Long-Range Forecasting of 2m-Temperature with Machine Learning - - PowerPoint PPT Presentation

▶

Sep 09, 2023 203 likes •345 views

Long-Range Forecasting of 2m-Temperature with Machine Learning Etienne Vos Ashley Gritzman Sibusisiwe Makhanya Thabang Mashinini Campbell Watson IBM Research Motivation Why Long-Range Temp. & Precip. Forecasts? Temperature and

SLIDE 1

Long-Range Forecasting

f 2m-Temperature with

Machine Learning

Etienne Vos Ashley Gritzman Sibusisiwe Makhanya Thabang Mashinini Campbell Watson IBM Research

SLIDE 2

Motivation

Why Long-Range Temp. & Precip. Forecasts?

Temperature and precipitation are important climate variables that can have adverse

effects on the economy and society

Sectors affected include: Agriculture, forestry, fisheries, energy, health, tourism
Long-range forecasts can assist in mitigation and preparedness of anticipated impacts

SLIDE 3

Motivation

Why Long-Range Temp. & Precip. Forecasts?

Temperature and precipitation are important climate variables that can have adverse

effects on the economy and society

Sectors affected include: Agriculture, forestry, fisheries, energy, health, tourism
Long-range forecasts can assist in mitigation and preparedness of anticipated impacts

Why Use Machine Learning?

ML approaches require less time and resources to train than numerical climate models
Predictions from ML approaches can be interpretable (e.g. Toms et al., 2019)
In some cases, ML can improve upon numerical climate models (e.g. Ham et al., 2019)

SLIDE 4

Single-Target CNN & LSTM

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

… …

Week: Input horizon (3)

… …

Forecast horizon (25) 1 2 3 23 24 25 Lead time: 22

SLIDE 5

Single-Target CNN & LSTM

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

… …

Week: Input horizon (3)

… …

Forecast horizon (25) 1 2 3 23 24 25 Lead time: 22

Predictor Variables: 2m Temperature 150mb Geopotential 500mb Geopotential Predictand Variable: 2m Temperature Spatial Resolution: 3° x 3° Temporal Resolution: Weekly Dataset: ERA5 Reanalysis

SLIDE 6

Single-Target CNN & LSTM

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

… …

Week: Input horizon (3)

… …

Forecast horizon (25) 1 2 3 23 24 25 Lead time: 22

Predictor Variables: 2m Temperature 150mb Geopotential 500mb Geopotential Predictand Variable: 2m Temperature Spatial Resolution: 3° x 3° Temporal Resolution: Weekly Dataset: ERA5 Reanalysis

2 3 4 2 3 4 2 3 4

Target locations

CNN Model

Stack full global maps as channels Fixed Forecast horizon (e.g. 25 weeks)

SLIDE 7

Single-Target CNN & LSTM

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

1

2 3 4 5 6 7 26 27 28 29 30

… …

Week: Input horizon (3)

… …

Forecast horizon (25) 1 2 3 23 24 25 Lead time: 22

Predictor Variables: 2m Temperature 150mb Geopotential 500mb Geopotential Predictand Variable: 2m Temperature Spatial Resolution: 3° x 3° Temporal Resolution: Weekly Dataset: ERA5 Reanalysis

2 3 4 2 3 4 2 3 4

LSTM

Target locations

LSTM Model

Extract time-series from target location Fixed Forecast horizon (e.g. 25 weeks)

LSTM

…

SLIDE 8

Climatology PCC

SLIDE 9

Climatology PCC

SLIDE 10

Results

Panama City (Low Latitude Location)

SLIDE 11

Results

Panama City (Low Latitude Location) Perth (Mid/High Latitude Location)

SLIDE 12