New climate and earth-science research published between October 20 and October 22, 2026, covers global temperature trends, forest carbon loss, geophysical monitoring, and weather modification. A reanalysis of the Berkeley Earth record confirmed a warming rate of 0.206 degrees Celsius per decade since 1970, noting that evidence for recent acceleration remains weak. In the Amazon, a deep learning study found that 80 percent of carbon loss from 2019 to 2022 resulted from small-scale disturbances invisible to standard monitoring. Regarding geophysical hazards, researchers introduced an InSAR technique that improves north-south displacement precision tenfold, while an analysis of Nepal's 2026 rock-ice avalanches revealed that current hazard tools fail to capture full downstream risks. Finally, an independent review of Utah's 2016-2025 cloud seeding program reported a 2.7 percent precipitation increase per storm, which is lower than previous operational claims. Synthesized from 102 manifests produced by 19 monitored earth-science preprint sources in the last 7 days, including EarthArXiv, ESS Open Archive.
Answer updated Oct 23, 2026 12:00 UTC · rebuilt twice daily from the rolling 168-hour window
The manifests behind this answer
Robust warming rates, marginal evidence of acceleration and unstable extrapolation: a reproducible reanalysis of statistical trend models for the Berkeley Earth global temperature record
A reanalysis of the Berkeley Earth global temperature record (1850-2024) confirms a warming rate of 0.206 ± 0.022 °C/decade since 1970, consistent with NASA GISTEMP and HadCRUT5 datasets. The study finds that evidence for a recent acceleration in warming is statistically weak and that standard polynomial trend models a
EarthArXiv - Official Climate Attribution Risk · 2026-10-21 · 3 claims · manifest 1792597240026887662 source →
Evaluating pre-event hazard screenings against the 2026 Budhi Gandaki and Bhotekoshi-Trishuli rock-ice avalanche cascades, central Nepal
The 2026 rock-ice avalanche cascades in central Nepal demonstrate that current hazard screening tools-specifically 14° runout models and GLOF inundation maps-fail to capture the full extent of downstream risk to settlements and hydropower infrastructure. While the HAND (Height Nearest Drainage) 0-50m screening performe
ESS Open Archive - Official Ocean Cryosphere SEA Level · 2026-10-22 · 3 claims · manifest 1792696023374854808 source →
Previously undetected disturbances are the dominant driver of carbon loss in protected Amazonian forests
This study quantifies Amazonian carbon loss from 2019 to 2022 using a high-resolution deep learning framework that integrates spaceborne LiDAR with Sentinel-1/2 imagery. It identifies that 80% of carbon losses stem from small-scale disturbances that remain largely invisible to standard Landsat-based monitoring. These f
EarthArXiv - Official Hydrology Water Security · 2026-10-22 · 3 claims · manifest 1792680087267996865 source →
Near-complete 3D deformation mapping from Sentinel-1 through extended burst overlap interferometry
A new InSAR processing technique significantly enhances the ability to map 3D surface deformation using Sentinel-1 satellite data. By extending burst overlap regions, the method achieves 86% coverage and a 10-fold improvement in north-south displacement precision compared to standard intensity offset methods. This adva
ESS Open Archive - Official Hazards Resilience · 2026-10-20 · 3 claims · manifest 1792532548529991058 source →
Storm-Level Independent Verification of Operational Cloud Seeding: Seven Seasons of Utah's Program, 2016-2025
An independent, storm-level analysis of Utah's winter cloud seeding program (2016-2025) challenges the efficacy of current operational claims. By applying a wind-rotation test to 181 generator sites and 525 seeded storm periods, the study finds a modest precipitation increase of 2.7% per storm, significantly lower than
EarthArXiv - Official Atmosphere AIR Quality · 2026-10-20 · 3 claims · manifest 1792466412186066521 source →