Add BrightSurf on Google Email

Researchers develop machine learning model to improve Amazon carbon storage estimates

Researchers developed a machine learning model using high-resolution satellite imagery to estimate aboveground carbon stocks in the Amazon. The study found that accounting for uncertainties in forest degradation classification led to lower estimates of mean carbon density, suggesting earlier estimates may have been over-optimistic.

SourceOregon State University·JournalCarbon Balance and Management·TypeComputational simulation/modeling·DateFeb 20, 2023

New technique maps large-scale impacts of fire-induced permafrost thaw in Alaska

A new technique maps the effects of fire-induced permafrost thaw in Alaska, revealing widespread topographic change and vegetation shifts. The study used a machine learning-based approach to quantify thaw settlement across 3 million acres of land, with results showing a significant loss of evergreen forest and shrubland encroachment.

SourceFlorida Atlantic University·JournalEnvironmental Research Letters·TypeComputational simulation/modeling·DateFeb 14, 2023

Scientific AI’s ‘black box’ is no match for 200-year-old method

A new study uses Fourier analysis to understand how deep neural networks learn complex physics. By analyzing the equation of a fully trained model, researchers were able to identify crucial information about how the network learns and generalizes. This breakthrough could accelerate the use of scientific deep learning in climate science.

SourceRice University·JournalPNAS Nexus·TypeComputational simulation/modeling·DateFeb 13, 2023

Devastating cost of future coastal flooding for many developing nations predicted in new study

A new study predicts devastating socioeconomic impacts of future extreme coastal flooding on developing nations caused by climate change. Without adaptation measures, the number of people affected could increase from 34 million in 2015 to 246 million by 2100, with expected annual damage costing over five percent of national GDP.

SourceUniversity of Melbourne·JournalFrontiers in Marine Science·TypeComputational simulation/modeling·DateFeb 6, 2023

Remapping the superhighways travelled by the first Australians reveals a 10,000-year journey through the continent

Researchers used sophisticated models to estimate the peopling of Sahul, revealing a 10,000-year journey across the continent. The ancestors of Aboriginal people first entered the continent 75,000-50,000 years ago from Timor, expanding southward and northward to settle all parts of New Guinea and Australia.

SourceFlinders University·JournalQuaternary Science Reviews·TypeData/statistical analysis·DateFeb 2, 2023

Reducing water flow model uncertainty

Researchers connected global climate modeling with local hydrological data to improve water flow models. They found that using smaller tree samples and younger trees yielded better model validation results.

SourceUniversity of Connecticut·JournalJournal of Advances in Modeling Earth Systems·TypeComputational simulation/modeling·DateJan 19, 2023

Changes in the approximation of snow by climate models over typical vegetation in the Northern Hemisphere

Researchers analyzed relationships between snow cover and depth for three vegetation types in the Northern Hemisphere. Different interactions between land cover and snow processes were found, which climate models struggle to reproduce. This study highlights the need to improve snow parameterization schemes.

SourceInstitute of Atmospheric Physics, Chinese Academy of Sciences·JournalAtmospheric and Oceanic Science Letters·DateJan 13, 2023

Offshore and coastal risk analyses may misrepresent wave storms from extreme weather like bomb cyclones

Researchers found that global wave models can vary in their estimates of extreme wave heights by up to 20 feet, which can lead to underprotected areas during extreme events. The study emphasizes the need for considering multiple models and modern observational data to better assess offshore and coastal risks.

SourceUniversity of Central Florida·JournalScience Advances·TypeData/statistical analysis·DateJan 11, 2023

Researchers propose a more effective method to predict floods

A team of researchers from Xi'an Jiaotong-Liverpool University and other institutions has identified a flexible and user-friendly model for predicting flood frequency in a changing environment. The fractional polynomial-based regression method is more effective than existing models, which often fail to account for factors like climate ...

SourceXi'an Jiaotong-Liverpool University·JournalJournal of Hydrology·TypeComputational simulation/modeling·DateJan 9, 2023

Climate change could cause “disaster” in the world’s oceans, say UC Irvine scientists

Climate-driven heating of seawater is causing a slowdown of deep circulation patterns in the Atlantic and Southern oceans, which could lead to a complete shutdown by 2300. This would intensify global warming and reduce ocean uptake of carbon dioxide, resulting in declining marine ecosystems.

SourceUniversity of California - Irvine·JournalNature Climate Change·TypeComputational simulation/modeling·DateJan 4, 2023

Tipping points complicate the evaluation of complex climate models

Researchers warn that commonly used methods for estimating equilibrium climate sensitivity in complex climate models may be underestimating temperature rise. The study highlights the need for longer simulations to account for potential late tipping points, which could significantly influence global warming.

SourceUniversity of Copenhagen - Faculty of Science·JournalProceedings of the Royal Society A Mathematical Physical and Engineering Sciences·DateJan 3, 2023

Linking fossil climate proxies to living bacteria helps climate predictions

A new study reveals that certain types of lipids found in ancient fossils are produced by specific living bacteria. By identifying these microorganisms and understanding how they produce the lipids, scientists can create more accurate climate reconstructions. This discovery also sheds light on the early evolution of life on Earth.

Catching up to climate change by tracking big-picture patterns

By analyzing satellite imagery and tracking net primary productivity, researchers aim to predict how ecosystems will respond to climate change. Their model suggests that the rate at which ecosystems adapt to changing conditions is a critical component to reducing uncertainty about future projections.

Finding simplicity within complexity

A University of Houston researcher has developed a method to describe complex systems using the least number of variables possible, reducing complexity from millions to just one. This advancement speeds up science with efficiency and ability to understand and predict natural system behavior.

SourceUniversity of Houston·JournalNature Machine Intelligence·DateDec 8, 2022

Mekong Delta will continue to be at risk for severe flooding

A study by Hokkaido University reveals that the El Niño Southern Oscillation (ENSO) has caused more frequent heavy rainfall in the Mekong Delta, resulting in increased flooding risk. The findings suggest that better weather predictions and preparation can help mitigate the negative effects of floods and droughts in the region.

SourceHokkaido University·JournalScientific Reports·TypeExperimental study·DateDec 7, 2022

Gwangju Institute of Science and Technology researchers embrace uncertainty to make microgrids better

A new optimization model by GIST researchers reduces operating costs and load shedding in microgrids, achieving a 20% decrease in average ENS. The model accounts for variations and uncertainty in renewable energy supply using an ANN-based prediction model, with predicted power output accuracy of 9.7%.

SourceGIST (Gwangju Institute of Science and Technology)·JournalApplied Energy·TypeComputational simulation/modeling·DateDec 6, 2022

Antarctic ice: a better knowledge of the ocean improves the predictability of sea ice variability

Researchers found that using coupled atmosphere-ocean-sea ice circulation models with observation-based datasets, they can predict Antarctic sea ice variability over decadal time scales. The initialization of subsurface ocean temperature and salinity fields significantly improves prediction skills, especially in the west Antarctic region.

Gully erosion prediction tools can lead to better land management

Researchers developed a modeling framework using remote sensing data to predict gully erosion susceptibility. They found that spatiotemporal changes in land cover and precipitation were crucial in predicting gully formation in agricultural areas, with 7.4% of the study area having an elevated risk of developing gullies.

SourceUniversity of Illinois College of Agricultural, Consumer and Environmental Sciences·JournalJournal of Hydrology Regional Studies·TypeData/statistical analysis·DateNov 23, 2022

Early warning signals for climate tipping should be taken very seriously

Scientists have found early warning signals for climate tipping in the Amazon Rainforest, West-Central Greenland ice sheet, and Atlantic Meridional Overturning Circulation. The study provides a clear mathematical understanding of increased variability, sensitivity, critical slowing down, and being close to a tipping point.

SourceUniversity of Copenhagen - Faculty of Science·JournalJournal of Physics A Mathematical and Theoretical·DateNov 17, 2022

Dam safety: New study indicates probable maximum flood events will significantly increase over next 80 years

A new study by UNSW and the University of Melbourne found that estimates of Probable Maximum Precipitation (PMP) for large dams in Australia are expected to increase by 14-38% on average over the next 80 years. This is due to climate change and outdated modelling, which has not been updated since at least 20 years ago.

SourceUniversity of Melbourne·JournalWater Resources Research·DateNov 14, 2022

Subtropical clouds key to southern ocean teleconnections to the tropical pacific

A research team at UNIST has identified subtropical low cloud feedback as a key mechanism driving teleconnections between the Southern Ocean and tropical precipitation. Their findings suggest that this impact is stronger than previously thought, with implications for mid-latitude climate predictions.

SourceUlsan National Institute of Science and Technology(UNIST)·JournalProceedings of the National Academy of Sciences·DateNov 13, 2022

The world will probably warm beyond the 1.5-degree limit. But peak warming can be curbed

New research suggests that the world will probably warm beyond the 1.5-degree limit set by the 2015 Paris Agreement, but peak warming can be minimized by adopting more ambitious climate pledges and decarbonizing faster. The study models scenarios to explore what degree of warming would likely follow different courses of action, highlig...

SourceDOE/Pacific Northwest National Laboratory·JournalNature Climate Change·DateNov 10, 2022

Long-term memory in temperature series: the “forgetfulness” of state-of-the-art climate models in equatorial and coastal regions

Researchers found significant biases in climate models' simulation of long-term memory in temperature series, especially in equatorial and coastal regions. This may be due to imperfectly simulated coupled processes between oceans and atmosphere, leading to uncertainty in regional climate change predictions.

SourceInstitute of Atmospheric Physics, Chinese Academy of Sciences·JournalAtmospheric and Oceanic Science Letters·DateNov 9, 2022