Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,19 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# ACE2-ERA5
|
2 |
+
|
3 |
+
Ai2 Climate Emulator (ACE) is a family of models designed to simulate atmospheric variability from the time scale of days to centuries.
|
4 |
+
|
5 |
+
**Disclaimer: ACE models are research tools and should not be used for operational climate predictions.**
|
6 |
+
|
7 |
+
ACE-climSST is the original `ACE' model, as described in [ACE: A fast, skillful learned global atmospheric model for climate prediction](https://arxiv.org/abs/2310.02074).
|
8 |
+
It is trained on output from the FV3GFS atmospheric model forced with annually-repeating climatological sea surface temperature.
|
9 |
+
|
10 |
+
Code for doing inference with ACE models can be found here: [https://github.com/ai2cm/ace](https://github.com/ai2cm/ace)
|
11 |
+
with corresponding documentation here: [https://ai2-climate-emulator.readthedocs.io/en/stable/](https://ai2-climate-emulator.readthedocs.io/en/stable/)
|
12 |
+
|
13 |
+
Briefly, the strengths of ACE-climSST are:
|
14 |
+
- long-term stability
|
15 |
+
- highly accurate time-mean climate compared to its target dataset
|
16 |
+
|
17 |
+
Some known weaknesses are:
|
18 |
+
- responses to more El Niño-like sea surface temperature and long-term warming trends are not accurately captured
|
19 |
+
- small but non-zero drifts in total dry air mass of the atmosphere
|