Environmental sciences articles from across Nature Portfolio

The degradation of agricultural land poses a vital threat to productivity and is, in some cases, hard to reverse. Curbing this trend is essential to avoid cascading impacts on natural ecosystems, biodiversity and human health.

Satellite images have been analysed by a machine-learning algorithm to produce high-resolution maps of wetland environments — a boon for conservation efforts.

Antimicrobial resistance risk has long been framed around the fate of parent antibiotics in the environment. Community-level selection assays show that transformation products can remain bioactive and sustain antimicrobial resistance selection.

SPA-DEC is a dynamic carbon calculator that translates a facility’s metered energy and water use, together with waste and other operational sources, into the footprint of a single spa or wellness visit. Across two case-study facilities in 2024 it found typical visits of about 3.5 to 5 kilograms carbon dioxide equivalent, with on-site renewables reducing per-visit emissions by roughly 14 to 33%

Considering international trade in methane emissions accounting substantially reshapes the geographical distribution of methane emissions from rice cultivation, according to a global assessment of trade-adjusted emissions

The degradation of agricultural land poses a vital threat to productivity and is, in some cases, hard to reverse. Curbing this trend is essential to avoid cascading impacts on natural ecosystems, biodiversity and human health.

Wei Chen explains how differential pulse voltammetry is used to resolve the electron transfer processes underlying microbe–mineral element cycles in soils, sediments and aquifers.

Satellite images have been analysed by a machine-learning algorithm to produce high-resolution maps of wetland environments — a boon for conservation efforts.

Antimicrobial resistance risk has long been framed around the fate of parent antibiotics in the environment. Community-level selection assays show that transformation products can remain bioactive and sustain antimicrobial resistance selection.

Access to safe drinking water remains a global challenge. A low-cost, self-powered floating device points to a new direction in smart water treatment by unifying contaminant detection and disinfection in a single platform.

Machine learning studies in water treatment often report high retrospective accuracy but rarely show whether models can support operational decisions. We analyse 423 full-text studies and a bounded six-plant wastewater-treatment scenario. Plant deployment (2.8%), real-time testing (5.2%), future-facing validation (18.2%) and uncertainty reporting (8.5%) remain uncommon. We propose an operation-aware evidence standard linking sensing context, validation timing, uncertainty and control role to deployment claims.