Climate models René D. Garreaud Departement of Geophysics Universidad de Chile .

Climate models

René D. Garreaud

Departement of GeophysicsUniversidad de Chile

www.dgf.uchile.cl/rene

Modelación Numérica de la Atmósfera

• ¿Por que empleamos modelos climáticos?• ¿En que consiste un modelo climático?• ¿Qué son variables y parámetros?• ¿Qué son los GCM, AOGCM y RCM?• Aplicaciones y limitaciones

¿Porque empleamos modelos climáticos?

• Suplemento de observaciones climáticas (interpolación física)

• Experimentos de gran escala (cambio en forzantes) y análisis de sensibilidad.

• Generación de escenarios climáticos pasados y futuros.

• Dynamical downscaling (escalamiento hacia abajo basado) de climas pasados y futuros.

¿Porque me gustan tanto las modelos?

• Aunque son una representación incompleta del mundo real, un buen modelo captura gran parte de la dinámica del sistema climático terrestre.

• Método alternativo (modelos conceptuales cualitativos) aguanta cualquier cosa…

• Los números que entrega un modelo pueden no ser “exactos” pero dan un rango plausible de cambio climático pasados o futuros.

Surface (land/ocean) Synoptic Stations Met. Observations (T,Td,P,V,…) @ 0, 6, 12, 18 UTC are transmitted in real-time to

WMO and Analysis Centers

From where do we get climate data? Almost all climate data is initially meteorological data, acquired to assist weather nowcast and forecast (especially for aviation)

Red de Radiosondas (OMM, GTS)

Perfiles verticales (20 km) de T, HR, viento, presión, cada 12 / 24 hr

Surface and UpperAir Observations

Satellite Products

AssimilationSystemModel

Gridded Analysis

Griddingmethod

DATA SOURCES AND PRODUCTS

Isobars: lines of constant pressure at a given level(e.g. sea level pressure chart) and time (snapshot).

High (anticyclones) and Low (cyclones) pressure centers.

HH

L

L

SLP map, October 17 2010

We now can reinterpret troughs and ridges seen in the geopotential height in the upper troposphere (e.g., 300 hPa) as tongues of cold air abnormally at low latitudes and warm air abnormally at high latitudes

Geopotential @ 300 hPa (contours)Air temperature averaged between 700 and 400 hPa (colors)

Table 1. Main features of datasets commonly used in climate studies

Dataset Key referencesInput data -Variables

Spatial resolution -Coverage

Time span -Time resolution

Station GHCN

Peterson and Vose (1997)

Sfc. Obs Precip and SAT

N/A Land only

1850(*) – present Daily and Monthly

GriddedGHCN

Peterson and Vose (1997)


5° 5° lat-lon Land only

1900 – present Monthly

GriddedUEA-CRU

New et al. 2000Sfc. Obs Precip and SAT

3.75° 2.5° lat-lon Land only


GriddedUEA-CRU05

Mitchell and Jones (2005)




GridddedU. Delware

Legates and Willmott (1999a,b)



1950 – 1999 Monthly

GriddedSAM-CDC data

Liebmann and Allured (2005)

Sfc. Obs Precip

1° 1° lat-lon South America

1940 – 2006 Daily and Monthly

GriddedCMAP

Xie and Arkin (1987)Sfc. Obs.; Sat. data Precip

2.5° 2.5° lat-lon Global

1979 – present Pentad and Monthly

GriddedGCPC

Adler et al. (2003)Sfc. Obs.; Sat. data Precip

2.5° 2.5° lat-lon Global


NCEP-NCAR Reanalysis (NNR)

Kalnay et al. 1996Kistler et al. 2001

Sfc. Obs.; UA Obs; Sat. data Pressure, temp., winds, etc.

2.5° 2.5° lat-lon,17 vertical levels Global

1948 – present 6 hr, daily, monthly

ECMWF Reanalysis(ERA-40)

Uppala et al. (2005)Sfc. Obs., UA Obs, Sat. data Pressure, temp., winds, etc.

2.5° 2.5° lat-lon,17 vertical levels Global

1948 – present 6 hr, daily, monthly

http://www.cdc.noaa.gov

http://www.ncdc.noaa.gov

http://iridl.ldeo.columbia.edu/

Sistema TerrestreTodos juntos y al mismo tiempo

Hidrosfera

Criosfera

Litosfera

Biosfera

EspacioExterior

Atmósfera

Sistema Terrestre

Subsistema Tiempo de ajuste a cambio climático de escala global

Océano superficial Meses - Años

Océano Profundo Años - Décadas

Campos de Hielo Años - Décadas

Biosfera continental Años - Décadas

Litosfera > Miles de años

Atmosfera

Oceano Superficial

Oceano Profundo

Atmosfera

Oceano Superficial

Oceano Profundo

Atmosfera

Oceano Superficial

Oceano Profundo

SistemaModelado

dT = dias-años dT = siglos y +dT = años - décadas

Mayor largo de simulacion requiere incorporar mas subsistemas terrestres a la simulacion (variables vs parámetros) hasta llegar a modelos del sistema terrestre completo (atmos-ocean-crio-bio-lito)

Prescrito

Sistema TerrestreDividir para modelar

Hidrosfera

Criosfera

Litosfera

Biosfera

EspacioExterior

Atmósfera

Global Models (GCM)

Type SST Sea Ice

Land Ice

Biosphere Land use

AGCM P P P P P

CGCM C C P/C P/C P

OGCM C C P P P

ESM C C C C C

External parameters: GHG, O3, aerosols concentration; solar forcing

Com

plex

ity

1980-

1990-

2005-

A: Atmospheric Only; C: Coupled; O: Ocean; ESM: Earth-system models

X(t) = A·X(t-1) ·Y(t) + B·Z(t-1) + x

Y(t)= C·X(t-1)·Y (t-1) + B·Z (t)+ y

Z(t)= D·Z(t-1)·Y (t) + E·X(t-1) + z

x = y = z = 0

A, B, C, D: External parameterOrbital parameters, CO2 Concentration, SST (AGCM), Land cover

X, Y, Z: Time-dependent variablesPressure, winds, temperature, moisture,….

My first toy modelA system of coupled, non-linear algebraic equations

x y z Randoms error

Set to zero Deterministic model

X(t) = A·X(t-1) ·Y(t) + B·Z(t-1) + x

Y(t)= C·X(t-1)·Y (t-1) + B·Z (t)+ y

Z(t)= D·Z(t-1)·Y (t) + E·X(t-1) + z

x = y = z = 0

To run the model, we need:

• Initial conditions: X0, Y0, Z0

• The values of the External Parameters … they can vary on time

• A numerical algorithm to solve the equations

• A computer big enough

My first toy model

Weather forecastModel predicts daily values

Climate PredictionModel does NOT predict daily valuesbut still gives reasonable climate state

(mean, variance, spectra, etc…)

ObsMod

X0 = -0.502

X0 = -0.501

A slight difference inthe initial conditions

Large differenceslater on

Non-linearequations

The Lorenz’s (butterfly) chaos effect

A=2

A=1

Two runs of the model, everything equal but parameter ANote the “Climate Change” related to change in A

Nevertheless, simulations after two-weeks are still “correct” in a climatic perspective and highly dependent upon external parameters models can

be used to see how the climate changes as external parameters vary.

Examples of External Parametersthat can be modified:

1. The relatively long memory of tropical SST can be used to obtain an idea of the SST field in the next few months (e.g., El Niño conditions). Using this predicted SST field to force an AGCM, allows us climate outlooks one season ahead.

2. Changes in solar forcing (due to changes in sun-earth geometry) are very well known for the past and future (For instance, NH seasonality was more intense in the Holocene than today). Modification of this parameter allow us paleo-climate reconstructions (still need to prescribe other parameters in a consistent way: Ice cover, SST, etc.…hard!)

3. Changes in greenhouse gases concentration in the next decades gases give us some future climate scenarios.

gFpVkfdtVd

R

1ˆ

SfcConvRADPQQQSTV

t

)(

0p

V

pRT

pgz

)(

Momentum eqn.

Energy eqn.

Mass eqn.

Idea gas law

Atmospheric circulation is governed by fluid dynamics equation + ideal gas thermodynamics

ECdt

dqv

rr SEC

dt

dq

Water substance eqns.

Water Vapor

Ice crystalCloud droplets

SnowRain droplets

Graupel/Hail

rcv EEC

dt

dq

cccc KAEC

dt

dq

rsrrccr FPPFEKA

dt

dq

¿¿¿Where is precipitation???W

arm

clo

ud

Cold clouds

Previous system is highly non-linear,with no simple analytic solution

?!.... We solve the system using numerical methods

applied upon a three-dimensional grid

diabQ

x

Tu

t

T

diab

it

iti

t

it

it Q

x

TTu

t

TT

11

111

Once selected the domain and grid, the numerical integration uses finite differences in time and space

Numerial method(stable & efficient)

Sub-grid processes must be parameterized, that is specified in term of large-scale variables

lat

lon

z

lat ~ lon ~ 1 - 3 z ~ 1 km t ~ minutes-hoursTop of atmosphere: 15-50 km

Global Models (GCM)

Ejemplo 1: Cambio climático antropogénico (siglo XXI)

Parámetros variables: Concentración de gases con efecto invernadero.

Parámetros fijos: Geografía, geometría tierra sol.

Componentes variables: Circulación atmosférica y oceánica, biosfera simplificada

GCMs

Escenarios DesarrolloEconomico-Social

EJEMPLO 1: Clima del Siglo XXI

© IPCC 2007: WG1-AR4 (2007)

EJEMPLO 1

Ejemplo 2: Clima del Cretáceo

Parámetros “alterados”: Concentración de gases con efecto invernadero, geografía, geometría tierra sol.

Componentes variables: Circulación atmosférica y oceánica, biosfera simplificada

Ruddiman: Earth’s Climate, Chapter 4

Early Cretaceous world


GCMs (cambios CO2 + geografía)

Modelos además presentan problemas en climas pasadosTermostato tropical? Múltiples equilibrios?

Head et al. Nature 2009

La aparente existencia de un termostato tropical y suno representación en GCMs es un problema…

Sin embargo una buena culebra ayuda

Head et al. Nature 2009


GCMs (cambios CO2 + geografía)

Reconstrucción incluyendo Titanoboaotorga más credibilidad a GCMs

Reconstrucción c/culebra

Ejemplo 3: Aridificación del desierto de Atacama

En este caso haremos un experimento de sensibilidad y no una simulación paleo-climática. No es necesario alcanzar equilibrio.

Parámetros “alterados”: geografía (Andes) y temperatura superficial del oceano.

Parametros fijos: Concentración GEI, geometría tierra-sol, TSM, geografía.

Componentes variables: Circulación atmosférica.

High Andes & dry Atacama

Condiciones actuales hiper-áridas a lo largo del desierto costero. Sin embargo, abundante evidencia geológica de un pasado remoto (Ma) menos extremo en cuanto a déficit de precipitación (e.g., formación de yacimientos de Cu)

Calama12 mm/decada

Iquique6 mm/década

High Andes & dry Atacama

Posicionamiento de Sud América en rango actual de latitudes (80-100 Ma)

Conceptually, both Andean uplift (enhanced bloking of moist air) and SEP cooling (less evaporation from ocean) may increase dryness of the Atacama desert…it would be nice to use a “simple” climate model to study these two conditions.

We use PLASIM, an Earth System Model of Intermediate Complexity from Hamburg University:

• Atmospheric component: PUMA• Simple slab model for SST and Sea Ice• SIMBA for biosphere

We performed 50 year long simulations altering one Boundary condition at a time

Model Validation

0.3*Topo minus Control (DJF)

900 hPa winds and Precip % Precip (P/Pc)

PLASIM Topography Experiments

PLA

SIM

“H

umbo

ldt”

Exp

erim

ents

uSST: SST() only

wSEP: warmerSoutheast Pacific

PLASIM “Humboldt” Experiments

uSST minus Control (DJF)

900 hPa winds and Precip % Precip (P/Pc)

Ejemplo 4: Clima durante el LGM

Regional Models (LAM, MM)

x ~ y ~ 1-50 km z ~ 50-200 m t ~ secondsLx ~ L y ~ 100-5000 km Lz ~ 15 km

z

y

x

Ly

Lx

Lz

Regional Models (LAM, MM)

Regional models gives us a lot more detail (including topographic effects) but they need to be “feeded” at their lateral boundaries by results from a GCM.

Main problem: Garbage in – Garbage out

Model:

• PRECIS – UK

Single domain

• Horiz. grid spacing. 25 km

• 19 vertical levels

• Lateral BC: HadAM every 6h

• Sfc. BC: HadISST1 + Linear trend

Simulations

• 1961-1990 Baseline

• 2071-2100 SRES A2 y B2

• 30 years @ 3 min 4 months per

simulation in fast PC

Proyecto CONAMA – DGF/UCHhttp://www.dgf.uchile.cl/PRECIS

-1° 0° 1° 2° 3° 4° 5° C

Surface Temperature Difference A2-BL

30°S

50°S

75°W

Precipitation Difference A2-BL

30°S

50°S 75°W

-50 0 +50 mm/mes

Climate models René D. Garreaud Departement of Geophysics Universidad de Chile .

Documents

Transcript of Climate models René D. Garreaud Departement of Geophysics Universidad de Chile .