Weather forecasting
Created on August 25|Last edited on August 25
Comment
Predicting coarsely-correlated metrics
Our time series models can track multiple weather indicators:
- temperature (T deg C)
- atmospheric pressure (p mbar)
- relative humidity (rh %)
- wind velocity, horizontal component (Wx)
- wind velocity, vertical component (Wy)
- water vapor concentration (H2OC mmol)
In the charts below, we can compare how well our models predict all these different (and likely correlated) indicators, with the following legend:
- ground truth: black line
- baseline: peach line, literally repeating the previous value for each timestep
- linear model: red line, simplest one-layer approach
- dense models: lines in blue, multiple layers of different sizes, closer to violet hue as model complexity increases
Observations:
- we predict more stable/less variable metrics more closely (relative humidity or temperature in the first chart)
- the more complex model (violet) is often closer to the ground truth than the simpler models (red, light blue)—though this is not super reliable
- some lines are tricky to distinguish—we can swap out a symbol for the line of interest to make the chart easier to read
Run set
14
Distinguishing three highly-correlated vapor pressure variants
Below are three highly correlated metrics:
- VPact = actual vapor pressure, solid line
- VPmax = maximum vapor pressulte, dotted line
- VPdef = deficit, or the difference between these two
We can zoom into different regions to see detailed comparisons. Notice how the predictions of models of intermediate complexity (blues, especially violet) are closer to the black ground truth line than the simple models (red linear, green smallest dense model) and the overly large/complex models (magenta, largest layer sizes).
Run set
14
Add a comment