Geophysical Forecasting Techniques

Continue discussing Seismic prospecting -significant of Seismic wave frequency Seismic wave frequency is used to predict porosity by relating how frequency-dependent attenuation and velocity changes are affected by the presence of pores. Higher frequencies are more sensitive to the rock matrix, while lower frequencies are more strongly influenced by the fluid in pores, allowing for the use of different frequency bands to model porosity distribution, especially when combined with machine learning and geostatistical methods. This is achieved through techniques like seismic inversion and analysis of attributes like acoustic impedance, Vp/Vs ratio, and shear modulus, which are sensitive to porosity variations. How frequency relates to porosity High vs. low frequencies: High-frequency seismic waves tend to travel through the rock matrix, while low-frequency waves can travel through the pore space. This difference in behavior is due to seismic wave dispersion caused by the presence of pores. Pore fluid interaction: The velocity of the seismic waves is sensitive to the fluid properties within the pores, and at certain frequencies, wave propagation is influenced by the flow of fluids between the pores and the surrounding matrix. Seismic attenuation: The presence of pores can cause seismic waves to lose energy through attenuation. The degree of attenuation is frequency-dependent, meaning the relationship between frequency and attenuation can be used to infer porosity. Methods to predict porosity using seismic frequency Seismic inversion: This technique converts seismic reflection data into a more quantitative representation of subsurface properties. Inversion can be used to predict properties like acoustic impedance, which can then be correlated with porosity. Machine learning: Machine learning models, such as convolutional neural networks (CNNs), can be trained to recognize complex, non-linear relationships between various seismic attributes (including frequency-dependent ones) and porosity from well log data. Geostatistical analysis: This combines statistical methods with geological information. It uses relationships between petrophysical parameters (like porosity) and seismic attributes to build a more geologically constrained porosity model. Rock physics templates: These are used to create relationships between seismic properties and rock properties like porosity, based on known rock physics principles. They help in creating basin and lithology-dependent templates for porosity prediction. Key takeaway A single frequency is not enough; a combination of techniques and frequency bands is needed to predict porosity. Integrating seismic data with other information like well logs and geostatistical models improves the accuracy of porosity predictions. Machine learning is a powerful tool that can learn complex relationships between seismic attributes and porosity that are difficult to model with traditional methods.

Physically Consistent Forecasting Critical time series cannot rely on statistical correlation alone. When regimes shift, purely data-driven models collapse. In real systems, physics remains valid. Technical Architecture - Input Multivariate historical data + structural physical variables - Pinneaple Pipeline: . pinneaple_timeseries (temporal engineering and physics-aware feature preparation) . pinneaple_models (hybrid backbone: temporal dynamics + physical structure) . pinneaple_solvers (structural constraints and conservation laws enforcement) . pinneaple_train (combined loss: predictive error + physical violation penalty) Output Robust forecasts with stability under distribution shifts. Use Case Load forecasting in electrical grids under extreme events. Heat waves, substation failures, abrupt consumption spikes. Purely statistical models extrapolate poorly. The hybrid model: - Learns temporal patterns - Respects system physical limits - Penalizes structural violations during training - Maintains stability under extreme regimes The Principle - pinneaple_models defines the hybrid dynamics. - pinneaple_train enforces structural consistency during learning. When statistics fail, physics anchors the model. This is not just forecasting. It is constraint-aware engineering.

Diffusion Models for Weather Forecasting 🔭 Numerical Weather Prediction is the dominant approach to weather forecasting. It solves large systems of partial differential equations and relies on high-fidelity physics-based simulations. Recently, Machine Learning-based Weather Prediction has emerged as a competitive alternative, with several models surpassing individual NWP systems. However, they still fall short of the performance of NWP ensembles. Ensembles simulate multiple physically plausible trajectories and explicitly represent forecast uncertainty, while most machine learning models collapse this uncertainty into point estimates such as the mean. This typically produces over-smoothed or blurry predictions. In generative modelling, the most effective way to learn a full data distribution is through diffusion models, which denoise Gaussian noise step by step to generate samples that follow the underlying distribution. This provides a natural direction for weather forecasting. Starting from Gaussian noise defined on a spherical domain, the model can iteratively denoise to obtain a physically consistent weather state conditioned on past observations. To capture spatial dependencies across the globe, a graph transformer operating on a spherical mesh is used to extract features from K-hop neighborhoods. This framework has demonstrated performance that exceeds existing ensemble-based methods.

New paper – A foundation model for the Earth system Abstract “Reliable forecasting of the Earth system is essential for mitigating natural disasters and supporting human progress. Traditional numerical models, although powerful, are extremely computationally expensive. Recent advances in artificial intelligence (#AI) have shown promise in improving both predictive performance and efficiency, yet their potential remains underexplored in many Earth system domains. Here we introduce Aurora, a large-scale foundation model trained on more than one million hours of diverse geophysical data. Aurora outperforms operational forecasts in predicting air quality, ocean waves, tropical cyclone tracks and high-resolution #weather, all at orders of magnitude lower computational cost. With the ability to be fine-tuned for diverse applications at modest expense, Aurora represents a notable step towards democratizing accurate and efficient Earth system predictions. These results highlight the transformative potential of AI in environmental forecasting and pave the way for broader accessibility to high-quality #climate and #weather information.” Bodnar, C., Bruinsma, W.P., Lucic, A. et al. A foundation model for the Earth system. Nature 641, 1180–1187 (2025). https://lnkd.in/eh8wQ2wx

Volumetric Method Principle: Estimates hydrocarbons in place (STOIIP/GIIP) based on the reservoir’s geometry, porosity, saturation, and formation volume factor. Applies before production begins (static method). Strengths: Useful in early field life (before production data). Straightforward and quick. Requires geological and petrophysical data. Weaknesses: Accuracy depends on data quality (porosity, thickness, area). Assumes uniformity—doesn't capture heterogeneity or compartmentalization. Does not account for reservoir connectivity. 🔍 2. Material Balance Method (MBE) Principle: Uses the law of conservation of mass to estimate Original Hydrocarbon in Place (OHIP) by relating cumulative production to pressure depletion. Strengths: Applicable after some production data is available. Good for estimating drive mechanisms. Integrates PVT and production data. Weaknesses: Assumes average reservoir pressure is known accurately. Requires reliable PVT data. Sensitive to aquifer behavior assumptions. 🔍 3. Decline Curve Analysis (DCA) Principle: Projects future production using historical trends (rate-time data), assuming reservoir behavior remains consistent. Types include: Exponential Harmonic Hyperbolic Strengths: Simple and fast. Requires only production data. Effective in mature reservoirs. Weaknesses: Poor prediction in early life or unstable production. Doesn’t directly estimate hydrocarbons in place. Assumes constant operating conditions and no interventions. 🔍 4. Reservoir Simulation (Numerical Modeling) Principle: Uses mathematical models and computer simulations to predict reservoir performance under different scenarios. Integrates geology, petrophysics, PVT, SCAL, and production history. Strengths: Handles complex reservoir geometries. Simulates different development strategies. Powerful for optimization and forecasting. Weaknesses: Data- and labor-intensive. Requires skilled personnel and calibration. Can produce misleading results if poorly constrained. 🔍 5. Analog/Analytical Models Principle: Estimates reserves by comparing with similar, previously developed fields (analogs). Strengths: Quick and low cost. Useful for frontier areas with little data. Weaknesses: Assumes similarity—can be misleading. Not suitable for unique or heterogeneous reservoirs. 🔍 6. Probabilistic Methods (Monte Carlo Simulation) Principle: Applies probability distributions to input variables (porosity, saturation, area, etc.) to generate a range (P90, P50, P10) of reserves. Strengths: Accounts for uncertainty. Provides risk-based estimates. Useful for decision-making and portfolio management. Weaknesses: Requires proper input distributions. Computational resources needed. Can give false confidence if assumptions are wrong.

Geotechnical and Geophysical Engineering: Seeing Below the Surface Geotechnical engineers often rely on physical testing and boreholes to understand the subsurface. But sometimes, we need a broader picture—what’s happening between boreholes, over a larger area, or at depths that are difficult to reach. That’s where geophysics comes in. Here’s what geotechnical engineers should understand about working with geophysicists—and what geophysicists should know in return. 1. Different Tools, Same Goal Geophysics uses indirect methods—like seismic, resistivity, GPR, or electromagnetic surveys—to map subsurface conditions. Geotechnical engineering uses direct testing—like CPTs, SPTs, and boreholes. They’re not competing approaches. They’re complementary. 2. Where Geophysics Makes a Real Difference Geophysics is particularly useful when: Bedrock depth varies across a site (e.g., sloping or dipping bedrock beneath a building or bridge) The contact between soil and rock is unclear, but critical for deep foundations There’s a need to locate voids, buried structures, or utility conflicts Large infrastructure spans an area where borehole coverage is limited You need to estimate dynamic soil properties (e.g., Vs30 for seismic design) In these cases, geophysics can save time, reduce uncertainty, and guide borehole placement more effectively. 3. What Geotechnical Engineers Should Understand About Geophysics Geophysical results are indirect and require interpretation, not just measurement The data is resolution-limited and may smooth over sharp soil transitions Geophysicists need input from boreholes or known stratigraphy to calibrate models correctly Simply asking for a “bedrock map” without defining what you mean by "bedrock" can create confusion 4. What Geophysicists Should Understand About Geotechnical Practice Engineers are looking for design parameters, not just material boundaries A resistivity or seismic velocity change is interesting—but needs to translate to something usable: “This is where we hit refusal” or “This is where side resistance develops.” Resolution near the surface matters—especially for shallow foundations, pavement design, and utility coordination Close collaboration during survey planning improves alignment between what’s collected and what’s actually needed 5. Start the Discussion Early Some of the best results happen when geotechnical engineers bring in geophysicists during the planning stage of site investigations—not as a troubleshooting tool after borehole data is unclear. This allows both sides to frame the right questions and use resources efficiently. Have you used geophysics to refine or adjust your geotechnical designs? Or seen gaps between geophysical output and what the design team needed?