AGU 23 - iPosterSessions - an aMuze! Interactive system

Atmospheric Blocking

An atmospheric block is a large, persistent high pressure system that "blocks" the jet stream. It causes stagnation of typically eastward propagating Rossby wave packets and causes extreme weather events in the mid-latitudes. Its first order dynamics are still not fully understood.

(Atmospheric waves indicated by the jet stream. Video source: NOAA)

(Blocking Weather Patterns. Video source: Met Office - UK)

A deadly example is the 2003 European heatwave that resulted in over 70,000 deaths and a dry Loire River (J.-M. Robine et al., C. R. Biologies 331 (2008)).

OPEN

Local Wave Activity (LWA)

Local Wave Acitivty (LWA) is a measure of wave activity density, based on deviations from a zonally symmetric, time-invariant, hypothetical wave-free reference state (Huang and Nakamura, 2016).

Column-mean LWA and $500 hPA height — Column-mean LWA and 500 hPA height (Nakamura and Huang, 2018)

LWA can be used to:

(i). Quantify the amount of the mean-flow modification by the eddy

(ii). Distinguish the longitudinal location of an isolated large-amplitude event such as blocking

Nakamura and Huang (2018) proposed a traffic jam theory in predictions of blocking onsets:

(i). Jet stream has a capacity for the LWA Flux

(ii). When the capacity is exceeded, blocking manifests as congestion.

This is a local (pointwise) property.

The theory identifies non-linear suppresion of the flux caused by large amplitude waves as a quadratic modification to the LWA-LWA Flux relationship:

The traffic jam theory further approximates the quadratic relationship by:

$F \sim(C-\alpha \hat{A}) \hat{A}$

$=-\alpha\left(\hat{A}-\frac{C}{2 \alpha}\right)^2+\frac{C^2}{4 \alpha} \leq\frac{C^2}{4 \alpha}$

such that capacity is given by the maximum flux:

$F_{C, \text { Theory }}=\frac{C^2}{4 \alpha}=\frac{\left(u_{R E F}+c_g-2 \alpha A_0\right)^2}{4 \alpha}$

In comparison, the scatter plots give out the schematics of a direct quadratic regression for LWA flux capacity, without the traffic jam theory approximations:

$F_{C, Direct}$ = Collection of all parabolic vertices

OPEN

The Blocking Climatology

The traffic jam theory captures the blocking climatology: most blocking events lie within the regions of low capacity.

The flux capacity estimated from the theory, $F_{C, Theory}$, aligns with blocking frequency better than the direct calculation, $F_{C, Direct}$. The stationary local wave activity $A_0 $ contains information about blocking by construction.

OPEN

Predicting Individual Events

Can we predict blocks by looking for instances where the flux capacity is exceeded?

Composites of Flux: Blocking vs False Positives

Composites of Capacity: Blocking vs False Positives

Whilst the flux capacity diagnoses blocking climatology in terms of spatial sturctures, it over predicts individual events by giving out 43,684 false positives (flux exceedance is captured, but no blocking event is associated), which is 13 times a day, in contrast to real blocking frequency of 0.04 times a day.

Using two types of composite analyses, we distinguished the false postive predictions from real blocking events by upstream spatial structures.

Hence, we propose a regional amendment called traffic merging to the local traffic jam theory. Merger addresses 2 extra regional constraints to the blocking predictions, reducing the amount of false predictions while preserving the true positives.

OPEN

Regional Feature: Merging

1. Span of large LWA Flux from upstream

2. Capacity shrinkage from upstream

Image Source: Peake and Fowler

OPEN

Key Points

1. Predictors for the onset of blocking based on the novel traffic jam mechanism are evaluated in atmospheric reanalysis.

2. The theory captures the spatial structure of the blocking climatology, but not individual events, tending to overpredict blocking onset.

3. Using regional features improves predictors: a decrease in capacity favors blocking, as lane closures increase the chance of a traffic jam.

OPEN