GOES, Meteosat, and MTSAT ...hbarbosa/uploads/Teaching/IntroFisAtmos2019/… · Water in the atmosphere H 2 O Equivalent in meters Liquid or solid water in the atmosphere 0.0001 m

GOES,Meteosat,andMTSAThttps://www.ssec.wisc.edu/data/comp/wv/wvmoll.mpg

Totalcolumnwatervapor

Waterintheatmosphere

H2O EquivalentinmetersLiquidorsolidwaterintheatmosphere 0.0001m

Vaporintheatmosphere 0.025m

Waterinsoil,lakes,riversandglaciers 50to75m

Oceans 2800m

Stevens&Bony,PhysicsToday2013

Imageby:RobertA.Rohde,GlobalWarmingArt.

Majorgreenhousegases

WaterVapor

CarbonDioxide

Ozone

Methane,NitrousOxide

Water Vapor 60%

Carbon Dioxide 26%

O3 8%

CH4 N20 6%

CloudsalsohaveagreenhouseeffectKiehlandTrenberth1997

TheNaturalGreenhouseEffect:clearsky

https://www.nasa.gov

Stevens&Bony,PhysicsToday2013

Latentheat

Waterintheclimatesystem

Itsphysicalpropertiesdetermine•  Howstronggreenhouseeffectis;•  Planetaryalbedo;•  Thermodynamicstructureofthetroposphere;•  Largescalecirculation;•  Hydrologicalcycle;•  Aerosolshygroscopicgrowth

Austral Summer Precipitation 79-06

GPCP,mm/day

Nov-Mar

Precipita;on,PWVandVaportransportGPCP+ERA401989-2009

Arrautetal,J.Clim,2012mm/day

Nov-Mar

Aerialriversandlakes

mm

SouthAmericanMonsoon

MoistureTrans.&PrecipNov-Mar,ERA40

ITCZtothesouthMarengoetal2012

é  gradTocean/contZhou&Lau1998

é  MoistureTrans.Ocean=>Cont.Rodwell&Hoskins,2001

SubtropicsNicolineetal.2002

AndesbarrierByerle&Paegle2002

MoisturerecyclingEltahir&Bras,1994

ConvergencezoneCarvalhoetal2004

Prec(m

m/day)

Literature:LLJandprecipatLPB(e.g.Nicolinietal.,2002)LLJandMCS(Salioetal.,2007)

Amazonandmoisturerecycling?

Deforestationandmoistureflux?

Moisturefluxandprecipitation?

RiverofSmoke

Rela;onshipbetweenaerosolsandprecipita;onintheLaPlataBasin

AERONET(Aerossols)+TRMM(Precipita;on)+BRAMS(simula;ons)Reduc;oninprecipita;onwithincreaseinaerosols

SilvaDiasetal.,2014BRAMS:Simula;onswithcloudmicrophysicsconfirmthemeasurements

Considerthefluxacrossthisline

Correla^onmaps

Nov-Mar Jul-Aug

Correla^onmaps

Arrautetal,J.Clim,2012

Could we do that for every point?

Nov-Mar

Correla^onmaps

ρi, j,k,l =covt (Pi, j,MTk,l )

stdt (Pi, j )stdt (MTk,l )

Whatifbothvariablesare2Dxt?Createlinksonlywhenρi,j,k,l>thresholdρi, j,k,l =

covt (Pi, j,MTk,l )stdt (Pi, j )std

t (MTk,l )

ComplexNetworks�  Inthecontextofnetworktheory,acomplexnetworkisagraph(network)withnon-trivialtopologicalfeatures

Proper^esofRealNetworks�  Scale-freenetworks�  Mostverticeshavefewneighbors(degree),butsomehave

veryhighdegree(power-lawdegreedistribution)

� Highclusteringcoefficient�  Ifxisconnectedtoyandz,thenyandzare

likelytobeconnected

�  Smallworldnetworks�  Mostverticesarejustafew

edgesawayonaverage.

Examples:�  peoplethatarefriendsonfacebook�  computersthatareinterconnected� webpagesthatpointtoeachother�  proteinsthatinteract�  braincellstransmittinginformation�  phone-callnetworks�  transportationnetworks�  transmissiongrids

MotterandYang,PhysicsToday70,1,32(2017)

Theunfoldingandcontrolofnetworkcascades

Ourintenttodayis…GiveexamplesofhowweappliedComplexNetworksforunderstandingmoisturetransportoverSouthAmerica1.  Propagationofextremeevents2.  Cascadingmoisturerecycling3.  Climatechange&Deforestation

DJFPrecip&Moist.Flux

Cannetworksdetectmoisturebeingtransported

southandcausingprecipitationontheway?

Prec(m

m/day)

Synchroniza^onofextremeprecipita^onevents

network divergence = Δstrength(i) = in-strength(i) - out-strength(i):

0.20.3

0.9

0.40.6

0.2

0.1

0.4

represent strongest synchronizations (2%) as directed and weighted network links:

in-strength(i) = sum of all weights at links pointing to i:

out-strength(i) = sum of all weights at links pointing from i:

definition:A = (Aij)1i,jN

IS(i) =NX

j=1

Aij

OS(i) =NX

j=1

Aji

�S(i) = IS(i)�OS(i)

network construction

Directedweightednetwork

Boersetal,NatureComm.2014

Boersetal,NatureComm.2014

OUTstrenghINstrengh

In(i) = A(ij)j=1

N

∑ Out(i) = A( ji)j=1

N

∑

OUTstrenghINstrengh

Boersetal,NatureComm.2014

~80km/hcontraryto

moistureflux!Boersetal,NatureComm.2014

NetworkDivergence

ΔS(i) = In(i)−Out(i)

http://floodlist.com/america/nasa-satellites-measure-flooding-rain-in-peru-and-bolivia

Canwepredict?

L

L

HH x

Boersetal,NatureComm.2014

El Nino Observed Not observed

Forecasted 7% 7%

Not forecasted 1% 85%

Heidke Skill Score = 0.57 of max 0.60

WhatistheroleofAmazon?

PrecLPB?

PrecSE?

2-LayerMoistureTransportModel

ObservedET

ObservedPrecipObservedWinds

Moisture(complex)network

Zempetal,Atmos.Chem.Phys.2014

56% Molion (1975); Recent (20-35%):

23% Brubaker (1993);30% Eltahir & Bras (1994);

34% Trenberth (1999); 24% Nobrega et al (2005);

Van der Ent (2010)

Scale effect: 4% for one grid-box

28% for entire Amazon

Cascading

Differentpathsforwater,andpossiblecascadingbeforegettingto

“final”destination!Zempetal,ACP2014

Cascading�  For45%ofthewalks,thedirecttransportthemostimportant

�  For55%,awalkwithatleastonestopismoreefficient! Fig. - Distribution of optimal paths

For 0 steps, local recycling For 2 steps, direct transport For 3 or more steps, path with cascading

0 2 4 6 8 Number of steps

Zempetal,ACP2014

DRY

WET

%ofAmazonETonPrecip

Cascading

20-24%ofprecipitationoverLPBcomesfrom

ETAmazon

Cascadingtransport=+6%

[0 – 1]

Cascadingtransport=+10%

Savanna

Seasonal Forest

Effects of water defining the transition forest-Savanna

Malhi et al., PNAS 2010

Nobre & Shukla, J Climate 1991Cox et al, Theo. App. Clim. 2004Sampaio et al., Geo. Res. Let. 2007Hirota et al., Science 2011Nepstad et al., Science 2014Levine et al., PNAS 2016Lapola et al., PNAS 2018

Probabilityoffindingforest

Non-linearresponse

One-waycouplingPèVegFullycoupledsystemPçèVeg Zempetal.,NatureComm.(2017)

Self-amplifiedlossincreasesnon-linearly

Zempetal.,NatureComm.(2017)

êevapotranspirationé=> êmoisturetransport

êprecipitationForest=>Cerrado

Deforesta^on

1ddynamicalsystem

atmos:soil:

1.  Run until equilibrium2.  Deforest N-th box3.  Run until equilibrium4.  Deforest N+1 box5.  etc..

Deforesta^onshowshysteresis

Boersetal.,NatureScientificReports(2017)

Oritnevercomeback,ifcouplingisverystrong

Hysteresisresponseduringreforesta^on

Rainfall 1.5 m to 3 m

Average discharge of 219,000 m3/sec of water

1/6 of all fresh water that drains into the world's oceans

HydrologicalCycle

https://www.nees.uni-bonn.de/research-/systematics-evolution-ecology/biogeography-and-macroecology-biomaps/worldmaps/worldmaps-of-plant-diversity

Biodiversity

Amazon~10%ofallknownbiodiversity

Biomass=>Carbonstorage

Saatchietal.,PNAS2011

Balaetal.,PNAS2007

Global temperature additional increase of +0.7 oC with removal of all tropical forests

CO2 in a world without forestsAbout +200 ppm

D.Lawrence&K.Vandecar,NatureClimateChange(2015)

Worldwithouttropicalrainforests

DecreaseIncrease

AmazonTallTowerObservatory(ATTO),325m

Aerosollifecycle

Cloudlifecycle

Aerosol-cloudinteractions

Carbonbalance

Evenmorenon-linearcomplex

system!

A B123

D

4

E

Exis%ng

A AerosolMicrophysics•  Sca7ering•  Absorp;on•  Concentra;on•  Par;clesizeandmass•  Chemicalcomposi;on•  CCNspectra

B AerosolMicrophysics

C Bioaerosolsampler

D Sunphotometer

E Filtercollec;on

New

1 Ceilometer(IR)

2 Micro-pulsedlidar(VIS)

3 Ramanlidar(UV)

4 All-skyIRcamera

5 Fogsampler

5

C

Schematic view of the aerosol measurements at the ATTO site. Detailed aerosol particle microphysics measurements are already in operation at the 60 m triangular tower (A), and the tall tower (B). Filter collection for elemental chemical composition, and organic and elemental carbon concentration (E) takes place at the walkup tower, and for analysis of fungi and spores (C), at the tall tower. New instruments will allow measuring the vertical profile of aerosol and cloud properties, and collecting in-cloud water for chemical and biological analysis.

Aerosol measurements at ATTO

Aerosol-CloudPhysicsinstrumenta^onsite,containing:Joss-Waldvogeldisdrometer(JWD),RadarWindProfiler(RWP),W-bandCloudRadar(CldRad),GPSforwatervaporcolumn,MP3000,Sodaranddual-polariza^onLidar.Thissystemwillallowaninnova^veviewofprecipita^onforma^onoverCentralAmazonia

Aerosol-CloudPhysicsinstrumenta;onsite