GeospatialGeospatial

Remote Sensing for Beginners: What You Can Learn from Space

Introduction to remote sensing — satellite types, sensors, what you can measure from space, and how to get started with free data from Sentinel and Landsat.

Ryan Bethencourt
April 24, 2026
10 min read

What Is Remote Sensing?

Remote sensing is the practice of observing Earth's surface from a distance – typically from satellites orbiting hundreds of kilometers above the ground. Rather than walking through a forest to count trees, a satellite captures an image of the entire forest in a single pass. Rather than driving through a city to assess flood damage, a satellite images the whole city in seconds. Remote sensing transforms local, labor-intensive observations into global, automated measurements.

The technology rests on a simple physical principle: everything on Earth reflects or emits electromagnetic radiation. Sunlight hits a forest canopy, and the leaves reflect certain wavelengths while absorbing others. A satellite sensor captures this reflected light and records it as a digital image. Because different materials (vegetation, water, soil, concrete) reflect light differently, scientists can determine what covers the surface by analyzing the spectral signature of each pixel.

Note
The term "remote sensing" was coined in 1958 by Evelyn Pruitt of the U.S. Office of Naval Research. The first civilian Earth observation satellite, Landsat 1, launched in 1972. Today, over 1,000 Earth observation satellites orbit the planet, generating petabytes of data annually.

Types of Satellites and Sensors

Optical Sensors

Optical sensors work like a camera: they capture reflected sunlight in visible and near-infrared wavelengths. The resulting images look like aerial photographs (in the visible bands) but also contain information invisible to the human eye (in the infrared bands). Sentinel-2, the workhorse of modern optical remote sensing, captures 13 spectral bands ranging from blue visible light (443 nm) to shortwave infrared (2190 nm) at resolutions between 10 and 60 meters.

The main limitation of optical sensors is clouds. When clouds cover the surface, the sensor sees the cloud tops, not the ground. In tropical regions, cloud coverage can exceed 70% on any given day, making it challenging to get clear images. Temporal compositing – combining multiple partially cloudy images to produce a cloud-free composite – is the standard solution.

Radar Sensors (SAR)

Synthetic Aperture Radar (SAR) sensors transmit microwave pulses toward the ground and measure the backscattered signal. Unlike optical sensors, SAR works day and night and sees through clouds, smoke, and light rain. This makes SAR invaluable for monitoring tropical forests, detecting floods during storms, and imaging polar regions during winter darkness.

SAR images look fundamentally different from photographs. Bright areas indicate rough surfaces (like urban areas or choppy water) that scatter the radar signal back toward the satellite. Dark areas indicate smooth surfaces (like calm water or flat pavement) that reflect the signal away. Interpreting SAR data requires understanding these backscatter physics, but the information it provides is complementary to optical data and essential for many applications.

Thermal Sensors

Thermal sensors measure heat radiation emitted by the surface. They detect land surface temperature, which reveals urban heat islands, volcanic activity, wildfire hotspots, and industrial heat sources. Landsat carries a thermal band at 100-meter resolution, and MODIS provides daily thermal observations at 1-kilometer resolution. Thermal data is critical for urban planning (identifying neighborhoods that overheat), agriculture (detecting crop water stress), and fire monitoring.

What You Can Measure from Space

Vegetation Health

Healthy vegetation reflects strongly in the near-infrared and absorbs visible red light for photosynthesis. The Normalized Difference Vegetation Index (NDVI) captures this relationship in a single number between -1 and 1. High NDVI (above 0.5) indicates dense, healthy vegetation. Low NDVI (below 0.2) indicates bare soil, water, or urban surfaces. Tracking NDVI over time reveals crop growth stages, drought stress, seasonal patterns, and deforestation events. NDVI is the single most widely used metric in remote sensing.

Water Quality and Extent

Satellites can map surface water extent (lakes, rivers, floods, reservoirs) by exploiting the strong absorption of near-infrared light by water. More advanced analysis using visible bands estimates water quality parameters like turbidity, chlorophyll-a concentration (indicating algal blooms), and dissolved organic matter. These measurements enable continental-scale water resource monitoring that would be impossible with ground-based sampling alone.

Urban Heat Islands

Cities are typically 1-3 degrees Celsius warmer than surrounding rural areas due to concrete, asphalt, reduced vegetation, and waste heat from buildings and vehicles. Thermal satellite data maps these urban heat islands at high spatial resolution, revealing which neighborhoods are hottest and where green infrastructure (parks, street trees, green roofs) could provide the most cooling benefit. This information is increasingly critical as cities adapt to climate change.

Crop Health and Agricultural Monitoring

Agricultural remote sensing is one of the most commercially valuable applications. Satellites monitor crop type, growth stage, health, water stress, and predicted yield across millions of hectares. Vegetation indices track crop development through the growing season. Multitemporal analysis distinguishes crop types based on their growth calendars. Anomaly detection flags fields experiencing stress weeks before visual symptoms appear on the ground, giving farmers time to intervene.

Tip
If you are interested in agricultural monitoring, start with Sentinel-2 NDVI time series. Download a year of NDVI data for a farming region and plot it over time. You will clearly see planting dates, peak growth, and harvest timing – the fundamental rhythm of agriculture visible from space.

Getting Started: Free Data Sources

The barrier to entry for remote sensing has never been lower. Here are the primary free data sources and how to access them:

Sentinel-2

The European Space Agency's Sentinel-2 mission provides global optical imagery at 10-meter resolution with a 5-day revisit cycle. Data is freely accessible through the Copernicus Open Access Hub, Google Earth Engine, and cloud providers like AWS and Microsoft Planetary Computer. Sentinel-2 is the best starting point for most remote sensing projects due to its combination of resolution, revisit frequency, and data quality.

Landsat

Landsat is the longest-running Earth observation program, providing continuous global coverage since 1972. The current generation (Landsat 8 and 9) offers 30-meter resolution with a 16-day revisit cycle. The historical archive is invaluable for studying long-term change – deforestation trends, urban growth, glacier retreat, and coastline change over five decades. Data is freely available through USGS EarthExplorer and Google Earth Engine.

MODIS

The Moderate Resolution Imaging Spectroradiometer flies on NASA's Terra and Aqua satellites, providing daily global coverage at 250-meter to 1-kilometer resolution. While the resolution is too coarse for detailed mapping, the daily revisit makes MODIS ideal for tracking rapid changes like fire spread, flood extent, and snow coverage. MODIS data products (NDVI, land surface temperature, fire detections) are pre-processed and freely available through NASA's LAADS DAAC.

Using APIs vs. Downloading Data

Traditional remote sensing workflows involve downloading large image files (often several gigabytes per scene), installing specialized software, preprocessing the data, and running analysis locally. This works well for researchers who need fine-grained control, but creates a steep learning curve for developers and analysts who want quick answers.

API-based access inverts this workflow. Instead of downloading data and running analysis locally, you send a request specifying what you want (location, date, analysis type) and receive results back as structured data. Here is what that looks like in practice:

Get vegetation health for a region via API
import requests

API_KEY = "sk-sci-your-api-key"
BASE = "https://api.scirouter.ai/v1"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}

# Get NDVI (vegetation health) for a farming region
response = requests.post(f"{BASE}/geospatial/ndvi",
    headers=HEADERS,
    json={
        "latitude": 42.03,
        "longitude": -93.63,
        "radius_km": 5,
        "source": "sentinel-2",
        "date_range": ["2026-07-01", "2026-07-31"]
    })

result = response.json()
print(f"Mean NDVI: {result['mean_ndvi']:.3f}")
print(f"Healthy vegetation (>0.5): {result['healthy_pct']:.1f}%")
print(f"Stressed vegetation (0.2-0.5): {result['stressed_pct']:.1f}%")
print(f"Non-vegetated (<0.2): {result['bare_pct']:.1f}%")
Output
Mean NDVI: 0.687
Healthy vegetation (>0.5): 71.2%
Stressed vegetation (0.2-0.5): 18.4%
Non-vegetated (<0.2): 10.4%

The API handles data retrieval, cloud masking, atmospheric correction, and analysis. You get structured results without managing any geospatial data infrastructure. For developers building applications that need satellite intelligence – crop monitoring dashboards, environmental compliance tools, real estate analytics – APIs are the fastest path from concept to production.

Your First Remote Sensing Project

The best way to learn remote sensing is to start with a question about a place you know. Pick a familiar location – your hometown, a vacation destination, a region in the news – and observe it from space. Here is a suggested learning path:

  • Week 1: Explore your region visually using the Satellite Analyzer Studio or EO Browser (free, web-based). Toggle between true color and NDVI views. Notice how vegetation, water, and urban areas appear differently.
  • Week 2: Compare two dates – summer and winter, before and after a storm, or pre- and post-construction. Observe how the landscape changes and what signals those changes produce in the imagery.
  • Week 3: Try programmatic access. Use SciRouter's API or Google Earth Engine to extract NDVI time series for a crop field and plot the seasonal growth curve.
  • Week 4: Build something useful. Create a simple dashboard that monitors vegetation health in a region you care about, using API calls to fetch updated satellite data weekly.

Next Steps

Now that you understand the fundamentals, dive deeper into specific applications:

Remote sensing puts the entire planet at your fingertips. Whether you are a farmer monitoring crops, a researcher studying climate change, or a developer building the next geospatial application, the data is free, the tools are accessible, and the insights are waiting. Get a free API key and start exploring Earth from space.

Frequently Asked Questions

What is remote sensing in simple terms?

Remote sensing is the science of gathering information about something without physically touching it. In practice, it almost always refers to using satellites or aircraft to observe Earth's surface from above. Satellites carry sensors that measure light (or radar signals) reflected or emitted by the ground, water, and atmosphere. By analyzing these measurements, scientists can determine what is on the surface — forests, cities, water, crops — and monitor how it changes over time. Think of it as taking a very detailed, very high-altitude photograph, except that many sensors see wavelengths of light that human eyes cannot.

Is satellite data really free?

Yes, a substantial amount of satellite data is freely and openly available. The European Space Agency's Sentinel constellation (Sentinel-1 radar, Sentinel-2 optical) and NASA/USGS Landsat missions provide free global coverage with frequent revisits. MODIS data from the Terra and Aqua satellites is also free. These datasets are sufficient for most research, environmental monitoring, and agricultural applications. Higher-resolution commercial imagery (sub-meter from Maxar, daily 3-meter from Planet) requires paid access, though many providers offer free tiers for academic and non-commercial use.

What can I actually measure from satellite imagery?

The list is surprisingly long. From satellite imagery you can measure vegetation health and density (using NDVI and other vegetation indices), surface water extent and quality, urban footprint and building density, land surface temperature, snow and ice coverage, soil moisture, crop type and growth stage, air pollution (NO2, aerosols), ocean color and chlorophyll concentration, elevation and terrain (from radar interferometry), and ground displacement (subsidence, landslides). The specific measurements depend on which satellite and sensor you use — different sensors are designed for different purposes.

Do I need programming skills to work with satellite data?

Not necessarily, but programming greatly expands what you can do. For visual exploration and simple analysis, platforms like Google Earth Engine (JavaScript or Python), EO Browser (web-based, no coding), and QGIS (desktop application with a graphical interface) provide point-and-click access to satellite data. For automated analysis, batch processing, and integration into applications, Python is the standard language, with libraries like rasterio, geopandas, and xarray for data handling. API services like SciRouter let you access satellite analysis results with simple HTTP requests from any programming language.

What is the difference between Sentinel and Landsat?

Sentinel and Landsat are both free, global satellite constellations, but they differ in resolution, revisit frequency, and heritage. Landsat (NASA/USGS) has been operating since 1972, providing the longest continuous record of Earth observation at 30-meter resolution with a 16-day revisit cycle. Sentinel-2 (ESA) launched in 2015 and provides 10-meter resolution with a 5-day revisit cycle. For most new projects, Sentinel-2 is preferred due to its higher resolution and more frequent imaging. For historical analysis going back decades, Landsat is essential. Many workflows combine both to maximize temporal coverage.

Try this yourself

500 free credits. No credit card required.