Atmospheric science,
open access.

Papers, reports, and research outputs from LADAR field campaigns — freely available to researchers, institutions, and the public.

No publications yet. LADAR is currently in its founding phase. Field campaigns have not yet commenced. This page will be populated with research outputs as the platform becomes operational. Research areas and methodology are outlined below.

Papers & reports coming

All research outputs will be published here as open-access documents. No paywalls. No embargo periods.

Follow progress

What we study.

LADAR's research spans fundamental atmospheric physics and applied environmental studies. The platform is deliberately flexible — enabling a wide range of investigations from a single deployable system.

3D Thermal Mapping

Vertical profiling and spatial mapping of temperature fields. Investigations into urban heat islands, thermal plumes, inversion layers, and surface–atmosphere heat exchange.

Stack 1, Stack 4

Atmospheric Stability

Computation of Richardson numbers, detection of mixing layers, identification of stable and unstable stratification regimes. Understanding turbulence onset and suppression.

Stack 1

Flow Field Estimation

Inference of wind vectors from drone motion and IMU data for turbulence studies. Direct measurement with ultrasonic anemometry for wind shear detection and vorticity approximation.

Stack 1, Stack 3

Pollutant Dispersion

Mapping of PM2.5, PM10, CO₂, NO₂, and O₃ distributions. Advection–diffusion modelling, source localisation, and time-varying exposure reconstruction in urban environments.

Stack 2

ML Field Reconstruction

Physics-informed machine learning applied to sparse in-situ measurements to generate continuous volumetric representations of temperature, pollutant, and airflow fields.

All stacks

Event Capture

Adaptive sampling of extreme or transient atmospheric phenomena: heatwaves, storm fronts, inversion events, and rapid pollutant episodes. Onboard edge intelligence enables rapid response deployment.

Stack 5

How we work.

Physics-informed modelling

Rather than treating reconstruction as a purely data-driven problem, LADAR integrates known atmospheric physics — heat transfer equations, Navier–Stokes constraints, advection–diffusion dynamics — as regularisation terms within the machine learning pipeline. This improves reconstruction fidelity when measurement density is low.

Sparse-to-dense field reconstruction

Flight paths are planned to maximise spatial coverage within endurance constraints. Collected point measurements are then used as input to interpolation and ML reconstruction algorithms to generate continuous volumetric field estimates with associated uncertainty bounds.

Reproducible workflows

All data processing pipelines, calibration procedures, and reconstruction algorithms are fully documented and versioned. Raw data and processed outputs are both published, enabling independent replication and validation of all results.

Publication policy

  • All papers published as open-access preprints on arXiv prior to submission
  • Datasets deposited in open repositories with DOI assignment
  • Processing code published under open-source licences
  • No embargo periods on any research output
  • Null and negative results published alongside positive findings