juliensimon/mars-perseverance-weather
收藏Hugging Face2026-03-26 更新2026-04-12 收录
下载链接:
https://hf-mirror.com/datasets/juliensimon/mars-perseverance-weather
下载链接
链接失效反馈官方服务:
资源简介:
---
license: cc-by-4.0
pretty_name: "Mars Perseverance MEDA Weather"
language:
- en
description: "Surface weather measurements from the MEDA instrument on NASA's Perseverance rover: pressure, temperature, humidity, and thermal infrared radiation on Mars."
task_categories:
- tabular-regression
- time-series-forecasting
tags:
- space
- mars
- perseverance
- meda
- weather
- nasa
- planetary-science
- open-data
- tabular-data
size_categories:
- 10M<n<100M
configs:
- config_name: default
data_files:
- split: train
path: data/meda_weather.parquet
default: true
---
# Mars Perseverance MEDA Weather
*Part of the [Planetary Science Datasets](https://huggingface.co/collections/juliensimon/planetary-science-datasets-69c24caca4ab3934c9856994) collection on Hugging Face.*


Surface weather measurements from the **Mars Environmental Dynamics Analyzer (MEDA)** on NASA's
Perseverance rover in Jezero Crater, Mars. Covers **sol 1** to **sol 1619** with
**69,780,040** measurements across **1524** sols.
## Dataset description
MEDA is a suite of environmental sensors on the Perseverance rover that measures Martian weather
at ~1 Hz cadence. This dataset combines three derived data products from the PDS Atmospheres Node:
- **PS** — Atmospheric pressure (Pa) from the pressure sensor
- **RHS** — Relative humidity (%) and humidity sensor temperature (K)
- **TIRS** — Thermal infrared upward/downward longwave irradiance (W/m2)
Records are merged on spacecraft clock (SCLK) to produce a unified weather timeline.
## Schema
| Column | Type | Description |
|--------|------|-------------|
| `sclk` | int64 | Spacecraft clock count (unique timestamp) |
| `lmst` | string | Local Mean Solar Time |
| `ltst` | string | Local True Solar Time |
| `sol` | int64 | Martian sol (day) number since landing |
| `pressure_pa` | float64 | Atmospheric pressure (Pa) |
| `pressure_uncertainty_pa` | float64 | Pressure measurement uncertainty (Pa) |
| `transducer` | int64 | Pressure transducer ID (1 or 2) |
| `relative_humidity_pct` | float64 | Local relative humidity (%) |
| `relative_humidity_uncertainty_pct` | float64 | Humidity uncertainty (%) |
| `humidity_sensor_temp_k` | float64 | Humidity sensor temperature (K) |
| `humidity_sensor_temp_uncertainty_k` | float64 | Humidity sensor temperature uncertainty (K) |
| `volume_mixing_ratio` | float64 | Water vapor volume mixing ratio |
| `volume_mixing_ratio_uncertainty` | float64 | Volume mixing ratio uncertainty |
| `downward_lw_irradiance_wm2` | float64 | Downward longwave irradiance (W/m2) |
| `downward_lw_irradiance_uncertainty_wm2` | float64 | Downward LW irradiance uncertainty (W/m2) |
| `upward_lw_irradiance_wm2` | float64 | Upward longwave irradiance (W/m2) — proxy for ground temperature |
| `upward_lw_irradiance_uncertainty_wm2` | float64 | Upward LW irradiance uncertainty (W/m2) |
Additional TIRS quality flag columns: `rsm_head_outside_tirs_up_fov`, `wheel_outside_tirs_down_fov`,
`sun_outside_tirs_fov`, `rover_low_tilt`, `tirs_ground_not_in_shadow`, `rover_hga_off`, `skycam_off`, `rover_still`.
## Quick stats
- **69,780,040** measurements across **1524** sols (sol 1--1619)
- Mean surface pressure: **63793.2 Pa** (range 585.5--999999999.0 Pa)
- **52,738,487** humidity readings available
## Usage
```python
from datasets import load_dataset
ds = load_dataset("juliensimon/mars-perseverance-weather", split="train")
df = ds.to_pandas()
# Daily pressure cycle for a given sol
sol_100 = df[df["sol"] == 100]
sol_100.plot(x="ltst", y="pressure_pa", title="Sol 100 pressure")
# Seasonal pressure variation (Mars has ~25% annual pressure swing)
daily_avg = df.groupby("sol")["pressure_pa"].mean()
daily_avg.plot(title="Mars surface pressure by sol")
# Ground temperature proxy from upward thermal IR
df["ground_temp_proxy"] = (df["upward_lw_irradiance_wm2"] / 5.67e-8) ** 0.25
daily_temp = df.groupby("sol")["ground_temp_proxy"].agg(["min", "max"])
daily_temp.plot(title="Ground temperature range by sol")
# Humidity readings (sparse — mostly nighttime)
humid = df[df["relative_humidity_pct"].notna()]
humid.groupby("sol")["relative_humidity_pct"].mean().plot()
```
## Data source
[NASA PDS Atmospheres Node](https://pds-atmospheres.nmsu.edu/PDS/data/PDS4/Mars2020/mars2020_meda/data_derived_env/) —
Mars 2020 MEDA derived environmental data, maintained by New Mexico State University.
## Update schedule
Monthly (1st of each month at 08:00 UTC) via [GitHub Actions](https://github.com/juliensimon/space-datasets).
New sols are ingested incrementally.
## Related datasets
- [mars-craters](https://huggingface.co/datasets/juliensimon/mars-craters) — Mars crater catalog
- [neo-close-approaches](https://huggingface.co/datasets/juliensimon/neo-close-approaches) — Near-Earth object approaches
- [exoplanets](https://huggingface.co/datasets/juliensimon/exoplanets) — NASA Exoplanet Archive
## Pipeline
Source code: [juliensimon/space-datasets](https://github.com/juliensimon/space-datasets)
## Citation
```bibtex
@dataset{meda_weather,
author = {Simon, Julien},
title = {Mars Perseverance MEDA Weather},
year = {2026},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/juliensimon/mars-perseverance-weather},
note = {Based on NASA PDS Mars 2020 MEDA derived environmental data}
}
```
## License
[CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/)
提供机构:
juliensimon



