willychap commited on
Commit
bb8fe52
·
verified ·
1 Parent(s): 2f84dd3

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +24 -12
README.md CHANGED
@@ -15,11 +15,20 @@ pipeline_tag: other
15
 
16
  # CAMulator
17
 
18
- **CAMulator** is an AI emulator of NSF NCAR's CAM6 atmosphere, trained and run
19
- within the [CREDIT](https://github.com/WillyChap/miles-credit) framework. It
20
- rolls a 1 degree (192x288), 32-level, 6-hourly atmospheric state forward for
21
- climate-length simulations (years to decades), driven by prescribed SST, sea-ice,
22
- solar, and CO2 forcing.
 
 
 
 
 
 
 
 
 
23
 
24
  CAMulator is a research tool. It emulates a specific CAM6 configuration and is
25
  not a substitute for an operational forecast or a full Earth-system model.
@@ -28,7 +37,7 @@ not a substitute for an operational forecast or a full Earth-system model.
28
 
29
  - Inference toolbox (code): https://github.com/WillyChap/miles-credit (branch `camulator_huggingface`, dir `climate/`)
30
  - CREDIT framework: https://github.com/NCAR/miles-credit
31
- - Paper: CAMulator (Chapman et al.) -- add DOI/link
32
  - This model + data: https://huggingface.co/willychap/camulator
33
 
34
  ### Inference quickstart
@@ -137,11 +146,14 @@ full training archive is not hosted here; the inputs needed to *run* the model
137
  ### Citation
138
 
139
  ```bibtex
140
- @misc{camulator,
141
- title = {CAMulator: an AI emulator of the CAM6 atmosphere},
142
- author = {Chapman, William E. and others},
143
- note = {CREDIT framework, NSF NCAR},
144
- year = {2025}
 
 
 
 
145
  }
146
  ```
147
- *(Replace with the published reference / DOI when available.)*
 
15
 
16
  # CAMulator
17
 
18
+ **CAMulator** is an auto-regressive machine-learned emulator of NSF NCAR's CAM6
19
+ atmosphere, trained and run within the
20
+ [CREDIT](https://github.com/WillyChap/miles-credit) framework. Given prescribed
21
+ sea-surface temperature, sea-ice, incoming solar radiation, and CO2, it rolls a
22
+ 1 degree (192x288), 32-level, 6-hourly atmospheric state forward for
23
+ climate-length simulations (years to decades). It conserves global dry-air mass,
24
+ moisture, and total atmospheric energy, remains numerically stable over decadal
25
+ rollouts, and reproduces the annual climatology together with major modes of
26
+ variability such as ENSO and the NAO -- at roughly a **350x speedup over CAM6**,
27
+ making it an efficient way to generate large climate ensembles.
28
+
29
+ The model and method are described in Chapman et al. (2025), *CAMulator: Fast
30
+ Emulation of the Community Atmosphere Model*
31
+ ([arXiv:2504.06007](https://arxiv.org/abs/2504.06007)).
32
 
33
  CAMulator is a research tool. It emulates a specific CAM6 configuration and is
34
  not a substitute for an operational forecast or a full Earth-system model.
 
37
 
38
  - Inference toolbox (code): https://github.com/WillyChap/miles-credit (branch `camulator_huggingface`, dir `climate/`)
39
  - CREDIT framework: https://github.com/NCAR/miles-credit
40
+ - Paper: Chapman et al. (2025), *CAMulator: Fast Emulation of the Community Atmosphere Model*, [arXiv:2504.06007](https://arxiv.org/abs/2504.06007)
41
  - This model + data: https://huggingface.co/willychap/camulator
42
 
43
  ### Inference quickstart
 
146
  ### Citation
147
 
148
  ```bibtex
149
+ @article{chapman2025camulator,
150
+ title = {CAMulator: Fast Emulation of the Community Atmosphere Model},
151
+ author = {Chapman, William E. and Schreck, John S. and Sha, Yingkai and
152
+ Gagne II, David John and Kimpara, Dhamma and Zanna, Laure and
153
+ Mayer, Kirsten J. and Berner, Judith},
154
+ journal = {arXiv preprint arXiv:2504.06007},
155
+ year = {2025},
156
+ doi = {10.48550/arXiv.2504.06007},
157
+ url = {https://arxiv.org/abs/2504.06007}
158
  }
159
  ```