High Sensitivity of Seasonal Precipitation to Local SSTs

  • Slides: 12
Download presentation
High Sensitivity of Seasonal Precipitation to Local SSTs Peter Good, Rob Chadwick, Christopher E.

High Sensitivity of Seasonal Precipitation to Local SSTs Peter Good, Rob Chadwick, Christopher E. Holloway, John Kennedy, Jason A. Lowe, Romain Roehrig, Stephanie S. Rushley www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Stubborn Tropical Precipitation Biases El Nino/La Nina Atlantic seasonal cycle East Pacific seasonal cycle

Stubborn Tropical Precipitation Biases El Nino/La Nina Atlantic seasonal cycle East Pacific seasonal cycle Atlantic spatial pattern East Pacific spatial pattern Indian Ocean spatial pattern Time variability Spatial variability www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Dependence of Seasonal-Mean Precipitation on local SST loge(P/P 0) ≈ kqsat * (qsat -

Dependence of Seasonal-Mean Precipitation on local SST loge(P/P 0) ≈ kqsat * (qsat - qsat, 0) + <other processes> Exponential dependence of precipitation on local SSTs through: 1) Boundary layer moist static energy 2) Total column water vapour 3) Local boundary layer temperature gradients www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Estimation of kqsat • Estimate using interannual variability over the tropical oceans. • Take

Estimation of kqsat • Estimate using interannual variability over the tropical oceans. • Take qsat and log(P) anomalies from the long-term mean for each grid-point/year. • Combine data together from all gridpoints/years using a method that averages out and removes the influence of remote processes. • Finally, calculate kqsat via a linear regression of the sorted data. www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Observational estimates of kqsat www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Observational estimates of kqsat www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Dependence of biases on kqsat El Nino/La Nina Atlantic seasonal cycle East Pacific seasonal

Dependence of biases on kqsat El Nino/La Nina Atlantic seasonal cycle East Pacific seasonal cycle Atlantic spatial pattern East Pacific spatial pattern Indian Ocean spatial pattern Time variability Spatial variability www. metoffice. gov. uk © Crown Copyright 2020, Met Office

kqsat constrains model biases across multiple tropical precipitation indices www. metoffice. gov. uk ©

kqsat constrains model biases across multiple tropical precipitation indices www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Constraint on South American ENSO Precipitation Biases www. metoffice. gov. uk © Crown Copyright

Constraint on South American ENSO Precipitation Biases www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Link to shallow circulations www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Link to shallow circulations www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Summary • A new metric, kqsat, quantifies the dependence of seasonal-mean precipitation variability (in

Summary • A new metric, kqsat, quantifies the dependence of seasonal-mean precipitation variability (in space and time) on local SST. • 43 out of 47 CMIP 5/6 models have a value of kqsat lower then the central observed estimate. • Models with more realistic kqsat have lower tropical precipitation biases across a wide range of metrics of temporal and spatial variability. • The magnitude of kqsat is linked to the strength of shallow circulations associated with shallow convection. • A focus on shallow circulations provides a route to reducing persistent model biases in tropical precipitation. www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Questions? www. metoffice. gov. uk © Crown Copyright 2016, Met Office

Questions? www. metoffice. gov. uk © Crown Copyright 2016, Met Office

Precipitation Observations www. metoffice. gov. uk © Crown Copyright 2020, Met Office

Precipitation Observations www. metoffice. gov. uk © Crown Copyright 2020, Met Office