Practical explanations of the methods behind land cover classification, trend detection, and spectral indices — written for people who'll actually use the numbers.
How planetary-scale cloud platforms let anyone analyse decades of satellite imagery without downloading files or writing code.
Sea-surface temperature anomalies and multi-date reflectance change give reef conservation an early, wide-area warning system.
How supervised machine-learning classification turns satellite imagery into thematic maps — and why every map needs an accuracy assessment.
How NDVI is calculated from red and near-infrared reflectance, what the values actually mean, and the three places it commonly breaks down.
Why a slope alone can't tell you if a trend is real, and how a non-parametric significance test separates genuine change from year-to-year noise.
A practical comparison of the three most common classifiers, and why your validation method matters more than which one you pick.
Resolution, revisit time, and historical depth compared — and how to choose between them for a given analysis.
Why Kappa exists alongside overall accuracy, and the legitimate criticisms of relying on it too heavily.
Why a relative comparison, not absolute temperature, is what actually defines an urban heat island.
How zonal statistics summarize raster data within boundaries, and why the mean alone can mislead.
Put these methods to work on your own Google Earth Engine, without code, on your laptop or your phone.
Get the free apps →Deep-dive explainers for every index and analysis — formula, bands, interpretation, limitations and citation.