Power Budgets in Optical Communication Systems

from Aerosol Scattering

1

January 2026 1

Research interests:

• solar energy harvesting
• desalination/cooling
• drawing science from environmental data
• optical monitoring/sensing at cost and for emergency preparedness

Luat T Vuong (teaching optics)

Research interests:

• air quality
• atmospheric chemistry and physics
• aerosol-cloud interactions�

Markus Petters (clearing the air)

WCGEC workshop, Jan 29, 2026

Areas/expertise:

• optical interactions
• structured/complex light
• free-space optical sensing through/of air turbulence
• optical computing

Areas/expertise:

• aerosol-light interaction
• atmospheric turbulence
• aerosol measurements

​

​

​

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Turbulence affects power budgets for optical communications/sensing

Reference L. B. Stotts and L.C. Andrews, SPIE 2023

Transmitter

Turbulence

Receiver

Turbulence reduces signal strength:

Speckling of spatial beam profile

Propagation →

Most optical communication is digital on-off intensity modulation

Goal: understand aerosol power budget + mitigate with polarization structured light

2

January 2026 2

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Initial Plan: Experiments with aerosol chamber

Experiments are slow, but preliminary results indicate opportunities for turbulence and fire sensing

Could not measure aerosol effect:

• optical density too small,
• optical path too short

Modified chamber:

Specular (direct) transmission and side windows for measuring scattering

3

January 2026 3

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Pivot to Data Science Turbulence Prediction

Initially NYS data indicates humidity-dependent prediction accuracy.

4

January 2026 4

Combine ground-based NYS anemometer data with ERA-5 dataset

Predict C_n², atmospheric optical turbulence strength parameter

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

OTCliM Approach

OTCliM: ground truth are ground measurements; Satellite data presents multi-year weather variables

• Climatology Creation: From these predictions, the authors build climatologies:
• Seasonal/annual distributions of turbulence strength (top-right).
• Monthly summaries separated by stable vs unstable conditions (bottom-right).

​

• Model Predictions: The trained model is applied to the full ERA5 dataset (2018–2022), generating predicted multi-year C_n² time series.
• Extrapolated Time Series: The bottom panel shows the model’s extrapolated C_n² values, extending beyond the single year of measurements.

​

• ERA5 Inputs: Multi-year ERA5 weather variables (temperature, radiation, heat flux, wind shear). The blue box highlights the 1-year overlap period used for training
• Training Step: During the overlap year, a machine learning model is trained to link ERA5 variables → measured optical turbulence (C_n²)
• Ground Truth: Flux tower data provide the actual C_n² time series for that same year which is used as the “target” to fit the model.

5

January 2026 5

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Repeated OTCliM with CA anemometer data

Both models converge. CA more readily.

CA C_n² measurements are an order of magnitude larger than NYS

Anemometers at ~20 ground stations in both NYS and CA.

These measure wind flux/velocity to provide ground truth.

6

January 2026 6

New York State

California

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Feature importance and per-station comparison plots

January 2026 7

Surface heating (ΣR, QH) and wind shear (α, ΣX) are the most influential factors driving optical turbulence:

Both thermal and mechanical mixing dominate the model’s predictions across all stations

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

Conclusion and Future funding

Fire Science Innovations through Research and Education (NSF) deadline full proposal April

Combustion and Fire Systems and Atmosphere Cluster (NSF) deadline rolling

Physics of Sensing (AFOSR) deadline soon (Luat is currently at the Program Review for this)

Communications, Circuits, and Sensing-Systems (NSF) deadline rolling

Atmospheric Process Research (DOE) Whitepaper before March 6

“observational, laboratory experiments, and modeling research on atmospheric components such as clouds, aerosols, precipitation, and turbulence interactions”, predictive modeling.

Sensors and Surveillance Systems (NSWC Crane) Basic research; applied research; advanced technology development; and advanced component development and prototypes that facilitate the maturation of Sensors and Surveillance System technologies and manufacturing capabilities.

8

January 2026 8

Arial Tolentino

Travis Crumpton

Sarah Petters

Seed ideas, support also from AFSOR and CECERT

We achieve successful translation of turbulence prediction models with NY and CA, differences of which may relate environmental or sensing conditions.

​

Future work: time-series/geographic extensions, quantification of radiative transfer and surface heating, impact of turbulence on other events such as fire.

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols

New York State

California

CIMIS (California) vs Mesonet (NYS)

• NYS Mesonet
• Research-grade surface energy balance and turbulence observations
• Designed to support boundary-layer physics, flux studies, and micrometeorology
• Flux instruments mounted ~8–10 m AGL
• Fluxes computed every 30 minutes on-site
• Physics complete but spatially sparse
• Sited to sample diverse land use and terrain types

​

• Cal CIMIS
• Operational network optimized for irrigation management and water use
• Emphasis on agricultural representativeness and spatial coverage
• Standard meteorological sensors only
• No direct turbulence or flux measurements
• Sensors sampled every minute but stored as hourly averages or totals
• Spatially dense and operationally robust
• Stations are intentionally sited in open, irrigated grass or agricultural settings to minimize obstruction
• Surroundings are actively maintained
• Grass kept short (~7–8 cm)
• Irrigated and fertilized to maintain a uniform reference surface

• Similarities
• Core meteorology (Air temp, Rel humidity, Wind speed, wind direction, Solar radiation, Precipitation, & Soil temperature)
• Long-term continuity
• Station-specific metadata (location, land cover)
• Designed for public and research use

Energy Systems Research Workshop – Seed Grants

Luat T Vuong and Markus Petters

Optical Power Budgets: Turbulence and Aerosols