Remote sensing means measuring the Earth from a distance, usually from a satellite or an aircraft, without touching the ground. Every map you have seen of vegetation, water, cities or disasters probably started as remotely sensed data, and it is the raw material that most GIS analysis is built on.

The basic idea is simple. The Sun shines on the Earth, surfaces reflect different wavelengths of light, and a sensor records how much of each wavelength comes back. Healthy vegetation reflects a lot of near-infrared light and absorbs red light, water absorbs near-infrared almost completely, and bare soil sits somewhere in between. By comparing the bands, you can tell what covers the ground without ever visiting it.

Sensors come in two kinds. Passive sensors record reflected sunlight or emitted heat, and they need daylight for the optical bands. Active sensors send out their own signal and measure the echo: radar, such as Sentinel-1, images through clouds and at night, and LiDAR measures distance with laser pulses to build elevation models. The choice between them decides what you can see and when you can see it.

The main free missions are worth knowing by name. Landsat, run by the USGS, has imaged the Earth since 1972 at 30-metre resolution and is the longest continuous record we have. Sentinel-2, from Europe's Copernicus programme, gives 10-metre imagery every five days. Sentinel-1 adds radar. MODIS and VIIRS give daily global coverage at coarse resolution, which is ideal for fires, floods and weather. All of them are free to download.

Getting started is easier than it looks. Pick a question, such as how green is this field, find a Sentinel-2 scene in the Copernicus Browser, open it in QGIS, and calculate NDVI, the standard vegetation index. That single workflow teaches you the whole chain: data, bands, index and map. From there you can move to change detection, comparing two dates to find what grew, flooded or burned, and that is where remote sensing earns its keep.