Home / Methods / Remote sensing

Methods · Remote sensing

Seeing Earth's most common minerals from space, and why river sediment scientists should care

Quartz and feldspar are nearly invisible to the satellites we use most. A thermal-infrared approach is closing that gap, and the payoff for hydrology is a better map of where river sediment actually comes from.

Stylised satellite view of a catchment, with quartz-rich and feldspar-rich source areas shaded and a sediment plume entering a river channel.
Fig. 0   Mapped surface mineralogy across a catchment can be read as a first-order map of potential sediment sources. Illustration.
Ayres River Editorial TeamReviewed by a water-resources engineer
Published 30 June 2026
7 min read · ~800 words
AI-assisted, human-reviewed. Research synthesis and a first draft were produced with AI tooling. The interpretation, the hydrology context, and the conclusions were checked and edited by the editorial team and a qualified reviewer before publication. See cited sources for the underlying material.

A recent advance in spaceborne mineral mapping has a quiet relevance for anyone who works on river sediment. The headline is that quartz and feldspar, the two most abundant minerals in the continental crust, can now be approached globally from orbit. The hydrology question underneath it is older: where does the sediment in a river actually come from?

For most of the satellite era, the answer from imagery was incomplete, because the workhorse sensors are nearly blind to the very minerals that dominate the land surface.

Why quartz and feldspar hide from ordinary satellites

The reflectance bands most remote-sensing instruments use sit in the visible, near-infrared, and shortwave infrared, a region usually written as VNIR-SWIR. Clays, carbonates, and iron oxides show clear absorption features there, so they map well. Quartz and feldspar are effectively featureless across that range. A pixel of clean quartz sand and a pixel of feldspar-rich granite can look almost the same to a SWIR sensor, which is a problem when those two source areas behave very differently once their grains reach a channel.

The signal that does distinguish them lives further out, in the thermal infrared. Silicate minerals carry strong silicon-oxygen (Si-O) bond vibrations that produce emissivity features in the longwave infrared, roughly the 8 to 13 micron band. Thermal-infrared, or TIR, sensing reads those features. The recent work highlighted by AGU is a method that pushes TIR mineral mapping toward consistent, near-global coverage of quartz and feldspar rather than scattered local case studies.

VNIR-SWIR (0.4–2.5 µm) LWIR / TIR (8–13 µm) quartz: flat, no features Si–O feature appears signal
Fig. 1   Schematic. Quartz and feldspar stay flat through the VNIR-SWIR range but show diagnostic Si-O emissivity features in the longwave thermal infrared, which is what makes TIR mapping useful for them.

The instrument that proved it could work

The reference case for spaceborne TIR mineral work is ASTER, the thermal and reflective sensor flown on NASA's Terra satellite. ASTER measured multispectral thermal emission from the surface and remains the main example of an orbital instrument used for broad mineral mapping, including silica-rich terrain. Its limitation is coarse spectral detail in the thermal range, which is precisely what newer methods aim to improve. The direction of travel is from a handful of broad thermal bands toward richer, more separable signatures for the common rock-forming minerals.

The hydrology payoff: tracing sediment sources

This is where the topic turns into a water-resources tool. Suspended sediment is rarely uniform across a catchment. It is mobilised from specific lithologies, soils, and disturbed surfaces, and the mix that arrives at an intake or a reservoir reflects which of those sources are active. Sediment fingerprinting already uses geochemical and mineralogical tracers to attribute river sediment to its sources. A consistent map of surface mineralogy, including the dominant silicates, gives that work a spatially complete starting layer rather than a patchwork of field samples.

Practically, a quartz-rich headwater and a feldspar-rich, weathering-prone slope are different sediment problems. They erode at different rates, deliver different grain sizes, and load the water column differently. Knowing where each sits in the catchment helps explain downstream turbidity signals and helps target where erosion control will actually move the needle. This connects directly to how disturbance, such as the road-building and harvest pressures we covered in forest roads, sediment, and source-water risk, translates into measurable load at a treatment plant.

What it does not replace

Surface mineralogy is a control on sediment supply, not a measurement of sediment in the water. It says nothing on its own about discharge, transport capacity, or whether a source is currently connected to the channel. A bare quartz outcrop with no delivery pathway contributes little, while a small but well-connected gully can dominate the load. The mapping narrows the field of candidate sources; gauging, sampling, and an understanding of how the channel itself reworks material, the kind of process we discussed in meandering versus braided rivers, still decide the answer. Read this advance as a better basemap for sediment work, not a substitute for the fieldwork that grounds it.

For more on how we approach these topics, see about this blog or browse the latest posts.

Cited sources

References & underlying material

  1. Seeing Earth's Most Common Minerals from Space. AGU, Eos editor highlights, 2026. eos.org
  2. Mineral Classification of Land Surface Using Multispectral LWIR and Hyperspectral SWIR Remote-Sensing Data: a case study, Sokolov lignite open-pit mines. Remote Sensing, MDPI. mdpi.com
  3. Satellite-derived mineral mapping and monitoring of weathering, deposition and erosion. Scientific Reports, Nature. nature.com
  4. Mapping quartz, carbonate minerals, and mafic-ultramafic rocks using multispectral thermal infrared ASTER data. Research literature on ASTER TIR mapping. researchgate.net