The investigation focused on sediments from headwater streams, the small upstream channels that feed larger rivers. These habitats are especially important because their shallow waters and biologically active sediments are where terrestrial material first enters a river network. Bacteria living among sediment grains break down organic matter, transform nitrogen compounds and influence the chemical conditions experienced downstream. To determine how these communities respond to environmental pressures, Lucía Cabello-Alemán and colleagues examined samples collected across three distinct biogeoclimatic regions of Spain: Mediterranean lowlands, the Sierra Nevada and the Cantabrian Mountains. Within that broad gradient, the researchers compared four land-use categories, allowing them to distinguish the effects of climate and landscape setting from those associated with human activity.

The team used 16S ribosomal RNA metabarcoding to identify bacteria in the sediment. The 16S rRNA gene is present in bacteria and contains regions that vary among evolutionary lineages, making it a standard molecular barcode for profiling microbial communities. In practice, researchers extract environmental DNA, amplify selected sections of this gene and sequence the resulting fragments. Computational comparisons with reference databases then provide an inventory of the bacterial groups present in each sample. This approach does not necessarily identify every organism to species level, nor does it measure whether a detected microbe is actively growing, but it is powerful for revealing community composition and comparing biodiversity across many sites.

The analysis showed pronounced regional differentiation. Bacterial communities from the Mediterranean lowlands, Sierra Nevada and Cantabrian Mountains differed substantially from one another, with strong taxonomic turnover indicating that many bacterial lineages were replaced as environmental conditions changed. The researchers also observed distinct patterns of alpha diversity, a measure describing the variety of bacterial taxa within individual samples. In contrast, differences between land-use types were weaker and less consistent across regions. Land use did appear to alter particular components of the microbial community, but those effects were comparatively subtle and lacked the same broad spatial coherence as the regional signal. In other words, the identity and diversity of sediment bacteria were structured primarily by the environmental context of the region rather than by a single local category of human land use.

That finding does not mean agriculture, urban development or other human pressures are biologically irrelevant. Instead, it suggests that their effects are nested inside a larger environmental hierarchy. Temperature, precipitation, hydrology, geology, vegetation and the chemistry of material entering a stream can all vary among biogeoclimatic regions. These factors influence the amount and type of organic carbon available to microbes, the oxygen conditions within sediment pores, water residence time and the supply of nitrogen and other nutrients. They can also determine whether a stream experiences persistent flow, seasonal drying or intense pulses of runoff. Such variables act together to create regional habitat templates, potentially overwhelming or masking more localized land-use effects when microbial communities are compared across a large geographic area.

The researchers next examined what the bacterial communities might be able to do, rather than only which taxa they contained. They used functional inference, a computational method that predicts the presence of metabolic genes from taxonomic profiles and reference genomes. The method can suggest whether a community has the genetic potential to perform processes such as nitrification, denitrification, nitrogen fixation or dissimilatory nitrate reduction to ammonium, commonly called DNRA. These predictions are not direct measurements of gene activity or chemical fluxes: a predicted gene may not be expressed under the conditions present in a stream. Nevertheless, functional inference can provide a broad map of the biochemical capabilities associated with microbial community composition and generate testable hypotheses for future work using metagenomics, transcriptomics or direct process measurements.

The predicted nitrogen-cycling profiles partly mirrored the taxonomic differences among regions. Mediterranean lowland sediments were enriched in genes associated with predicted nitrification and denitrification. Nitrification is the aerobic conversion of ammonium into nitrite and nitrate, while denitrification generally occurs under oxygen-limited conditions and reduces nitrate through gaseous intermediates, ultimately returning nitrogen to the atmosphere as nitrogen gas. These processes can remove biologically available nitrogen from water, although incomplete denitrification can also produce nitrous oxide, a potent greenhouse gas. Sediments from the Sierra Nevada showed stronger predicted potential for nitrogen fixation and DNRA. Nitrogen fixation converts atmospheric nitrogen into biologically usable forms, whereas DNRA reduces nitrate to ammonium, retaining nitrogen in a form that can remain available to organisms. The Cantabrian Mountains displayed a more even functional profile, without the same pronounced enrichment of one set of predicted pathways.

The correspondence between bacterial identity and predicted function was clearest at the regional scale and less obvious at the local scale. This distinction is important because different bacterial groups can sometimes perform similar biochemical tasks, a phenomenon known as functional redundancy. Conversely, closely related organisms may possess different genes and respond differently to changing conditions. As a result, a shift in taxonomic composition does not automatically translate into a proportional change in ecosystem functioning. In these headwater sediments, broad regional differences were sufficiently strong to be reflected in inferred nitrogen-cycling potential, but land-use effects were more modest and appeared to affect selected community components rather than reorganize the entire functional system. The study therefore supports a hierarchical view in which large-scale environmental conditions establish the main structure of microbial communities, while local disturbance modifies that structure.

One of the most striking findings was the presence of a consistent core microbiome across contrasting samples. Despite the regional turnover and differences in predicted metabolic potential, some bacterial taxa were shared among streams exposed to different environmental and land-use conditions. A core community may represent lineages capable of tolerating a wide range of freshwater sediment environments, or it may reflect fundamental biochemical requirements common to headwater ecosystems, such as the breakdown of organic matter and the transformation of nitrogen compounds. These widespread taxa could become useful indicators in biomonitoring, particularly if future research determines whether their abundance or activity changes predictably under pollution, warming, drought or nutrient enrichment. The findings also caution against interpreting every difference between samples as evidence of ecosystem collapse: a community can undergo substantial turnover while retaining a stable functional or taxonomic foundation.

The authors argue that freshwater monitoring should incorporate regional baselines instead of applying one universal expectation to every stream. A bacterial profile that signals disturbance in a Mediterranean lowland may be normal in a mountain watershed, while a nitrogen-cycling pattern characteristic of the Sierra Nevada may not be appropriate as a benchmark for the Cantabrian Mountains. Combining taxonomic surveys with functional approaches could make assessments more informative, especially as climate change alters temperature, rainfall and stream flow across the Iberian Peninsula. At the same time, the researchers emphasize the limits of inference: predicted genes must ultimately be validated with measurements of gene expression, enzyme activity and actual nitrogen transformations. Even with that caveat, the study provides a continent-spanning microbial perspective on stream health and shows why the invisible biology of river sediments cannot be separated from the climate and geography of the landscapes above them.