search

CIGLR-ai-lab / GreatLakes-TempSensors

Collaborative repository for optimizing the placement of temperature sensors in the Great Lakes using the DeepSensor machine learning framework. Aiming to enhance the quantitative understanding of surface temperature variability for better environmental monitoring and decision-making.
MIT License
0 stars 0 forks source link

Convert surface temperature data to daily anomalies #20

Closed DaniJonesOcean closed 2 months ago

DaniJonesOcean commented 2 months ago

I've created an example notebook of how to calculate a daily seasonal climatology and the anomalies:

https://github.com/CIGLR-ai-lab/GreatLakes-TempSensors/blob/main/notebooks/03_dcj_example_of_seasonal_cycle_decomposition.ipynb

You can also find this on our Google Drive:

https://colab.research.google.com/drive/1qO5wKroTnAI8qTTF_tK10hjE6WgIs9OX?usp=sharing

DaniJonesOcean commented 2 months ago

@eredding02 Please add some climatology and anomaly plots of your data (i.e. the data that is going into DeepSensor) so we can check it out

eredding02 commented 2 months ago

@DaniJonesOcean Using Dani’s example notebook, I have converted the GLSEA3 data to anomalies. As an example I used 2014-2016 data. I plotted a time series of two points on Lake Superior. We can see that 2016 was hotter than the total average on both points, which would explain why we only saw positive values when looking at a time series, as 2016 was my validation year. Image Image

Another visual way we can see this is comparing temperature anomalies of July 4th in 2015 versus 2016. This is all to say that there are negative values in the anomaly data, there was just a suspicious lack of them in 2016 due to increased temperatures. Image Image

DaniJonesOcean commented 2 months ago

@eredding02 This all looks great! Feel free to close the issue when you're ready