Scientific Computing Software Tools

AI is completely rewriting the rules of weather forecasting, and this video from NVIDIA is a perfect example of how fast things are moving. In just under 5 minutes, the video demonstrates Earth-2, a platform that allows you to run global weather forecasts in mere seconds using just a few lines of Python. You can seamlessly switch between data sources (like ERA5, GFS, IFS) and even swap out entire AI models (like FourCastNet, GraphCast, or Aurora) with a single line of code. But NVIDIA isn’t alone. We are witnessing an arms race among big tech to solve weather prediction: - Google DeepMind has GraphCast and NeuralGCM, which have already outperformed gold-standard physical models in many metrics. - Microsoft released Aurora, a foundation model trained on over a million hours of data, claiming to be 5000x faster than traditional numerical systems. - IBM & NASA recently open-sourced Prithvi, a "geospatial foundation model" designed not just for weather, but to be fine-tuned for specific climate applications. - Huawei has Pangu-Weather, which famously predicted the path of a typhoon more accurately than traditional methods. Why is this happening? - Compute: Traditional Numerical Weather Prediction (NWP) solves complex physics equations requiring massive supercomputers. AI models, once trained, infer results in seconds on a few GPUs. - Ensemble Forecasting: Because they are so cheap to run, we can generate thousands of scenarios (ensembles) instead of just a few. This is a game changer for predicting low probability extreme weather events. - Data Fusion: These models are proving incredibly good at learning patterns from historical data that pure physics equations might miss. For the geospatial practice, this is a big change. Weather is moving from a static dataset we download to a dynamic capability we run. You no longer need a supercomputer to generate high-resolution forecasts; you just need a GPU and a Python script. We may soon see fine-tuned weather models for specific geospatial use cases like hyper local wind for drones, precise precip for agriculture, or cloud cover for satellite tasking. The latency between data in and forecast out is shrinking to near zero, enabling true real time geospatial intelligence. Have you tried any of these models? What are your thoughts? 🌎 I'm Matt Forrest and I talk about modern GIS, earth observation, AI, and how geospatial is changing. 📬 Want more like this? Join 12k+ others learning from my daily newsletter → forrest.nyc

New review out in Trends in Plant Science from our group, tackling one of the most persistent problems in plant breeding: moving complex traits into elite varieties is hard, slow, and expensive, and breeders have long lacked the tools to plan it rigorously. Led by Seema Yaadav, the paper frames trait introgression not as a fixed backcross protocol but as a constrained multi-objective optimisation problem, balancing success probability, elite genome recovery, time, and cost. It maps the full toolkit to tackle this: predictive cross metrics, pan-genomes and introgressiomics libraries, speed breeding, doubled haploids, and recombination engineering. The vision is 'designed diversity'. Pipelines where the genotype is planned computationally before a single cross is made, then refined iteratively as data accumulates. The building blocks are largely here. The challenge now is integrating them into breeder-facing tools that work in practice. Great to see Seema lead this one. Congratulations also to co-authors Meredith McNeil (CSIRO), Peter Dodds (CSIRO), and Ben Hayes (UQ) and thanks to the Grains Research and Development Corporation for supporting our R&D in this space. 📄 https://lnkd.in/gz6tCkAT

🌍 Biggest Earth release of the year just dropped courtesy of NVIDIA 🌍 Production-grade weather prediction just became open and landed on Hugging Face. + Did I mention it's full stack? Meet the new Earth-2 models: → Nowcasting: km-scale severe weather prediction → Medium-range: 15-day forecasts → Global Data Assimilation: accurate initial conditions The best part is that it's incredibly accessible. NVIDIA shipped two frameworks to get developers up and running really fast: 🛠️ Earth2Studio - build inference pipelines with just a few lines of Python 🛠️ PhysicsNemo (for earth) - train new models with your custom data If you want all the gory details, there are three research papers and a comprehensive blog post to keep you busy (which I'll link in the comments). This is going to fundamentally change disaster prediction and response, transform agricultural planning, ++ hopefully give the commercial weather providers a serious run for their money, which should enfranchise lower resource regions to also predict and safeguard their climates. Open source for the win! 🤗

Precipitation is one of the most challenging variables to accurately simulate in global climate models as it depends on small-scale physical processes. In our latest research published in 𝘚𝘤𝘪𝘦𝘯𝘤𝘦 𝘈𝘥𝘷𝘢𝘯𝘤𝘦𝘴, we describe an advancement in our hybrid atmospheric model, NeuralGCM, which now leverages AI trained directly on NASA satellite observations to improve global precipitation simulations. Key results of this work: 👉 Physics-AI Integration: The model combines a traditional fluid dynamics solver for large-scale processes with AI neural networks that learn to account for the effects of small-scale physics, specifically precipitation. 👉 Improved Extremes: NeuralGCM demonstrates significant improvements in capturing the intensity of the top 0.1% of extreme rainfall events, better representing heavy precipitation than many traditional models. 👉 Long-Term Accuracy: In multi-year simulations, the model achieved a 40% average error reduction over land compared to leading atmospheric models used in the latest Intergovernmental Panel on Climate Change (IPCC) report. 👉 Daily Patterns: It more accurately reproduces the timing of peak daily precipitation, which is critical for hydrology and agricultural planning. We are already seeing the value of this approach in the field. A partnership between the University of Chicago and the Indian Ministry of Agriculture recently used NeuralGCM in a pilot program to help predict the onset of the monsoon season. NeuralGCM is part of our Earth AI program to better understand the physical earth in ways that benefit society. We have made the code and model checkpoints openly available to the community. Read the full details on the Google Research blog by Janni Yuval: goo.gle/4qH63sU Paper: https://lnkd.in/d7E4US4W

You might have seen news from our Google DeepMind colleagues lately on GenCast, which is changing the game of weather forecasting by building state-of-the-art weather models using AI. Some of our teams started to wonder – can we apply similar techniques to the notoriously compute-intensive challenge of climate modeling? General circulation models (GCMs) are a critical part of climate modeling, focused on the physical aspects of the climate system, such as temperature, pressure, wind, and ocean currents. Traditional GCMs, while powerful, can struggle with precipitation – and our teams wanted to see if AI could help. Our team released a paper and data on our AI-based GCM, building on our Nature paper from last year - specifically, now predicting precipitation with greater accuracy than prior state of the art. The new paper on NeuralGCM introduces 𝗺𝗼𝗱𝗲𝗹𝘀 𝘁𝗵𝗮𝘁 𝗹𝗲𝗮𝗿𝗻 𝗳𝗿𝗼𝗺 𝘀𝗮𝘁𝗲𝗹𝗹𝗶𝘁𝗲 𝗱𝗮𝘁𝗮 𝘁𝗼 𝗽𝗿𝗼𝗱𝘂𝗰𝗲 𝗺𝗼𝗿𝗲 𝗿𝗲𝗮𝗹𝗶𝘀𝘁𝗶𝗰 𝗿𝗮𝗶𝗻 𝗽𝗿𝗲𝗱𝗶𝗰𝘁𝗶𝗼𝗻𝘀. Kudos to Janni Yuval, Ian Langmore, Dmitrii Kochkov, and Stephan Hoyer! Here's why this is a big deal: 𝗟𝗲𝘀𝘀 𝗕𝗶𝗮𝘀, 𝗠𝗼𝗿𝗲 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆: These new models have less bias, meaning they align more closely with actual observations – and we see this both for forecasts up to 15 days, and also for 20-year projections (in which sea surface temperatures and sea ice were fixed at historical values, since we don’t yet have an ocean model). NeuralGCM forecasts are especially performant around extremes, which are especially important in understanding climate anomalies, and can predict rain patterns throughout the day with better precision. 𝗖𝗼𝗺𝗯𝗶𝗻𝗶𝗻𝗴 𝗔𝗜, 𝗦𝗮𝘁𝗲𝗹𝗹𝗶𝘁𝗲 𝗜𝗺𝗮𝗴𝗲𝗿𝘆, 𝗮𝗻𝗱 𝗣𝗵𝘆𝘀𝗶𝗰𝘀: The model combines a learned physics model with a dynamic differentiable core to leverage both physics and AI methods, with the model trained directly on satellite-based precipitation observations. 𝗢𝗽𝗲𝗻 𝗔𝗰𝗰𝗲𝘀𝘀 𝗳𝗼𝗿 𝗘𝘃𝗲𝗿𝘆𝗼𝗻𝗲! This is perhaps the most exciting news! The team has made their pre-trained NeuralGCM model checkpoints (including their awesome new precipitation models) available under a CC BY-SA 4.0 license. Anyone can use and build upon this cutting-edge technology! https://lnkd.in/gfmAx_Ju 𝗪𝗵𝘆 𝗧𝗵𝗶𝘀 𝗠𝗮𝘁𝘁𝗲𝗿𝘀: Accurate predictions of precipitation are crucial for everything from water resource management and flood mitigation to understanding the impacts of climate change on agriculture and ecosystems. Check out the paper to learn more:  https://lnkd.in/geqaNTRP

Plant breeding generates massive datasets, but most remain locked in silos due to trust barriers and technical incompatibilities that slow innovation. Our new study introduces "Data cohorts" (structured packages of interoperable breeding data) paired with a Data Trustee Platform that enables federated sharing while protecting proprietary interests. By implementing FAIR principles from the start and using dynamic licensing with secure analysis environments, we show how genomic prediction accuracy doubled when aggregating previously isolated datasets. The key? Standardized metadata, common genotype references, and quality metrics that let breeders access relevant data without compromising competitive advantage. When data flows freely but safely, everyone's breeding programs get stronger and innovation accelerates. https://lnkd.in/dq2dJUHk

⭐ AgroSynapsis Practical Tip 09: How to choose the right genotyping platform for my breeding program? First, we should focus on what really defines a genotyping platform: 👉 Reproducibility (across batches, labs, and years) 👉 Throughput (how many samples you can process efficiently) 👉 Missing data rate (and how much cleaning/imputation you’ll need) These three factors determine the data quality as tool of decision making. 🧬 Main genotyping platforms in plant breeding 1️⃣ SSRs (Simple Sequence Repeats) PCR-based markers targeting repeat regions in the genome. Moderate throughput Low missing data ⚠️ Limited reproducibility across labs (allele calling can vary) 👉 Still useful for: parentage, diversity, legacy datasets 2️⃣ SNPs (KASP assays / SNP arrays) Targeted genotyping of predefined SNPs at known genomic positions. KASP (targeted SNP panels): flexible, low-to-medium throughput SNP arrays (fixed or customized panels): scalable, high-throughput platforms ✔️ High reproducibility ✔️ Low missing data ✔️ Standardized workflows and QC 👉 Customized SNP arrays allow you to focus on: Trait-linked markers Breeding-specific germplasm Long-term program consistency 👉 Widely used for: Routine selection Genomic selection (especially arrays) Seed purity and quality control 3️⃣ GBS (Genotyping-by-Sequencing) A reduced-representation sequencing approach using restriction enzymes and barcoding to sample SNPs across the genome. ✔️ Very high throughput ✔️ Low cost per sample ❗ High missing data ❗ Lower reproducibility across runs ❗ Requires bioinformatics pipelines and stringent QC 👉 Typical uses: Very large early-generation populations Species without SNP arrays GWAS and biparental QTL mapping Diversity panels Entry-level genomic selection ⚖️ How platforms compare on key parameters GBS: High throughput ✅ | Missing data ❌ | Reproducibility ⚠️ → Powerful, but requires strict filtering and careful interpretation SSRs: Moderate throughput ⚠️ | Low missing data ✅ | Reproducibility ❌ → Informative but difficult to standardize SNP arrays / KASP panels: High throughput ✅ | Low missing data ✅ | High reproducibility ✅ → The most robust option for routine decisions 🎯 So… which platform for which objective? 👉 Routine selection (MAS): Use targeted SNP panels (KASP) or customized SNP arrays ✔️ Focus on a compact set of validated, trait-linked markers ✔️ More effective than hundreds of dispersed genome-wide SNPs 👉 Genomic selection (advanced programs): Use medium- to high-density SNP arrays ✔️ Consistent, comparable data across years and environments 👉 Discovery & early-stage selection: Use GBS ✔️ Explore diversity, discover markers, screen large populations ay low cost 👉 If you’d like to be informed about the upcoming workshops organized by AgroSynapsis, and receive early access and discounts, 𝗳𝗶𝗹𝗹 𝗼𝘂𝘁 𝗼𝘂𝗿 𝘀𝗵𝗼𝗿𝘁 𝘁𝗿𝗮𝗶𝗻𝗶𝗻𝗴 𝗶𝗻𝘁𝗲𝗿𝗲𝘀𝘁 𝗳𝗼𝗿𝗺 here: https://lnkd.in/g3tApqPz Genotyping #MolecularBreeding #MAS#AgroSynapsis

Acceleration breeding in corn (maize) is a modern breeding approach aimed at reducing the breeding cycle time and speeding up the development of new hybrids or inbreds. It integrates various tools like off-season nurseries, doubled haploids, marker-assisted selection, genomic selection, and controlled environments to fast-track breeding.🌽 Acceleration Breeding Process in Corn: Step-by-Step1. Parental SelectionChoose elite or diverse parents based on trait performance, heterotic grouping, or genotypic data.Use molecular markers or genomic prediction models for precision.2. Hybridization (Crossing)Perform controlled crosses to generate F1 seeds.Multiple crosses can be planned in staggered nursery setups (e.g., main + off-season).3. Speed Advancement of GenerationsUse the following techniques to rapidly advance generations:✅ Off-Season Nurseries / Counter-Season NurseriesGrow 2–3 generations per year by shifting between locations (e.g., India: Kharif + Rabi + off-season in South India or tropics).✅ Doubled Haploid (DH) TechnologyInduce haploid embryos using haploid inducers.Chromosome doubling → instant homozygous inbred lines (within 1–2 seasons).✅ Greenhouse/Controlled EnvironmentsControlled temperature, photoperiod, and irrigation allow cycle shortening (e.g., F1 to F2 in 6–8 weeks).4. Early Generation SelectionUse marker-assisted selection (MAS) or genomic selection (GS) to:Identify favorable alleles early.Avoid phenotyping low-potential material.This reduces the number of lines carried forward.5. Rapid Line AdvancementCombine speed breeding, DH, and MAS/GS to generate fixed lines in 1–2 years instead of 5–6.6. Testcross EvaluationEvaluate promising inbred lines in tes tcross combinations.Early-stage test crossing helps screen for combining ability.7. Multi-Location Hybrid TestingHigh-throughput phenotyping in target environments.Use precision phenotyping tools (e.g., drones, sensors).8. Product AdvancementSelect superior hybrids for:YieldStress toleranceGrain qualityAdaptationEnter best candidates into pre-commercial and national trial📊 Tools That Enable Acceleration BreedingTool            v.                BenefitDoubled Haploids Homozygosity in one stepGenomic Selection Predictive selection without phenotypingMarker-Assisted Selection   Early trait            l.   screeningOff-Season Nurseries Multiple generations/yearControlled Environments Fast generation cyclingSpeed Breeding Protocols Shorten growth duration using light & temp

𝗦𝗽𝗲𝗲𝗱 𝗯𝗿𝗲𝗲𝗱𝗶𝗻𝗴 + 𝗔𝗜 𝗳𝗼𝗿 𝗺𝗶𝗹𝗹𝗲𝘁𝘀 This is my hope 🌱 Two weeks ago we got the first good news from The International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) They developed Rapid Ragi speed breeding framework. _____________________________________ A brief pre-history. When I spoke recently with a millets grower, he reminded me of a simple, brutal truth in agriculture. One business cycle for him is one year. One experiment cycle for him is one year. And a person, even the luckiest one, gets just 50–70 of those cycles in a lifetime. That’s not much time. Especially if you’re trying to breed something better... Just for comparison: in fintech transaction takes seconds. They get yield every minute, literally. _____________________________________ However, it looks like we can make magic with time in agriculture too 😊 Surely, we can't grow plants in seconds! But a significant boost in productivity and time savings? Done! ✅ Rapid Ragi framework made it possible. It was developed specifically for finger millet. _____________________________________ The team behind Rapid Ragi optimized everything: → 9-hour photoperiods, 29°C controlled temperature, 70% humidity, high-density trays (105 plants per 1.5 sq. ft.), low-cost lighting, and targeted nutrition using Hoagland’s No. 2 solution. → They even discovered that harvesting at physiological maturity (not full maturity) could shave off extra days without compromising seed quality. _____________________________________ And this I call it "elegant science" Affordable, scalable and designed for real-world agriculture. What about AI in this framework? 1. AI can track phenology, predict trait performance, and automate decisions that usually take weeks. 2. It can analyze thousands of images, flag stress signals 3. It can help select the best lines faster than the eye or spreadsheet ever could. ____________________________________ Instead of summary: 1. Millets are among the world’s most climate-resilient and nutrient-dense crops. 2. I’m so glad that we’re moving — slowly but confidently — toward a future where they finally get the attention they deserve. 3. More research, more innovation, more tools like Rapid Ragi to accelerate progress - this is needed for all crops 🙏 What do you think about this innovative speed breeding framework?

NVIDIA Revolutionizes Climate Tech with ‘Earth-2’: The World’s First Fully Open Accelerated AI Weather Stack In a move that democratizes climate science, NVIDIA unveiled 3 groundbreaking new models powered by novel architectures: Atlas, StormScope, and HealDA. These tools promise to accelerate forecasting speeds by orders of magnitude while delivering accuracy that rivals or exceeds traditional methods. The suite includes three new breakthrough models: Earth-2 Medium Range: High-accuracy 15-day forecasts across 70+ variables.  Earth-2 Nowcasting: Generative AI that delivers kilometer-scale storm predictions in minutes.  Earth-2 Global Data Assimilation: Real-time snapshots of global atmospheric conditions. Full analysis: https://lnkd.in/gt_BugDZ Model weight: https://lnkd.in/gkUVqH5E Paper [Earth-2 Medium Range]: https://lnkd.in/gTf-f_Gd Paper [Earth-2 Nowcasting]: https://lnkd.in/gQf7muqz Paper [Earth-2 Global Data Assimilation]: https://lnkd.in/gu_-eZsn Technical details: https://lnkd.in/gPQ66Me2 NVIDIA NVIDIA AI