for a certain amount of time. Photosensitive materials contain chemicals that react to sunlight. Once the plates are exposed, they are exposed by soaking in chemicals in a darkroom. The ability to capture images using light formed the basis of remote sensing. As the cameras became more portable, people began taking pictures from hot air balloons. During the Civil War (1860-1865), photographs from hot air balloons were sometimes used to view the positions of enemy troops. During World War I, aerial images began being taking from planes. Throughout the World Wars, aerial imaging improved and became more and more vital. With the advent of the space race in the early 1960’s, satellites were produced to take images of the earth from space. Since then, satellite remote sensing has grown in its precision and abilities and satellite gathered information is used for many different military and research applications.
How it works
Remote sensing works by using sensors that perceive different bands of the electromagnetic spectrum to gather information about an object being studied
The electromagnetic spectrum
The electromagnetic spectrum consists of all wavelengths of light, not just the small part of it that humans can see. The visible part of the spectrum that humans can see consists of wavelengths that are between 0.4 and 0.7 micrometers (m). One micrometer is 1 millionth of a meter.  The visible spectrum contains all the colors that the human eye is capable of perceiving; the shortest wavelengths in are violet, and the longest wavelengths are red. As the wavelengths become shorter or longer than the visible spectrum range, the human eye cannot see them, but instruments have been developed that can see and record these wavelengths. During World War II, aerial infrared photographs were used to differentiate between vegetation and camouflage nets designed to look like vegetation.  Infrared is a range of wavelengths longer than red portion of the spectrum and is often used to identify vegetation because healthy plants have strong reflection in the infrared.
When light hits an object, some of that light is reflected, while some of it is absorbed by an object. The human eye perceives color because is sees the reflected wavelengths. More than one wavelength may be reflected, but usually one section of wavelength is more strongly reflected than another. For example, plants are green because that is the part of the visible spectrum most strongly reflected by the chlorophyll in the plants. Other wavelengths, such as blue or red, are still somewhat reflected, but to a lesser extent. However, although the green wavelengths are the portion of the visible spectrum with the strongest reflectance, infrared wavelengths have even stronger reflectance than green wavelengths, but because the human eye cannot see infrared, plants are seen as green.  Fortunately for scientists, remote sensing platforms are capable of detecting infrared reflectance, as well as reflected wavelengths from other portions of the spectrum outside the visible spectrum.
The different portions of wavelengths are often referred to as spectral bands. Ultraviolet, Infrared, Red, Green, and Blue are all types of spectral bands. Remote sensing instruments are designed to detect a different set of bands depending upon the purpose of the instrument. For example, the Landsat 8 satellite can detect 11 different spectral bands. Each of these bands helps identify certain features: Band 1 is good for coastal and aerosol studies, Band 2 (blue) helps with bathymetric mapping and distinguishing between different types of vegetation, and Band 5 (Near-Infrared) helps with mapping biomass content and shorelines.  Although most remote sensing instruments can only sense a dozen or so bands, certain instruments can detect hundreds of bands. This is known as hyperspectral imaging, and the data from a hyperspectral imager requires enormous data storage capacity as well as specialized software to analyze it.
Because the level of reflectance per spectral band varies from object to object, remote sensing analysts use spectral signatures to identify objects. By looking at the patterns of reflectance across the spectral bands, scientists can determine the object being observed and what it was made from. For example: vegetation has high reflectance in the green band, low red reflectance, and very high infrared reflectance; concrete has high reflectance in all bands; and water has medium reflectance in the blue band and almost no reflectance in the infrared. These patterns of reflectance are unique to different materials, so scientists are able to accurately identify materials and land features, even from space. This is why having satellites with more spectral bands is helpful, because the greater number of bands allows for greater accuracy in identification. 
Restrictions in Remote Sensing
Although remote sensing is extremely useful for many different applications, it does have some limitations. The primary limitations revolve around the resolutions of remote sensing instruments and the need for atmospheric correction.
There are three primary types of resolution in remote sensing: spatial resolution, temporal resolution, and spectral resolution.
- Spatial Resolution refers to the smallest area covered by one pixel in a remotely sensed image. For example, a satellite with a spatial resolution of 30 meters can see objects and areas greater than 30 meters. Below 30 meters, the data is one big blur, with all the wavelengths in that are averaged together.
- Temporal resolution refers to how long it takes for a remote sensing instrument to re-visit a point. Aerial images (from airplanes) have the shortest temporal resolution; they are able to return to a point within minutes or hours. Satellites have the longest temporal resolution, depending upon their orbit and spatial resolution. Satellites with larger (i.e. less precise) spatial resolutions gather data from large areas at once, covering large swaths of the earth’s surface. This allows them to complete their orbit and return to the same point more quickly than a satellite with a smaller spatial resolution. Satellites with smaller spatial resolutions can only detect a small swath of the earth’s surface at any given time, which makes it take longer to orbit the earth. Therefore, as the spatial resolution increases, the temporal resolution decreases.
- Spectral resolution refers to the number of spectral bands an instrument can sense. High resolution instrument have more bands while low resolution instruments have fewer bands. Hyperspectral imaging has the highest resolution, with the hundreds of spectral bands. 
The atmosphere is full of particles and gasses which interfere with the electromagnetic waves, causing them to bend and scatter somewhat as they pass through the atmosphere. Weather systems also cause problems because the light cannot travel through the clouds very easily. These problems require scientists to correct the data they gather for atmospheric disturbances. They use different algorithms, depending upon what kind of interference they are correcting for, in order to ensure the accuracy of the data.
- Graham, S. (1999, September 17). Remote Sensing : Feature Articles [Text.Article]. Retrieved February 28, 2016, from http://earthobservatory.nasa.gov/Features/RemoteSensing/
- ESA - Eduspace EN - Home - History of Earth observation. (2009, November 26). Retrieved March 8, 2016, from http://www.esa.int/SPECIALS/Eduspace_EN/SEM1NP3Z2OF_0.html
- ESA - Eduspace EN - Home - Introduction. (2010, June 11). Retrieved February 28, 2016, from http://www.esa.int/SPECIALS/Eduspace_EN/SEM7IQ3Z2OF_0.html
- What are the best spectral bands to use for my study? (n.d.). Retrieved March 8, 2016, from http://landsat.usgs.gov/best_spectral_bands_to_use.php
- ESA - Eduspace EN - Home - Spectral signatures. (2009, November 26). Retrieved March 8, 2016, from http://www.esa.int/SPECIALS/Eduspace_EN/SEMPNQ3Z2OF_0.html
- Satellite Applications for Geoscience Education. (n.d.). Retrieved March 10, 2016, from https://cimss.ssec.wisc.edu/sage/remote_sensing/lesson3/concepts.html