A phys.org article (dated around late September 2026) appears to cover concerns about the rapid growth of satellite constellations and escalating risks to Earth’s orbital environment, based on recent reports and analyses.
The exact page did not load successfully, but closely matching contemporary coverage, especially the European Space Agency’s (ESA) Space Environment Report 2026 (data through end of 2025, released mid- September 2026) and related discussions, points to this theme.
Key points from the ESA report and related coverage
Explosive growth in satellites: Commercial constellations continue expanding rapidly. In 2025 alone there were more than 300 launches putting over 4,000 new payloads into orbit. Around 10 new payloads are launched daily on average, while more than three intact satellites or rocket bodies reenter Earth’s atmosphere daily. Total active satellites are in the range of ~15,000- 17,000 (heavily driven by Starlink and similar systems), up sharply from a few thousand before the mega constellation era began around 2019.
Rising collision and debris risks: Not enough satellites are reliably leaving congested low- Earth orbits (LEO) at end- of- life. Projections show an unstable environment with collision rates potentially increasing exponentially under current trends (even assuming continued best- practice disposal rates). The “Kessler syndrome” risk, where collisions generate debris that triggers further collisions, is becoming visible earlier in models. ESA shortened its projection horizon from 200 to 100 years and still sees skyrocketing object numbers and collisions.
Space Environment Health Index worsened sharply: This single metric (comparing long- term effects against a sustainable 2014 benchmark) jumped roughly an order of magnitude (from ~4 to ~50) in one year, indicating current activities are far beyond sustainable levels for the orbital environment.
Other trends: Constellations are shifting altitudes more significantly year- to- year. A new metric tracks rising (still low absolute) on- ground casualty risk from surviving reentry fragments due to higher traffic. Fragmentation events and unidentified objects are adding complexity. Active debris removal is increasingly needed alongside better mitigation, as prevention alone is no longer sufficient.
Broader context from parallel reporting: Related analyses highlight threats to astronomy (trails and brighter night skies from large, proposed fleets, including potential million- satellite data- center constellations and mirror satellites), atmospheric pollution from launches/reentries, and the sheer scale of future filings (hundreds of thousands to over 1.7 million additional satellites proposed).
ESA emphasizes that mitigation adherence is improving in some areas (e.g., more controlled reentries, shorter disposal targets moving from 25 to 5 years), but efforts need to accelerate. The agency is pursuing “Zero Debris” goals by 2030, better tracking, design-for-demise technologies, and active removal demonstrations.
In short, the article highlights how the satellite “boom” is dynamically stressing orbital sustainability, with data showing the problem accelerating faster than mitigation is keeping pace. For the full original text or images, try the phys.org link directly or the ESA report PDF.
Drylands play a disproportionately large role in the global carbon cycle, despite low average productivity. They cover roughly 40- 45% of Earth’s land surface (~66– 67 million km²), support over 2 billion people, and dominate both the long- term trend and interannual variability (IAV) of the terrestrial carbon sink.
Key Carbon Stocks
Vegetation (biomass): Drylands hold ~22- 30% of global above- and below- ground biomass carbon (rough estimates around 790 Pg above -ground and 614 Pg below- ground in some inventories).
Soil organic carbon (SOC): Approximately 32% of the global SOC pool, estimates include ~248 Pg (0- 30 cm), 420- 470 Pg (0- 1 m), and ~646 Pg (to 2 m).
Soil inorganic carbon (SIC): Even more dominant, drylands contain the large majority (~80%) of the global SIC pool (hundreds of Pg to 1- 2 m depth). Secondary carbonate formation can act as a long- term sink.
Total ecosystem carbon stocks are substantial at the global scale, though densities per unit area are lower than in humid forests.
Fluxes and Dynamics
Drylands contribute ~40% of global net primary production (NPP) in some estimates and are the largest source of IAV in the land carbon sink (often ~30- 40% or more of global terrestrial net carbon flux IAV). Semi- arid ecosystems (e.g., tropical and subtropical grasslands, savannas, and shrublands) are especially important.
Productivity (GPP/NPP): Strongly limited by water availability. Rainfall pulses trigger rapid growth and microbial activity, producing large CO₂ uptake events in wet years and respiration pulses. CO₂ fertilization has contributed to greening and higher light- use efficiency in many areas.
Respiration and turnover: Soil respiration can release both recent and old carbon (radiocarbon studies show mean ages of ~520 years for respired CO₂ and ~2100 years for bulk SOC in some drylands). Old carbon is more vulnerable than models often assume.
Abiotic processes: Photodegradation, thermal degradation, wind and water erosion, and inorganic carbon dynamics (including mineral adsorption and desorption of CO₂) are more important than in humid systems. Recent work has highlighted abiotic carbon fluxes into desert soils that can rival biological fluxes locally.
Observed Trends and Variability
Satellite records (LAI, GPP, NDVI) generally show persistent greening across much of the global drylands from the 1980s/2000s onward, driven by a combination of rising CO₂, climate variability, and (in many regions) human activities such as cropland expansion, irrigation, and fertilization.
However, a 2026 Nature Climate Change study found that these greening masks increasing interannual variability (covering ~82% of drylands). LAI maxima are rising while minima are declining, creating a stronger “boom- and- bust” pattern. Increasing rainfall sensitivity is the primary driver, with inter- and intra-annual rainfall variability as secondary factors. Dynamic global vegetation models (DGVMs, e.g., from TRENDY) under- estimate observed greening rates and fail to capture the rising variability.
This heightened variability may signal approaching state transitions, increasing carbon- cycle uncertainty and threatening ecosystem services (e.g., forage, soil stability, water regulation).
Other recent findings note possible slowdowns in dryland GPP growth in some periods due to rising vapour pressure deficit (atmospheric dryness), contrasting with continued gains in humid regions.
Model Performance and Uncertainties
DGVMs and land surface models often:
- Capture spatial patterns of GPP reasonably well but struggle with trends, stocks, and IAV.
- Over- attribute changes to CO₂ fertilization while under- representing water constraints, land management, nitrogen dynamics, and abiotic processes.
- Project net carbon gains in dryland soils driven by CO₂ effects, but empirical evidence for multi-decadal soil sinks is limited.
Definitions of “dryland” (aridity indices, biome classifications) and sub-biome differences (e.g., Australian vs. southwestern US systems) strongly affect quantified contributions.
Climate Change, Land Use, and Future Outlook
Expansion: Drylands are projected to expand (though CO₂ effects on stomatal conductance and PET can moderate this). Expansion into formerly more productive areas may not increase global GPP.
Vulnerability: Higher temperatures, more extreme droughts or wet extremes, and desertification risks can release stored carbon (organic and potentially inorganic) and reduce sink strength.
Management potential: Sustainable practices (reduced tillage, cover crops, controlled grazing, restoration) can enhance SOC, though achievable rates in drylands are modest (tens of kg C ha⁻¹ yr⁻¹) and fall short of ambitious global targets. Avoiding degradation is critical for both carbon and livelihoods.
In summary, drylands are highly dynamic, water- limited systems whose carbon cycle exerts outsized influence on global atmospheric CO₂ variability and trends.
Observed greening is real but increasingly unstable; models lag observations; and both biotic (pulse-driven) and abiotic processes matter. Rising variability and potential state shifts heighten uncertainty for the future land carbon sink.
Improved observations (flux towers, radiocarbon, high- resolution remote sensing), better representation of dryland- specific processes in models, and adaptive land management are priorities.
Greening masks stability loss in drylands
Drylands are undergoing large- scale changes due to climate change, but understanding remains limited by observational gaps, conflicting satellite records, and model uncertainty. Using long- term trends and interannual variability in leaf area index (LAI) from multiple satellite products (1982- 2020), the study finds:
- Persistent greening across global drylands, with rates higher than those simulated by current dynamic global vegetation models (DGVMs). This contradicts some earlier reports of a recent greening slowdown.
- Increasing interannual variability in LAI that is spatially coherent and covers about 82% of global drylands. This is characterized by rising LAI maxima and declining minima.
- The dominant driver of the rising variability is increasing rainfall sensitivity, with both inter- and intra-annual rainfall variability as secondary contributors.
- The DGVMs examined (from the TRENDY ensemble) fail to reproduce the observed increase in variability and show divergent LAI trajectories compared with observations.
The authors note that rising variability may signal impending state transitions. These could increase uncertainty in the carbon cycle and disrupt important dryland ecosystem services.
Additional context from the paper
- Analyses rely on multiple LAI datasets (GIMMS4g, GLOBMAP, SNU, and MODIS for the more recent period) plus climate data (precipitation, temperature, VPD, PET) and land- cover products.
- Model comparisons use outputs from 13 DGVMs in TRENDYv13.
- Supporting figures examine spatial patterns of greening or browning, decomposition of coefficient of variation (LAIcv) into standard deviation and mean, product- specific trends, and drivers of variability.
- Data and code are publicly available via Zenodo repositories listed in the article.
Corresponding authors: Wen Zhang and William K. Smith (University of Arizona and collaborators from multiple institutions including Western University, NASA, USGS, and others).
The article is subscription and paywalled for the full text beyond the abstract and extended data summaries. Public datasets and code links are provided for reproducibility.
Journal information: Nature Climate Change
DOI: 10.1038/s41558-026-02733-7
Provided: University of Arizona
Authors: Wen Zhang,
David J. P. Moore,
Yang Li,
Natasha MacBean,
Andrew F. Feldman,
Christopher R. Schwalm,
Yanghui Kang,
Julia K. Green,
Benjamin Poulter,
Mostafa Javadian,
Russell L. Scott,
Zexi Shen,
Youngryel Ryu,
Sungchan Jeong,
Sasha C. Reed &
William K. Smith
Abstract
Drylands are already undergoing large-scale change due to climate change, yet understanding of this response remains limited by observational gaps, conflicting satellite records and substantial model uncertainty. Here, considering trends and interannual variability based on leaf area index (LAI), we show persistent (1982–2020) greening across global drylands and increasing interannual variability. Greening rates are higher than current dynamic global vegetation models and contradict reports of a recent greening slowdown. The interannual variability increase is spatially coherent, covering about 82% of global drylands, and is characterized by increasing LAI maxima and declining minima. Increasing rainfall sensitivity is the dominant driver of increasing variability with both inter- and intraannual rainfall variability acting as secondary drivers. The dynamic global vegetation models analysed cannot reproduce increasing variability and divergent LAI trajectories. Rising variability may indicate eminent state transitions that could intensify carbon cycle uncertainty and disrupt key dryland ecosystem services.
You must be logged in to post a comment.