Ground Truthing-Validation of Airborne or Satellite Remote Sensing Data

Many remote sensing studies employ the use of satellite data or aircraft collected spectral data. These datasets offer robust analysis on wide swaths of area. Remote sensing data has been used in a variety of applications for studies in climate change, precision agriculture, algae bloom monitoring, vegetation, soil, land use change, water quality assessment and many more. These types of studies almost always include some amount of in situ data collection for validation.

The field of remote sensing is incredibly diverse, and the sensors being used on satellites, drones, or otherwise often have different spatial and spectral resolutions. Some may be multi-spectral sensors which means that they have several bands targeting different parts of the spectrum. These bands are limited to their band width and may have data gaps between them. Some may be limited spatially measuring wide areas of land making each pixel’s spectra more complex. These limitations must be addressed in order to fully understand and interpret the data. That is why a high accuracy hyperspectral sensor is critical to ground truth/validate the readings.

https://landsat.gsfc.nasa.gov/satellites/landsat-8/

Spectral band resolution of Sentinel 2. Landsat 7, and Landsat 8. https://www.usgs.gov/media/images/comparison-landsat-7-and-8-bands-sentinel-2

Ground truthing is the act of comparing remote sensing data to what is known to be true on the ground. In order to conduct hyperspectral ground truthing it is necessary to use a highly accurate hyperspectral spectroradiometer to compare to the coarser remote sensing data. This is a common practice in remote sensing studies and gives the researcher the ability to calibrate their models or simply understand any assumptions being made with the coarser remote sensing data. Ground validation commonly includes a collection of spectral signatures from things like leaves, soil, or other vegetation using a leaf clip or contact probe. These direct spectral signatures can be used to train classification libraries and models. Validation will also commonly include traditional field samples from water or soil. Ground validation gives the researcher confidence in their analysis and a better understanding of the data that they are working with.

A hyperspectral radiometer like the Naturaspec™ is ideal to ground truth almost any satellite or drone. The Naturaspec is designed to be a light, efficient, and incredibly accurate making it an excellent fit for almost any kind of remote sensing application in the field or in the lab. Spectral evolution provides instruments that can validate all remote sensing satellites or drones targeting UV/Vis/NIR. Spectral Evolution’s focus on providing the highest hyperspectral resolution combined with the lowest signal to noise ratio from 350-2500nm makes our sensors the best choice for ground validation in your next study.