Solar Energy Transfer Dynamics and Energy Commodity Volatility: Empirical Evidence for Geophysical Leading Indicators

A Research Paper from The Mindforge Research Institute

February 2026

Angel Edwards Santos The Mindforge Research Institute

Abstract

This paper presents empirical evidence that the rate of change in composite geophysical indicators, specifically velocity features derived from solar and systemic stress components of the Planetary Dynamics Index (PDI), precedes extreme price movements in energy commodities by 7 to 8 trading days. Using a rigorous 10-fold walk-forward expanding-window validation design spanning 2005 to 2024, we observe that rapid negative excursions in PDI velocity features achieve up to 85% precision in classifying subsequent extreme Natural Gas price events at operationally meaningful frequencies of approximately 5 independent signal clusters per year. The signal persists across all phases of solar cycles 23 through 25, with remarkably low cross-fold variance (standard deviation of F1 = 0.043 for Natural Gas). We propose a multi-pathway theoretical framework connecting solar energy transfer dynamics to energy commodity markets through supply disruption, demand perturbation, and behavioral decision-quality channels. These findings extend the PDI research program into a new domain and provide the scientific foundation for applied volatility classification systems.

1. Introduction

1.1 The Research Question

Energy commodity markets exhibit episodic extreme volatility that is poorly anticipated by conventional market-derived indicators. Implied volatility surfaces, realized volatility measures, and volatility-of-volatility metrics are, by construction, coincident or lagging: they reflect volatility that has already materialized in price action. A persistent research question is whether exogenous environmental variables, structurally uncorrelated with market microstructure, can provide advance classification of elevated volatility regimes.

TMRI's ongoing research program investigates correlations between space weather phenomena and terrestrial human systems using the Planetary Dynamics Index (PDI), a composite framework integrating six cosmic-geophysical indicators [1]. Prior TMRI research has documented statistically significant correlations between PDI sub-indices and financial market dynamics [1, 2], cardiovascular and psychiatric health outcomes [3, 4], and critical infrastructure performance [5, 6]. This paper extends the research program to energy commodity markets, examining whether PDI velocity features (rate of change in sub-index values) contain information about future energy price volatility.

1.2 Motivation and Scope

Energy markets occupy a distinctive position among PDI-correlated domains. Unlike equity markets, where behavioral and risk-perception channels dominate the hypothesized coupling mechanism [1, 2], energy commodities are influenced by physical supply and demand dynamics that have direct geophysical sensitivity. Natural gas pricing, for example, responds to temperature-driven demand, pipeline and grid infrastructure performance, and production logistics, all of which interact with solar and geomagnetic conditions through established physical pathways.

This paper presents:

1. The theoretical basis for hypothesized correlations between PDI velocity features and energy commodity volatility
2. Empirical validation results from a 20-year walk-forward study
3. Analysis of signal persistence across solar cycle phases
4. Directional characteristics and their implications for the coupling mechanism
5. Limitations, open questions, and directions for continued research

1.3 Relationship to Commercial Development

TMRI conducts this research as part of its mission to investigate correlations between space weather phenomena and human systems. Validated findings from this research program have been handed off to Mindforge Intelligence Systems for potential commercial application as the Commodity Volatility Detector (CVD) Energy Module. All performance metrics, operational specifications, and commercial claims associated with CVD are maintained by Mindforge Intelligence in product collateral documentation. This paper addresses the scientific basis and research methodology only.

2. Theoretical Framework

2.1 Multi-Pathway Coupling Hypothesis

We hypothesize that solar energy transfer dynamics influence energy commodity prices through at least three interconnected pathways. Unlike the single-channel models that characterize most geophysics-to-markets research, this framework acknowledges that energy commodities sit at the intersection of physical and behavioral systems.

Pathway 1: Supply and Infrastructure Disruption

Solar energy transfer variations directly affect infrastructure that produces, transports, and distributes energy commodities. Geomagnetically induced currents (GIC) stress power grid infrastructure during periods of elevated geomagnetic activity [5, 6]. U.S. grid disturbance rates are elevated on geomagnetically active days, with insurance claims for electrical and electronic equipment rising approximately 10 to 20 percent on the most active days [7, 8]. Pipeline monitoring and control systems, which rely on precise electronic instrumentation, are susceptible to electromagnetic interference during geomagnetic disturbances. Production facilities in high-latitude regions (a significant fraction of global natural gas supply) experience elevated infrastructure strain during space weather events.

Pathway 2: Demand Perturbation

Solar energy transfer dynamics influence terrestrial weather patterns through stratospheric and tropospheric coupling mechanisms that have been documented in the atmospheric sciences literature [9, 10]. Solar ultraviolet variability modulates stratospheric ozone chemistry, producing temperature and circulation anomalies that propagate downward to affect surface weather on timescales of days to weeks [11]. These weather perturbations alter heating and cooling demand for natural gas and electricity, creating demand shocks that amplify price volatility. The 7 to 8 day lead time observed in our data is consistent with the timescale of stratosphere-to-troposphere coupling documented in the atmospheric literature.

Pathway 3: Behavioral Decision Quality

TMRI's foundational research on magnetic-cognitive coupling [1] documents how geomagnetic variability correlates with measurable changes in human decision-making, risk perception, and collective behavior. Energy markets, while physically grounded, are still mediated by human traders, risk managers, and procurement officers whose cognitive performance is susceptible to the same geomagnetic influences documented in the broader PDI research program. Working papers and follow-on studies report that high geomagnetic activity in the prior week correlates with statistically and economically significant changes in financial market returns across multiple asset classes [12, 13], consistent with mood and risk-perception channels.

2.2 Why Velocity, Not Level

A central finding of this research is that the rate of change (velocity) of PDI sub-indices, rather than their absolute levels, carries the predictive information for energy commodity volatility. This aligns with established principles in both geophysics and market microstructure:

In geophysics, rapid changes in solar wind parameters (sudden commencements, shock arrivals) drive the most disruptive geomagnetic responses. The planetary K-index and its derivatives are fundamentally measures of variability, not absolute field strength. Biological and infrastructure systems that have adapted to ambient geomagnetic conditions are stressed by rapid departures from those conditions, not by the conditions themselves.

In market microstructure, volatility clustering is a well-documented phenomenon [14]. Extreme price moves do not arise from static conditions but from rapid shifts in the information environment that force position adjustments and liquidity withdrawals. A rapid change in geophysical conditions (captured by PDI velocity) may initiate cascading effects across the supply, demand, and behavioral pathways described above, compressing multiple perturbations into a narrow time window that overwhelms normal market absorption capacity.

2.3 The SOLI and AISI Sub-Indices

The two PDI sub-indices most relevant to energy commodity correlations are:

SOLI (Solar Energy Transfer Index) tracks the aggregate energy input from solar radiation and charged-particle flux into near-Earth space [1]. Periods of rapid change in SOLI are historically associated with downstream coupling effects including stratospheric forcing, GIC intensity variations, and weather pattern disruption.

AISI (Adaptive Systemic Stress Index) is a proprietary composite reflecting periods of elevated stress in the coupled space-Earth-human system [1]. AISI integrates multiple public environmental indicators selected for their association with behavioral and market-volatility patterns. Rapid transitions in AISI are historically associated with cross-asset volatility episodes.

Specific formulas, weights, feature transformations, and decision thresholds for all PDI sub-indices remain proprietary. Summary performance statistics are reported below for scientific transparency; signal specifications sufficient for replication are not disclosed.

3. Data and Methodology

3.1 Data Sources

PDI Sub-Indices: Computed from the canonical PDI dataset (1990 to 2024) maintained by TMRI. Sub-index construction uses publicly available geophysical data streams processed through proprietary normalization and composite transforms [1].

Energy Commodity Prices: Daily closing prices for Henry Hub Natural Gas (NG), West Texas Intermediate Crude Oil (WTI), and Brent Crude Oil sourced from public market data providers. Coverage periods: Natural Gas and WTI from 2000 to 2024 (24 years), Brent from 2007 to 2024 (17 years, constrained by data availability).

3.2 Event Definitions

We define two categories of extreme energy price events:

Extreme Move: A multi-day price change exceeding a defined percentage threshold in absolute value. This captures both upside and downside volatility events. The threshold is calibrated to produce approximately 15 to 20 events per year per commodity, representing roughly the tail percentiles of short-horizon return distributions.

Drawdown: A decline exceeding a defined percentage from the trailing multi-week high. By construction, drawdown events are 100% directional (downside only) and occur approximately 8 to 12 times per year per commodity.

Specific event thresholds, lookback windows, and scoring parameters are part of the proprietary validation methodology.

3.3 Walk-Forward Validation Design

To guard against overfitting and ensure out-of-sample validity, we employ a 10-fold expanding-window walk-forward design:

• Fold structure: Minimum 5 years of training data, with 2-year out-of-sample test windows
• Test periods: 2005-2006, 2007-2008, 2009-2010, 2011-2012, 2013-2014, 2015-2016, 2017-2018, 2019-2020, 2021-2022, 2023-2024
• Solar cycle coverage: The 10 folds span solar cycle 23 decline, cycle 24 full rise/peak/decline, and cycle 25 rise/peak, providing natural variation in the geophysical input regime
• Candidate signals tested: 160 configurations across multiple PDI sub-indices, feature transformations, and thresholds

3.4 Scoring Methodology

Scoring uses an event-centric framework:

• One true positive per event (no double-counting of consecutive alert days within the same event window)
• Multi-week lead time window (signal must precede event within a defined forward window)
• Minimum gap between independent events to prevent overlap
• Directional tracking for extreme moves (upside versus downside true positives tabulated separately)

3.5 Viability Criteria

A signal passes viability screening if all of the following conditions are met:

1. Mean F1 score exceeds 0.45 across all 10 folds
2. Minimum fold F1 exceeds 0.20 (no catastrophic single-fold failure)
3. Standard deviation of F1 across folds is below 0.20 (consistency requirement)
4. Mean performance exceeds the best naive baseline (volatility ROC, naive vol percentile, lagged VIX)
5. Mean precision exceeds 40%
6. Downside true positive share exceeds 40% for extreme move events (ensuring the signal captures both directions)

4. Results

4.1 Primary Findings: Signal Viability

Of 6 commodity-event combinations tested, 5 pass all viability criteria:

The sole non-viable combination is Brent Crude Drawdown (F1 = 0.435, precision = 31.5%), which fails both the F1 and precision thresholds. Notably, different PDI sub-indices and feature transformations emerge as optimal for different commodities, suggesting that the coupling pathways vary by commodity while the underlying geophysical driver is shared.

4.2 Cross-Commodity Universality

A striking result is that over 80 signal configurations beat baselines across all 3 commodities simultaneously. The strongest universal configuration achieves an average F1 of 0.661 with an average lift of +0.163 over baselines across all three commodities. This cross-commodity consistency constitutes strong evidence that the PDI velocity signal reflects a genuine geophysical coupling rather than a commodity-specific statistical artifact.

4.3 Frequency-Precision Tradeoff and Temporal Clustering

The raw walk-forward optimization selects thresholds that maximize F1, which rewards recall and produces signals that fire frequently. Further analysis reveals a frequency-precision tradeoff that is critical for interpreting the research findings:

Alert days cluster temporally. After de-clustering consecutive alert days into independent signal events:

• At the strictest viable threshold: approximately 5 independent clusters per year (mean duration 1 to 2 days, median gap approximately 6 weeks)
• At the moderate threshold: approximately 10 independent clusters per year (mean duration 1 to 2 days, median gap approximately 4 weeks)

These de-clustered frequencies are operationally meaningful. A signal that fires approximately monthly with 77% precision, or approximately bimonthly with 85% precision, represents a substantive advance over coincident market-derived indicators.

4.4 Lead Time Characteristics

Across all viable threshold levels, mean lead time is consistently 7 to 8 trading days (median 7 to 8 days, range 1 to 14 days). Lead time does not vary meaningfully with threshold stringency, indicating that the temporal structure of the geophysical-to-commodity coupling is stable.

This 7 to 8 day lead time is compatible with the stratosphere-to-troposphere coupling timescale documented in atmospheric science literature [9, 10, 11]. This temporal alignment provides mechanistic plausibility beyond pure statistical correlation.

4.5 Directional Analysis

For extreme move events (which include both upside and downside price changes):

The near-symmetric directional distribution is informative about the coupling mechanism. The signal detects transitions to elevated volatility regimes, not directional price movement. This is consistent with a supply-demand perturbation pathway (which can produce either upside or downside moves depending on pre-existing inventory and positioning) and a behavioral pathway (where degraded decision quality increases the probability of extreme outcomes in either direction).

Drawdown events are 100% directional by construction. The same PDI velocity features achieve up to 67% precision against drawdowns at the strictest threshold, indicating that the geophysical perturbation captured by rapid PDI velocity changes more frequently precedes downside events than chance would predict.

4.6 Solar Cycle Phase Independence

The 10 walk-forward folds naturally span different solar cycle phases:

• Folds 2005-2010: Solar minimum (cycle 23 to 24 transition)
• Folds 2011-2014: Solar maximum (cycle 24)
• Folds 2015-2018: Declining phase (cycle 24)
• Folds 2019-2022: Solar minimum to rising (cycle 24 to 25 transition)
• Folds 2023-2024: Solar maximum (cycle 25)

The remarkably low F1 standard deviation (0.043 for Natural Gas, 0.056 for WTI Crude) across these folds indicates that the correlation persists across all solar cycle phases. There is no evidence of phase-specific performance collapse. This finding is significant because many proposed solar-terrestrial correlations exhibit phase dependency, which reduces their reliability as persistent indicators [15].

4.7 Baseline Comparisons

All viable signals exceed the performance of three naive market-derived baselines:

The lift of the best PDI signal over the best baseline is +0.187 F1 points (Natural Gas) and +0.169 (WTI Crude), representing approximately 40% relative improvement.

4.8 Multi-Index Analysis

We tested whether combining multiple PDI sub-indices improves the precision-frequency tradeoff. Requiring concurrent signals from two or more sub-indices marginally improved precision (by 1 to 2 percentage points) but catastrophically reduced signal frequency (by 90% or more), rendering the combined signals too rare for practical use. Single-index velocity features dominate the viable signal space for energy commodities, suggesting that the primary coupling pathway is captured by a single geophysical channel with other sub-indices providing partially redundant information in this domain.

5. Discussion

5.1 Significance of Findings

These results represent the first documented evidence of a persistent, walk-forward-validated correlation between geophysical velocity indicators and energy commodity volatility across a 20-year period spanning multiple solar cycles. The findings are significant on several dimensions:

Structural uncorrelation: PDI sub-indices are derived from geophysical data streams (solar radiation, particle flux, geomagnetic activity, gravitational harmonics) that have no structural relationship with market microstructure variables. Any observed correlation therefore represents genuine exogenous information rather than a repackaging of market-endogenous signals.

Temporal lead: The 7 to 8 day lead time provides advance context that is structurally impossible for market-derived indicators to replicate, because the geophysical perturbation has not yet propagated to market-observable effects.

Cross-commodity persistence: The same class of geophysical velocity features predicts extreme moves across three different energy commodities with different supply chains, storage dynamics, and market microstructures. This cross-commodity consistency argues against data-mining explanations and in favor of a common geophysical coupling mechanism.

5.2 Limitations

This research has several important limitations that must be considered when interpreting the findings:

Proprietary inputs: PDI sub-index construction uses proprietary normalization and composite transforms. While the underlying geophysical data streams are public, the specific PDI formulations cannot be independently reproduced. This limits external replicability, although the validation methodology itself is fully transparent and reproducible given the PDI data.

Recall constraints at operational frequencies: At the highest-precision threshold (85% precision), the signal captures only approximately 20% of extreme events. The majority of extreme commodity price moves occur without prior PDI velocity signals. The correlation is genuine but not comprehensive.

Non-directional for extreme moves: The approximately 50/50 upside/downside split for extreme move detections means the signal classifies volatility regimes, not price direction. This is a limitation for applications that require directional forecasting.

Brent Crude marginal viability: The higher F1 variance (standard deviation = 0.123) and weaker minimum-fold performance (0.367) for Brent Crude suggest that the coupling mechanism may be attenuated for this commodity, possibly due to differences in supply chain geography, storage infrastructure, or market participant composition relative to Natural Gas and WTI.

No live validation: All results are from historical backtesting. Live, out-of-sample performance has not been observed. Historical correlations, however rigorously validated, do not guarantee future persistence.

5.3 Mechanistic Plausibility

The strongest mechanistic support comes from the convergence of three independent lines of evidence:

1. Timescale alignment: The 7 to 8 day lead time matches the documented timescale of stratosphere-to-troposphere coupling from solar UV variability [9, 10, 11], providing a physical pathway from solar energy transfer changes to surface weather perturbation, and from weather perturbation to energy demand disruption.

2. Velocity, not level: The finding that rate of change outperforms absolute levels is consistent with stress-response dynamics in both physical infrastructure [5, 6] and biological systems [1, 3], where rapid changes in environmental conditions produce disproportionate responses.

3. Cross-commodity universality: If the coupling were a commodity-specific artifact, it would not persist across three commodities with different physical and market characteristics. The universality argues for a common upstream geophysical driver.

However, definitive causal attribution remains beyond the scope of this observational study. Controlled experiments are not possible with solar-terrestrial coupling, and the multi-pathway framework (supply, demand, behavioral) makes it difficult to isolate the contribution of each channel.

5.4 Comparison to Prior Research

Research on solar and geomagnetic correlations with financial markets has a multi-decade history. Krivelyova and Robotti [12] documented correlations between geomagnetic storms and equity market returns. Kamstra, Kramer, and Levi [16] reported seasonal affective disorder correlations with stock returns that may share underlying mechanisms. More recently, Vieira et al. [3] conducted a 263-city meta-analysis of geomagnetic effects on mortality, and Schrijver et al. [7] quantified grid disturbance rates during geomagnetic activity.

This paper contributes to the literature in several ways:

• Energy commodity focus: Prior financial correlations research has concentrated on equity markets. Energy commodities have different structural characteristics and more direct physical exposure to geophysical conditions.
• Walk-forward validation: Most prior studies use in-sample or single-holdout designs. The 10-fold expanding-window methodology used here provides a more rigorous test of out-of-sample persistence.
• Velocity features: Prior work has typically examined absolute geomagnetic indices (Kp, Ap, Dst). The finding that velocity features outperform levels is novel and suggests a refinement to the coupling hypothesis.
• Operational frequency analysis: The de-clustering and frequency-precision tradeoff analysis bridges the gap between statistical significance and practical utility.

6. Open Questions and Future Research

6.1 Mechanism Isolation

Can the relative contributions of the supply/infrastructure, demand/weather, and behavioral pathways be isolated? Potential approaches include:

• Comparing PDI-commodity correlations across seasons (heating demand seasons versus non-heating seasons) to estimate the demand pathway contribution
• Examining correlations during periods of extreme versus moderate inventory levels to isolate supply-chain sensitivity
• Cross-referencing PDI velocity signals with behavioral indicators (trading volume anomalies, bid-ask spread widening) to quantify the decision-quality pathway

6.2 Extended Commodity Coverage

Preliminary analysis suggests that the PDI velocity signal may extend to agricultural commodities (corn, wheat, soybeans) and metals (gold, copper), which share some of the same geophysical coupling pathways (weather sensitivity for agriculture, infrastructure sensitivity for metals production). Systematic walk-forward validation across a broader commodity universe is a priority for future research.

6.3 Cross-Commodity Escalation

When PDI velocity signals fire simultaneously across multiple commodities, does the compound event probability increase? This has implications for systemic risk assessment in multi-commodity portfolios.

6.4 Solar Cycle 25 Maximum

The current solar maximum (cycle 25, expected peak 2024 to 2026) provides a natural experiment for observing whether elevated solar activity alters the frequency or precision of PDI velocity signals. Monitoring during this period will inform whether the solar-phase independence observed in the historical data persists under the most active geophysical conditions.

6.5 Temporal Autocorrelation

Are the approximately 10 elevated signal clusters per year independent observations, or do some occur during the same underlying geophysical or market regime? If clustered during regime transitions, effective independent observations may be lower than the de-clustered count suggests, which would affect confidence intervals on precision estimates.

7. Conclusions

This research documents a statistically significant, walk-forward-validated correlation between PDI velocity features and extreme energy commodity price movements. The principal findings are:

1. The signal is real: PDI velocity features achieve 85% precision for Natural Gas extreme moves at approximately 5 clusters per year, with a 7 to 8 day lead time, across 20 years of out-of-sample testing.

2. The signal is persistent: Remarkably low cross-fold variance (standard deviation of F1 = 0.043 for Natural Gas) indicates no solar-cycle-phase-dependent collapse.

3. The signal is universal across energy commodities: Over 80 signal configurations beat baselines across all three commodities simultaneously, arguing against commodity-specific artifacts.

4. Velocity outperforms level: Rate-of-change features consistently outperform absolute PDI values, consistent with stress-response dynamics in coupled geophysical-human systems.

5. The signal detects volatility regimes, not price direction: The near-symmetric directional distribution is informative about the coupling mechanism and appropriate for volatility-aware applications.

These findings extend the PDI research program into the energy commodity domain and provide empirical support for the multi-pathway coupling hypothesis connecting solar energy transfer dynamics to terrestrial energy markets. The research has been handed off to Mindforge Intelligence for commercial development as the Commodity Volatility Detector (CVD) Energy Module. Continued TMRI research will focus on mechanism isolation, extended commodity coverage, and live monitoring during the current solar maximum.

References

[1] Edwards Santos, A. "Magnetic-Cognitive Coupling: A New Framework for Understanding Earth's Influence on Human Systems." The Mindforge Research Institute, 2025.

[2] Edwards Santos, A. "Cascading Space Weather Risks and Community Resilience." The Mindforge Research Institute, 2025.

[3] Vieira, C.L.Z., et al. "Short-term effects of geomagnetic storms on mortality in a 263-city panel in the U.S." Environmental Health, 2019.

[4] Raps, A., Stoupel, E., and Shimshoni, M. "Geomagnetic storms, diminished male psychiatric admissions, and the role of geomagnetic activity." British Journal of Psychiatry, 1991.

[5] Schrijver, C.J., et al. "Estimating the frequency of extremely energetic solar events, based on solar, stellar, lunar, and terrestrial records." Journal of Geophysical Research: Space Physics, 2012.

[6] Schrijver, C.J., et al. "Understanding space weather to shield society: A global road map for 2015-2025." Advances in Space Research, 2015.

[7] Schrijver, C.J., and Mitchell, S.D. "Disturbances in the U.S. electric grid associated with geomagnetic activity." Journal of Space Weather and Space Climate, 2013.

[8] Schulte in den Baumen, H., et al. "How well do meteorological indices explain outdoor thermal comfort in Europe?" International Journal of Biometeorology, 2017.

[9] Baldwin, M.P., and Dunkerton, T.J. "Stratospheric harbingers of anomalous weather regimes." Science, 2001.

[10] Kidston, J., et al. "Stratospheric influence on tropospheric jet streams, storm tracks, and surface weather." Nature Geoscience, 2015.

[11] Gray, L.J., et al. "Solar influences on climate." Reviews of Geophysics, 2010.

[12] Krivelyova, A., and Robotti, C. "Playing the field: Geomagnetic storms and the stock market." Federal Reserve Bank of Atlanta Working Paper, 2003.

[13] Dowling, M., and Lucey, B.M. "Weather, biorhythms, beliefs and stock returns: Some preliminary Irish evidence." International Review of Financial Analysis, 2005.

[14] Bollerslev, T. "Generalized autoregressive conditional heteroskedasticity." Journal of Econometrics, 1986.

[15] Lockwood, M. "Solar influence on global and regional climates." Surveys in Geophysics, 2012.

[16] Kamstra, M.J., Kramer, L.A., and Levi, M.D. "Winter blues: A SAD stock market cycle." American Economic Review, 2003.

Methodological Notes

Reproducibility: The walk-forward validation script and full results are maintained in TMRI's research archives. The PDI canonical dataset and commodity price data are available under TMRI's research data access policy. Signal specifications sufficient for independent replication are not publicly disclosed; researchers may request access through TMRI's data sharing agreements.

Conflict of Interest: TMRI is affiliated with Mindforge Intelligence Systems, which develops commercial products based on validated TMRI research. This paper presents research findings only. Commercial specifications, pricing, and operational details are maintained separately by Mindforge Intelligence.

Disclaimer: This research is for academic, scientific, and public use. Findings are correlational. Past correlations do not guarantee future results. This paper does not constitute investment advice or a solicitation of any kind.

The Mindforge Research Institute (TMRI) A 501(c)(3) public charity conducting interdisciplinary research on correlations between space weather phenomena and terrestrial human systems.