Radar nowcasting · Dust transport · Energy meteorology

Test first,then conclude.

I worked in operational weather forecasting and observation from 2014 to 2022, then returned to university for an M.S. in Meteorology (expected December 2026). I work on radar nowcasting, dust transport modeling and wind-power forecast evaluation, and every result here comes with its verification.

Looking for a PhD position (Fall 2027) or a research role in AI weather and energy forecasting.

Email me See work

Find meEmail

PPI · illustrative echo

PathFrom operational forecasting to research, built on real forecasts and their verification

2010–2014B.S. Atmospheric SciencesUniversity in China
2014–2022Operational forecastingShort-, medium-range and nowcasting; surface observation
2022–2026M.S. MeteorologyThesis: East Asian dust variability and transport
NowNowcasting and energyRadar–NWP verification, wind-power forecasts

peer-reviewed SCI articles, 1 as first author

Top 4^%

Kaggle code competition silver, 154th of 3,947 teams

fixes merged into pysteps, Satpy and Py-ART

hypotheses written down, then rejected by data

TopicsRadar nowcastingRadar–NWP blendingAI weather model evaluationDust transport modelingWind-power forecast evaluationForecast verificationSatellite remote sensingRapid hazard assessment

3D volume rendering of supercell radar reflectivity

Official wind-power forecast error by lead time

Lead-time verification of extrapolation and radar–NWP blending

Dust source regions and downstream cities (schematic)

Selected work

A few projectsworth showing.

A forecast evaluation for a power market, a first-author paper, and a nowcasting method comparison. Each with data and verification.

2026.08–10 · Power markets · Case study

Texas wind power: how much room does the official forecast leave?

Day-ahead error is 6.24%, level with a perfect-forecast ceiling; the high bias is curtailment. Driven by Pangu-Weather or GraphCast operational forecasts, the same model falls clearly behind the official forecast.

2023–2026 · First-author paper

Where the dust came from: the March 2023 East Asian dust storm

FLEXPART-WRF splits downstream PM₁₀ between two source regions; CALIOP overpasses show the model lofts Taklimakan dust about 0.9 km too low.

2026 · With an open-source benchmark

Radar extrapolation vs. radar–NWP blending: who wins?

Three U.S. cases: blending leads at every lead time in a supercell but loses at +1 h in stratiform rain. The evaluation became nowcast-bench, a package that reports sample sizes with every score.

All work

Findings

Write the hypothesis down,then let the data reject it.

Before an analysis I write down what I expect. These are the ones the data did not let through.

2026.10

AI weatherWith data

Pangu-Weather and GraphCast did not beat the grid operator's wind forecast

The gap has two parts: without 100 m wind, even perfect weather trails the official forecast by 0.7; forecast error adds 1.1.

Day-ahead in 2022, the same weather-to-power model driven by ECMWF HRES, Pangu-Weather and GraphCast gave MAE of 7.80%, 8.10% and 8.19% of capacity, against 5.97% for ERCOT's official forecast; every 95% interval of the difference excludes zero. After removing each source's annual mean bias, neither AI model differs significantly from HRES.

I had assumed: ~~Operational AI weather forecasts, plugged straight in, would match or beat the official wind-power forecast.~~

Evidence

Operational comparison

2026.09

Wind powerWith data

The official day-ahead forecast already matches a perfect-forecast ceiling

The remaining room is between available power and actual generation: curtailment.

ERCOT's 2024 day-ahead wind forecast has an MAE of 6.24% of capacity; my perfect-forecast ceiling built on reanalysis winds is 6.38%. The official forecast's high bias against actual generation comes mostly from curtailment.

I had assumed: ~~A model of my own could beat ERCOT's official wind-power forecast.~~

Evidence

Bias is curtailment Error by lead time

2026.09

Wind power

Seasonal coverage of the training data matters more than the model

Over a full 2024 test year, XGBoost, LSTM and a Transformer differ by less than 0.2 percentage points. Trained without autumn and winter, every model is biased low by 3.2–5.9 points (+0.5 to +1.3 when trained on a full year).

I had assumed: ~~A more complex model would bring wind-power error down.~~

Evidence

Model comparison

2026.07

Nowcasting

Which nowcasting method wins has to be tested case by case

In a stratiform case, the radar–NWP blend lost to plain extrapolation at +1 h; a storm-structure rule derived from the first two cases failed on the third.

I had assumed: ~~Blending in NWP would always beat extrapolation alone.~~

Evidence

Method comparison

All findings

Working together

Two waysto work together.

PhD positions in AI weather, radar or dust, and research roles in energy forecasting.

PhD / research position

Join your group

AI weather prediction, probabilistic forecasting, radar nowcasting or dust. I bring operational forecasting experience, code, and the habit of testing results.

5SCI articles3merged fixes

Email me

Collaboration

Work on a problem together

Data preparation, modeling and verification for dust, radar or energy-meteorology questions.

1first-author paper

Background

Have a positionor a question?

Tell me the problem you are working on, or the position.

Email me All work