??Reflections from #ClimateWeekNYC provided by our amazing #DataScientist Amy Piscopo ?? ? Thoughts I Keep Coming Back To, Two Weeks Later ?? ? A major theme across sectors at Climate Week was: ???DATA. DATA. DATA. ? A data nerd myself, I enjoyed connecting with others working with various types of climate data, from hyperlocal precipitation data at tomorrow.io ? to HiveTracks data for biodiversity monitoring! ?? ? Why the emphasis on data? In my field, #hydrology ??, streamflow and flood inundation models are driving valuable insights, such as flood forecasting for communities ??? and damage estimates for insurance ??. ? However, a model is only as good as its underlying data. Reliable, spatially diverse data is necessary to train, calibrate, and validate our models. Otherwise, the standard adage applies –?garbage in, garbage out?– no matter how sophisticated the model. ? For data derived from fixed infrastructure, gaps in data coverage can introduce bias and uncertainty into models. But how often do we talk about this in the context of human implications? Think of the benefits of an advanced flood alert ??, for instance, and how the accuracy of that alert may decrease with distance from the sensor. ? At Divirod, we understand not only the model implications, but more importantly, the human implications of data scarcity. We keep this in mind with each sensor placement in our effort to support equitable decision-making around #ClimateChangeAdaptation ?? Climate Group
Divirod Inc.的动态
最相关的动态
-
????? New publication about #Bias #Correction of #Climate #Models is out: ?? ?????????????????????????? ???????????????????? ???? ???????? ???????????????????? ?????????????? ???? ?????????????? ?????????? ??????????????????????: ?????????????????????? ???? ??????5-???????? ???????????? ?????????????????? ???????????????? ?????????????????????? ?? ?????????????? ???????? is critical across sectors like water and energy management, agriculture, and disaster planning, especially with growing demands for reliable forecasting and climate change adaptation. Despite advancements in climate model accuracy, their outputs often r???????????? ???????? ???????????????????? to ensure ?????????? ??????????????????????????. ?? In this study, we conducted a ?????????????????????????? ???????????????? of bias correction methods, ranging from ?????????????????????? ???????????????????? and ???????????????????????? ???????????????????? to advanced ?????????????? ???????????????? approaches across various temporal resolutions using ERA5-Land reanalysis data in the complex Alpine environment. ?? Key Highlights: - Compared the performance of bias correction methods for ?????????????????????????? and ??????????????????????. - Explored trends across ????????????, ??????????, and ??????????????. - Delivered insights to improve the applicability of ?????????????? ?????????? outputs for ?????????? ?????????? ???????????????? ????????????????????. ?? The findings aim to bridge the gap between ????????????????? ?????????????? ???????????? and ?????????? ????????????????????????, offering a robust framework for bias correction tailored to diverse temporal and spatial contexts. ?? Read more here: https://lnkd.in/dRgeMVT4 #ClimateResearch #BiasCorrection #DataMining #Hydrology #ERA5Land #MachineLearning #WaterResources
-
Excited to share my latest abstract titled "Forecasting the Next Decade of Mean Annual Rainfall Based on Historical Data and CHIRPS Data Using RStudio"! ?? This research explores innovative methodologies to predict rainfall patterns, contributing to better climate resilience strategies. I'm grateful for the support and guidance from my mentors and colleagues. #ClimateResearch #RainfallForecasting #RStudio #EnvironmentalScience #remotesensing #Climatechange #CHIRPS
-
-
Out of many inspirations at #NYClimateWeek over the past two days, a presentation by ClimateIQ at the Urban Systems Lab stands out as a fascinating example of #AI being applied for climate hazard mitigation and social impact. ClimateIQ is doing hyper-local data modeling of heat and flood scenarios for urban risk management. They start with data-rich cities as partners, which not only delivers high resolution insights to those populations but is also used to train machine learning that will then allow interpreting lower-data locations in more vulnerable communities so everyone benefits. The more we train the smarter we get, collectively! To get this project off the ground they partner with city governments around the world to co-develop this tool and validate the models, then empower the local stakeholders to collaborate on using this mitigation intelligence toward action with access to data in flexible ways. Incredible project, executed with world-class focus and clarity in partnership with Google .org In Possibility Ocean we’re exploring new ways to leverage Ocean Data. How might this approach unlock new possibilities in coastal resilience planning and marine ecology risk management? #nycw #climate #climatetech #oceantech #mitigation #resilience
-
It's happening today! Join us for an enlightening webinar at 1 pm ET on climate change where you'll explore diverse data sources, learn analysis techniques and discover how to showcase your findings in web applications. Sign up now: https://lnkd.in/gXjJy8X7 #GettingTechnical #HowToArcGIS #climatechange #GIS #webinar
-
-
"Detecting Floods on Fiji's Croplands", a short talk by Dr John Duncan, is now available as part of the "AI for Climate Action: From Data to Impact - Climate Week New York City 2024" event recording ?? In this short talk, John discusses how he and colleagues worked with the Fijian government and people to identify problems and help set up new data science capabilities to help manage Fiji's agricultural response to climate disasters. John's talk begins at 00:19:40 here ?? https://zurl.co/mbr0 Followed by a panel talk at 00:54:50 #ClimateChange #Sustainability #Geospatial #AI #DataScience UWA School of Agriculture and Environment
-
-
This Environment Day, let's focus on ???????? ??????????????????????: ??????????????????????????????, ?????????????? ????????????????????. At Ridgeant Technologies, we believe data science can be a powerful tool to heal our planet. Desertification and drought threaten our planet's health, but data offers a glimmer of hope. ???????? ?????????????? ???????? ???? ????: - ???????????????? ???????????????????? ??????????: Satellite imagery and environmental data pinpoint regions most at risk, allowing for targeted restoration efforts. - ???????????????? ???????? ????????????????????: Data analysis helps develop sustainable agricultural practices that prevent land degradation and promote healthy soil. - ?????????????? ????????????????: Tracking data on vegetation growth and water infiltration helps measure the effectiveness of restoration efforts. Together, let's use data science to cultivate a more resilient future for our land. #EnvironmentDay #LandRestoration #DataScienceForGood #SustainableFuture #SaveThePlanet #DataDrivenSolutions #Ridgeant
-
-
It’s tempting to see most Al “recycling” existing info, but some applications turn that upside down. This model outperformed other methods 97% of the time in predicting accurate 15-day forecasts. What this means is that in the new stage we’re moving into - the one of climate disasters, drought-impacted farming etc. - this is game changing tool. https://lnkd.in/dN8KPgk8
-
Accurately predict rainfall and master climate challenges! In the face of unpredictable weather, this paper brings a revolutionary rainfall prediction solution. Based on a large amount of historical data, an advanced parallel hybrid algorithm was developed, integrating the essence of multiple statistical models to significantly improve the accuracy of monthly rainfall prediction. Compared with traditional SARIMA, Holt-Winters, ETS and other models, the model in this article shows excellent advantages in prediction accuracy, and the error index is fully optimized. After rigorous verification, the model is not only technically leading, but also statistically significant. Choose this paper to make weather forecasts more accurate, provide solid support for your agriculture, water conservancy, disaster prevention and other fields, jointly cope with climate challenges, and create a better future! Find more information, please visit through: https://lnkd.in/ghMJ-ySC #ESCI #mathematics #Scopus
-
-
?? Water safety is under threat across the US! ?? What's the invisible danger lurking in our lakes and rivers? ?? Los Alamos National Laboratory has unleashed AI to tackle this growing menace! ???? ?? Ever wondered how climate change is turning quiet waters into toxic zones? ????? Discover how AI is diving deep into decades of data to predict and combat harmful algal blooms! ???? ?? Why should we care about these blooms, and what can we do to prevent them? Click to uncover the answers and see how technology could save our waters! ???? https://lnkd.in/gKzFDc8H ?? Follow Karmactive for more on how tech meets environmental challenges! ?? #WaterSafety #ClimateAction #TechForGood #AIInnovation #EnvironmentalImpact
-
Natural ecosystems are being degraded by local land use and global climate changes, but these have been hard to disentangle using traditional modeling methods. In a new paper by Kimberly Bourne, Ph.D., Shan Zuidema, Celia Chen, Mark Borsuk, and myself, we apply machine learning methods to a case study in Vermont and New Hampshire watersheds to demonstrate that local trends such as suburbanization likely drive impacts on river health over the next century. We show that these modeling approaches can better leverage the wealth of forecasted climate, meteorology, and land-use data to which policy-makers have access in order to improve decision-making, for example, to capture the benefits of densification and green infrastructure investments. More in Meteorological Applications (OA??) : https://lnkd.in/e-cHYb3g
