Curso Cambio Global

CUC Universidade de la costa

Aula 2 – 3 de Março de 2018

Amazonia: An unusual place were

biology forest interacts with climate
and human activities

Paulo Artaxo,
University of Sao Paulo, Brazil
E a Amazônia?
Naturally, the Amazon forest interacts strongly with the atmosphere
and climate. There are strong and complex links between the forest
biology, and the physics and chemistry of the atmosphere
Natural System The Transition
Agriculture expansion and climate variability are critical ingredients on Amazonian transition.
Energy balance and hydrological cycles changes are already observed in Amazonia.

Interactions between land use

change and climate change
are major drivers for changes
in Amazonia.
Tropical Forests and the Global Carbon Cycle

Tropical forest carbon budgets (Pg C yr-1)

Sinks are positive values; sources are negative values

Carbon sink and

source 1990-1999 2000-2007 1990-2007
Tropical intact forest 1.33 ± 0.35 1.02 ± 0.47 1.19 ± 0.41

Tropical regrowth 1.57 ± 0.50 1.72 ± 0.54 1.64 ± 0.52

forest
Tropical gross −3.03 ± 0.49 −2.82 ± 0.45 −2.94 ± 0.47*
deforestation

(Pan et al., 2011)

*-0.9 ± 0.5 Pg C yr-1 globally for 2005-2014

2014 fossil fuel emissions 9.8 ± 0.5 Pg C yr-1

36% (3.5 Pg C yr-1) remained in the atmosphere
(Global Carbon Project 2015)

5
Amazonia as a Complex Nonlinear Interactive System

Illustration from Anke Nölscher

Aerosol and cloud lifecycles
Química Atmosférica e Mudanças Climáticas

VOCs em fase gasosa: Ozônio

VOCs e produção de aerossóis secundários

Meia vida de metano controlada

pelo radical OH-

Monóxido de carbono (CO): importante no

balanço de carbono

Aumento de ozônio (2010-1850) Máximo diário

Conceptual overview of terrestrial carbon cycle –
chemistry – climate interactions

Kulmala et al, 2013

Arneth et al., 2011

 Photosynthesis: where radiation meets life

During photosynthesis, plants absorb carbon dioxide and sunlight to create fuel, glucose
and other sugars for building plant structures. This process forms the foundation of the
biological carbon cycle.
 1. GPP and plant respiration key drivers
Models require better representation of the diversity of plant traits & trade-offs dictating
GPP and respiration, and their responses to environmental drivers.

13
 Pilot study: Vegetation-hydrology interactions
Motivation: How tropical forests influence the storage and fluxes of water during drought is
not well understood but has important implications for how they feedback to novel climates

Scaling
framework of
vegetation-
hydrology
Individual-gap scale Cohort-hillslope scale Bulk canopy-watershed Basin scale
interactions 3D reactive transport REW coupled with scale/Unstructured Earth System Model
model demography model watershed model

OCO-2, GOSAT
Transpiration (mol m-2 s-1)

Sapflow
Atmospheric
Commensurate observations
measurements across scales LANL-FTS
14
O complexo sistema solo-planta-atmosfera

Ahmad B. Moradi
HOW MUCH CARBON DO PLANTS TAKE FROM THE ATMOSPHERE?
MODIS gross primary productivity (GPP) estimation from NDVI 2000-2010

Amazonia: 200 Tg of Carbon in the forest and soils

Global Net Primary Productivity NPP: South America is key…

Ecosystem Model Data Model Intercomparison (EMDI) project

Carbon cycling: Amazonia stores about 350 Tg C. If only a small fraction goes
to or from the atmosphere, large changes in atmospheric CO2 will occurs.
How tropical forests processes affects carbon, water and energy fluxes?

Net carbon flux:

Today: ZERO

Tree mortality:
significant INCREASE
Is the Amazonian
hydrological cycle
intensifying?
Maximum monthly, annual mean
and minimum monthly mean
Amazon river discharge at Óbidos
and in green maximum and
minimum daily mean river
discharge, (b) δ18O in precipitation
in Bolivia derived from tree rings
(Brienen et al. 2012) and (c) tropical
Atlantic sea surface temperature
from Extended reconstructed sea
surface temperature) (Gloor et al.
2013).
 Dry season length is increasing in Amazonia
Southern Amazonia dry-season length (DSL) (red line) and dry-season end, DSE (blue line
Dec. 10–15

Sept. 2–7

Southern Amazonia Sept-Oct-Nov-rainrate

(A) Annual time series of the dry-season length (DSL) (red line) and dry-season end, DSE (blue line) dates derived from the PM daily rainfall
data over the southern Amazonian domain show a decrease of DSL due to a delay of DSE. The unit is pentad (5 d). On the left axis, the 55th
pentad corresponds to September 2–7 of the calendar date and the 70th pentad corresponds to December 10–15. (B) Time series of austral
spring seasonal rainfall over southern Amazonia derived from the PM and GPCP datasets show decrease of rainfall consistent with the delay
of DSE shown in (A). Trends are significant at P < 5%. Rong Fu, PNAS 2013
 Amazonia is critical for
water vapor transport
over South America

What processes controls these fluxes?

 Simulated rainforest biomass under climate change and different plant trait diversity

Forest height structure recovers with biomass.

Annual biomass over 800 simulation years for 400 ha of Ecuadorian rainforest
from three different versions of the vegetation model LPJmL under a severe
climate change scenario (RCP 8.5 HadGEM2). 1T: annual temperature
difference to the mean temperature of pre-impact time (1971–2000) in K.
Visualization of model output showing 0.5 ha
of the 400 ha of Ecuadorian rainforest in a
selected year during pre-, mid-, and post-
impact time, respectively (top to bottom).
Different crown (stem) colors denote different
SLA (WD) values of individual trees.
 Projected distribution of natural biomes for RCP 2.4, 4.5 and 8.5.
Deforestation scenarios for 20%, 40% and 50% + Fire effect

Nobre et al., PNAS, 2016

 D e fo r e s te d a r e a (k m ² p e

10000
12000
14000
16000
18000
20000
22000
24000
26000
28000
30000

2000
4000
6000
8000

0
7 7 /8 8 *
8 8 /8 9
8 9 /9 0
9 0 /9 1
9 1 /9 2
9 2 /9 3
9 2 /9 4
9 4 /9 5
9 5 /9 6
9 6 /9 7
9 7 /9 8
9 8 /9 9
9 9 /0 0
0 0 /0 1
0 1 /0 2
0 2 /0 3
0 3 /0 4
0 4 /0 5
0 5 /0 6
0 6 /0 7
0 7 /0 8
0 8 /0 9
0 9 /1 0
1 0 /1 1
1 1 /1 2
1 2 /1 3
1 3 /1 4
Deforestation in Amazonia 1977-2017 in km² per year

1 4 /1 5
1 5 /1 6
1 6 /1 7
Large scale aerosol distribution in
Amazonia
• Severe health effects on
the Amazonian population
(about 20 million people)
• Climatic effects, with strong effects on
cloud physics and radiation balance.
• Changes in carbon uptake and ecosystem
functioning
 Aerosol emissions make
the high variability visible
– it also applies to aerosol
composition and the trace
gases!
20 years of aerosol optical thickness 2000-2016 in Amazonia
1. Introdução 2. Parametrizações 3. Iniciação de chuva 4.Raios e Precipitação 5.Conclusões

Por que estudamos as influências dos

aerossóis em nuvens sobre a Amazônia?
Estação Chuvosa
NCN ~ 500 cm-3
NCCN(1%) ~ 300 cm-3
NCCN(1%)/NCN ~ 0.6

Pöschl et al., Science 2010 Matin et al., Rev. Geo. 2010

 Aerosol particle acting as
Water vapor cloud condensation nuclei

Correct atmospheric
thermodynamics
conditions

All non linear processes

 Crystal shattering

Evaporation
Crystal
* collection
* * *
Partículas de * *
aerossóis, Collection
núcleos de Coalescence

condensação de SO2
nuvens e Diffusion
SO4-
produção de
chuva
Nucleation

Aerosols Soluble: Nitrates, DON, NH3, NH4+ , etc…

And all clouds need CCN to form
 Warm rain evolution over the western Warm rain evolution over the western
tip of the Amazon, Noon. tip of the Amazon, afternoon.
DSD20021005_1 DSD20021005_2
1000 1000
163826Z 1475m 195200Z 1597m
164111Z 1541m 195506Z 1890m
164542Z 2092m
100 100 195930Z 2338m
165904Z 2948m
200551Z 2781m
170418Z 3275m
201401Z 2980m
171156Z 3548m
172121Z 3631m 202142Z 3640m
172157Z 4250m 202635Z 4403m
10 10
N [cm ]

N [cm ]
-3

-3
1 1

0.1
0.1

0.01
0.01

0.001
0 10 20 30 40 50 0.001
0 10 20 30 40 50
Drop diameter [ m]
Drop diameter [ m]
 Aerossóis Atmosféricos
• Partículas sólidas ou líquidas suspensas em
um meio gasoso;
• Influenciam nos ciclos biogeoquímicos,
visibilidade, saúde, clima, dentre outros.
• Possuem diferentes formas, estruturas e
composição química, tornando seu estudo
complexo;
• Tamanho varia em 4 ordens de grandeza:
1nm- 100µm;
• Processos físico-químicos diferentes geram
diferentes modas de tamanho de
partículas;
• Material Particulado Fino (MPF):
d<2,5µm
• Material Particulado Grosso (MPG):
2,5µm<d<10µm;
• Podem ser primários ou secundários

33
Ian Colbeck & Lazaridis, 2014
 Amazonia: 3 different types of aerosols
Biogenic (primary and SOA) Biomass Burning Dust from Sahara

Each with VERY different properties and impacts

Aerosol size distribution
 Forçante radiativa e os feedbacks do
sistema climático global

IPCC AR5, Chapter 7, 2013

 Principais fontes de aerossol na
Amazônia
1. Emissão Biogênica
 primary biological aerosol particles: partículas
biológicas como fungos, pólen, detritos da
vegetação, etc. dominante na fração grossa
(MARTIN et al., 2010);
 Aerossóis secundários: Provenientes de bVOC (POSCHL et al., 2010).

2. Emissões de queimadas

• Predominam na região durante a estação seca;

NASA Saara,
3. Emissão e transporte de poeira mineral do solo África
Floresta
Ressuspensão da poeira do solo local e transporte de Amazônica
poeira do Saara (BEN-AMI et al., 2010).

4. Transporte a longa distância de aerossol marinho

• Aerossol marinho proveniente do Oceano Atlântico adentra o continente;
37
 Aerossóis Carbonáceos
• Responsáveis por >70% da massa de aerossol na Amazônia;

• Diferentes tipos de aerossóis carbonáceos:

• Carbono orgânico (OC):
• refere-se ao carbono presente em compostos majoritariamente orgânicos;
• pode ser primário ou secundário (SOA);
• Carbono elementar (EC):
• Fração de carbono com composição similar à fuligem;
• Altamente absorvedor de radiação;
• Black Carbon (BCe):
• Carbono altamente absorvedor de luz com propriedades ópticas semelhantes às da fuligem.

US EPA 38
How external (Biomass Burning, Sahara
dust, pollution) and internal emissions
(primary organic aerosol particles and
SOA) interact chemically and physically
altering aerosol and cloud properties.

Pöschl, et al., Science, 2010

Image Source: Jerome Fast

1. Introdução 2. Parametrizações 3. Iniciação de chuva 4.Raios e Precipitação 5.Conclusões

Nuvens convectivas formadas sobre a Amazônia

1. Introdução 2. Parametrizações 3. Iniciação de chuva 4.Raios e Precipitação 5.Conclusões

Nuvens convectivas formadas sobre a Amazônia

 Strong effects of aerosols on carbon uptake in Amazonia
Amazonia Rondonia Forest site 2000-2001
Dry Season - NEE increase: 46 %
0
Wet Season - NEE increase: 24 %
NEE (µmolm s )
-2 -1

-10

-20

-30
Increase in aerosol loading

0.0 0.2 0.4 0.6 0.8 1.0

Cirino et al., 2014 Relative Irradiance
 Mean Diurnal
Radiative Forcing
due to change in
surface albedo:
-8.0 + 0.9 W/m2

Mean Diurnal Aerosol

Forcing Efficiency:
Forest: -22.5 + 1.4 W/m2
Cerrado: -16.6 + 1.7 W/m2

Land-use change radiative forcing.

Forested areas are selected in red and
deforested areas are selected in blue.
Elisa Sena results, 2011
• First estimate of diffuse radiation fertilization due to Amazon BBA.
• BBA increases Amazon basin annual mean diffuse radiation by 3.4–
6.8% and NPP by 1.4–2.8%.
• Effect offsets 33–65% of the annual regional carbon emissions from
BBA.

• Estimate 30–60 Tg C a-1 of NPP enhancement

is within woody tissue  accounts for 8–
16% of the observed carbon sink across
mature Amazonian forests.
• Analysed satellite & AERONET AOD over southwest Brazil and
Bolivia for dry season (from 2001 to 2012).
• Observed dry season AODs strongly correlated with declining
deforestation rate.
• Simulated dry season PM2.5 declined by ~30% in the region.
Reduction in PM2.5 may be preventing roughly 1,700
premature adult deaths annually across South America.
 r = -0.75 Drought
years

Aerosol Optical Depth

P<0.01 r = -0.32

Dry season mean AOD

retrieved by MODIS
 Ozone and carbon uptake in Amazonia in the dry season

Ozone exposure reduces carbon uptake at the same order of magnitude as emissions from
deforestation. Potentially doubling the impact of biomass burning on the carbon cycle
Hydrological cycle critical for Amazonia

Pyrocumulus clouds

Natural clouds
Cloud Condensation Nuclei (CCN) Activity
Objective:
Investigate the effects of anthropogenic pollution on the CCN activity of particulate
matter over a tropical rainforest
Results:
• Dominated by low-hygroscopicity organic species
• Manaus plume affects the Aitken mode, with greater oxidation further downwind
(T2 compared to T3)

T0: Pöhlker et al., T2: Barbosa et al., T3:

Thalman et al., in prep, GoAmazon2014/5
Special Issue
 Rain rate (TRMM) versus Optical Depth (MODIS)

Amazonia

13:30 local-time map of rain rate (R) and the observed trend with aerosol loading in four
selected regions. Period: July and August 2007. b, The average R values are plotted for six
aerosol-loading sets (blue, including zero R grid squares; red, without zero R grid squares).
Note the R intensification as a function of AOD in all cases. (Koren et al., Nature 2012)
Cloud fraction and height as a function of
aerosols in Amazonia

Microphysics

absorption effects

Cloud top pressure (P) vs. AOD Cloud fraction vs. AOD.

Ilan Koren et al., Science 2008

Relationship between aerosols and precipitation in the La Plata Basin
AERONET (Aerossols) +
TRMM (Precipitation) +
BRAMS (simulations)
Reduction in precipitation with increase
in aerosols

BRAMS: Simulations with cloud

microphysics confirm the measurements Silva Dias et al., 2014
 Long range transport of
Sahara desert parti cles to
Amazonia
 African aerosol in central Amazonia

Smoke and dust AOD indicating the advection of African aerosol

toward Amazonia (Baars, 2011).
 GoAmazon2014/15 Experiment
The central idea...
Manaus is a city of 2
million people surrounded
by just forest in a radius of
1.500 Km. UNIQUE ti on
re c
situation. D i
ind
W
The aim of the GoAmazon Manaus
2014/15 experiment is to
analyze how the emissions
of pollutants of the city of
Manaus interacts with the
Amazonian natural
biogenic emissions from
the forest and how are the
impacts on the climate
over the forest and
ecosystem functioning.
How particles are formed from the interactions of forest
biogenic VOCs with urban emissions?

Lightning
Strikes
Cloud Droplet
Number
Anthropogenic Concentration
Emissions Aerosol Particle (CDNC)
Cloud
Condensation
Nuclei (CCN)
Aerosol Activity
Particle
Number
Conc’n (CN)

Natural
Processes of
Rainforest
Ecosystem
 GoAmazon Experiment 2014/15
7 ground sites (before at and after Manaus plume)
DoE G1 plane and the German G5 HALO plane for large scale

Study of the interactions of the urban plume of Manaus with the

forest, producing secondary organic aerosols, ozone and others
 Seven measuring sampling sites in Central Amazonia

ATTO

t io n
d ir ec
i nd
W
Manacapuru Manaus

500 km
 AERONET Aerosol optical depth before and after Manaus plume

Aerosol optical
depth of T0e versus
T3: Where are
Manaus emissions?

Absorption aerosol
optical depth :
Strong difference
between T0e and T3
Single Scattering albedo before and after Manaus plume
Different spectral dependence

Wet Season Dry Season

Before and after Manaus City

Ozone (O3) mean diurnal cycle at T3 site, considering periods impacted and not
impacted by the Manaus urban plume in the wet and dry seasons.

~25 ppb

~10 ppb

Enhancement of ozone at both wet and dry seasons

(Rizzo et al., 2016)
What drives light scattering and absorption for PM1?
T0a ATTO - Organics versus light scattering and absorption

The organics made up to 76% of the fine particles and when

investigated as a function of the scattering coefficient (σ450) different
patterns (with different slopes) were observed over time. BC also
shows different patterns but less pronounced
 Sulfate is related to BC, but with various ratios
Different sources, modulated by different Long range transport processes

T0a - ATTO

BC from
Africa wet
season
Organic aerosols from ATTO to Tiwa and Manacapuru (with BC)
Wet Dry
T0a

T2

T3
IEPOX-SOA fractions of OA
Amazon

W. W. Hu et al., ACP 2015

 Influence of anthropogenic emissions on isoprene-
derived particulate matter in central Amazonia

 Observational constraints of sulfate as a first

order predictor and NO as a modulator of
IEPOX-derived PM.
 NOy serves as indicator of integrated
exposure of airmass to NO chemistry.
 IEPOX-SOA factor obtained from PMF
analysis of AMS data is a proxy for
IEPOX-derived PM.
 Lower loadings of IEPOX-SOA factor observed
for polluted compared to background conditions

S. de Sá, et al., 2017

Wet season aerosol composition in
central Amazonia

Manaus
T3 Clear - T0

Suzane Sá, 2017

Shifts in aerosols with anthrop. influences

Clean Regional Small pollution Heavy pollution

Increase in organic aerosols from 0.7 to 4 ug/m³

Large scale
measurements

G5 HALO plane - “High

Altitude and Long
Range Research
Aircraft”.

DoE G1 plane in
two capaigns wet
and dry seasons
G-1 Flight Paths during GoAmazon
Phase 1 (Wet season) Phase 2 (dry season)

16 flights – 42.8 hours 19 flights – 53.7 hours

Feb 15th - March 26st , 2014 Sep 1st - Oct 10th , 2014
CRIDICON Flights G5-HALO plane dry season 20
 G-1 Flight
Aerial View of T3
Paths during GoAmazon

a. IOP1
Ozone, Isoprene and
oxidation products
downwind of Manaus
Optical Properties

500 m, 11 AM local, 13 March 2014

PARTICLE ORGANIC, GoAmazon2014/5, IOP1, 17 March 2014, 16:24 to 17:31
UTC

PARTICLE ORGANIC
GOES UP

Slide prepared by Scot Martin

PARTICLE SULFATE, GoAmazon2014/5, IOP1, 17 March 2014, 16:24 to 17:31
UTC

PARTICLE SULFATE
GOES UP

Slide prepared by Scot Martin

 How particles are produced in Amazonia?
Feb. 1- Mar. 31, 2014
500

dN/dlog D , (cm-3)
400

Dp (nm)
T0 site (ATTO) 100

10 p
(All data) 200
N: ~320 cm-3 Day time
10 0
00:00 06:00 12:00 18:00 00:00
UTC

dN/dlog10Dp, (cm-3)
500
2500
2000
D (nm)

T0 site (ATTO) 100

1500
(dry season)
p

1000
N: ~3100 cm-3 Day time 500
10
00:00 06:00 12:00 18:00 00:00
UTC
It rains a lot. Removal very high. How the particles are formed?
 Central Questions
• What controls the lifecycle – especially the
sources – of aerosols in the pristine
atmosphere?
• How do secondary aerosol particles originate?
• What is the role of nucleation “banana” events?

(Kerminen et al., 2012)

 Vertical profile of particle size distribution
under pristine condition during wet season
1200 March 7, 2014 6000 6000

5459 m
1000 5000 5000
dN/dlog10Dp (cm-3)

800 4000 4000

Altitude (m)
Altitude (m)
600 4244 m 3000 3000

400 3020 m 2000 2000

1798 m
200 1000 1000
598 m
0 0 0
10 100 500 0 500 1000 0 10 20
Dp (nm) N (Dp>10nm) sp (STP, Mm-1)
(STP, cm-3)
Entrained FT aerosol a source
of particle number in BL? Jian Wang, Nature 2017
Flight AC18, massive Cb (Layer E)
 HALO Plane up to 13 Km altitude

Andi Andreae, 2016 (STP normalized)

 AC18

CCN
CCN (SS
(SS ca.
ca. 0.5-0.6%)
0.5-0.6%) show
show similar
similar
structure,
structure, but
but stronger
stronger enhancement
enhancement
in
in UT.
UT. This
This is
is surprising
surprising because
because the
the UT
UT
aerosols
aerosols are
are mostly
mostly organic.
organic. Maybe
Maybe
they
they contain
contain more
more smaller
smaller particles?
particles?

Andi Andreae, 2016

Aerosol Particle Vertical Distribution
over Amazonia (ACRIDICON)

10.000 to 30.000 particles/cm³ at 12-15 Km versus 300 particles /cm³ at the ground
 The aerosols in the upper troposphere are completely
different from the boundary layer in size and composition
Fraction particles < 90 nm (UFF)

The ultrafine fraction [20 nm< CN

<90 nm] is low in the PBL, and
highest in the upper tropospheric
enriched layers
 The aerosols in the upper troposphere are completely
different from the boundary layer in size and composition
CCN Fraction @0.5% SS
Chemical properties: Volatile Fraction

Most of the particles in the upper

troposphere are volatile at 250 °C,
but they are not sulfuric acid
 AMS Aerosol composition: Organics
and nitrates. No sulfates

Andreae
Andi et al.,2016
Andreae, 2017
Enhancements are independent of BL pollution
CCN are also enhanced in UT, but
not as strongly as CN:
the high CN concentrations are
smaller particles and/or less
hygroscopic

Enhancements in UT similar over

clean area and polluted area:
SS=0.5%
=> Aerosol is not result of pollution,
AC12 but likely product of particle
AC09 production from clean BL air – BVOC
oxidation products
 CCN vertical Profiles HALO ACRIDICON-CHUVA campaign Sep. 2014

AC08 AC09 AC12 AC13 AC14

Aerosol enhancements in
UT are general
AC15 phenomenon, observed
AC16 AC17 AC18 AC19
on almost all flights
- enhanced CCN conc. at high altitude is observed in most of the flights (S~0.5%)
S~0.5% @ HASI Slide from Andi Andreae and Mira Kruger
Clouds as active aerosol processors in the atmosphere
Biogenic organic aerosol formation at low H2SO4 happens in UT!

processing reduces volatility (semi)volatile

Condensation to compounds
new Particles

Particle Boundary-Layer Aerosols

Growth
Biogenic Volatiles (semi)volatile
compounds

Andi Andreae, 2016

Convective precipitation brings these particles down to the
surface

Wang et al. (2016, in press)

 Aerosol Kappa Hygroscopicity Parameter

Results: Dominated by organics, with very low hygroscopicity;

Manaus plume affects the Aitken mode, oxidized T2->T3

T3
T2
T0

Pöhlker et al., 2017, Thalman et al., 2017

 One year CCN and Kappa distribution at ATTO Site

Transition season
Transition season

Nov. Dez. Jan. Feb.

rainy season
rainy season

Mar. Apr. Mai. Jun.

dry season
dry season

Jul. Aug. Sep. Oct.

- Somewhat enhanced values in the rainy season. From Mira Kruger and H. Barbosa
- the dominant kappa value is around 0.2, with larger values for larger diameters
 Amazonia: Liquid Organic Particulate Matter
Aerosols are part of the liquid atmosphere (not the solid one)…

Solid

Liquid

Adam Bateman (2016)

 More Secondary Aerosol Potential than Expected
Objective:
Test through in situ data sets the potential for secondary aerosol production and compare to
state of understanding with respect to known species and their reactive yields

IOP1 Wet Season IOP2 Dry Season

Measured Formation, mg m-3

Measured Formation, mg m-3

UTC UTC

Predicted Formation, mg m-3 Predicted Formation, mg m-3

Results:
• Secondary aerosol produced from oxidation by OH of ambient air
• Production much greater than predicted from modeled yields of measured
ambient precursors
• Suggests that production is dominated by unmeasured species
The unknown Western Amazonia

Rio
RioNegro
Negro
Rio
RioJapurá
Japurá

Rio
RioMadeira
Madeira

Rio
RioSolimões
Solimões
USP River Research vessel
UEA Research vessel
G5 HALO plane - “High
Altitude and Long Range
Research Aircraft” at the
“ACRIDICON: Aerosol,
Cloud, Precipitation, and
Radiation Interactions
and Dynamics of
CONvective Cloud
Systems”.
Uncertainties in Formulating Cloud and
Associated Processes in Climate Models

To predict redistribution of tracers in the atmosphere,

climate models must be able to realistically
determine mass transport due to convection.
 CAFÉ-Brazil 2020

• Use HALO measurements, in combination with ATTO

data, to study:

• Influence of VOC emissions on oxidation chemistry and

radical cycling at low NOx

• How this chemistry links to particle formation and

concentrations

• How clouds, VOCs and convection controls particle

production in the high troposphere?

• Constrain atmospheric chemistry and climate models

Sampling Amazonian
Aerosols in Chacaltaya,
Bolivia at 5,240 meters
Chacaltaya: 5.240
meters altitude
observatory in the
Andes, overlooking
Amazonia
Atmospheric observations at ATTO
ATTO Tower:
Permanent
observatory
at 325
meters height
Very
persistent air
mass
trajectories
Integration of this project with other
ATTO components
Conceptual overview of terrestrial carbon cycle –
chemistry – climate interactions

Arneth et al., 2011

Kulmala et al, 2013

 Fires increase Amazon forest productivity
through increases in diffuse radiation
Rap et al., 2015

Modeled 1998–2007 mean percentage changes

Amazon basin annual mean NPP enhancement caused by in (a–c) diffuse radiation, (g–i) GPP, and (j–l)
BBA as a function of BBA emissions (black: standard
NPP during the wet (defined here as December
BBA emissions; blue: 3 × BBA emissions; and green: 6
× BBA emissions), for each year during
to May) season, dry (June to November)
season, and August due to BBA emissions.
Aerosols effects on NEE – Manaus and Rondonia
(Glauber Cirino, INPA, 2013)

f max: ~ 0,80

NEE (max): ~ -20 µmol/m²s

AOT: ~ 0,5

Manaus K34 tower

Effects of aerosol particles Rondonia
on carbon uptake by the
forest: Diffuse radiation
plays a major role
(Glauber Cirino, 2014)

Manaus
 We can see a complex mixture of sources,
transformation and meteorological
effects that mix forest biology with the
chemistry and physics of the atmosphere

All GoAmazon data is freely available at the

DoE ARM site and in our Brazilian database

Thanks for the attention!!

 Ultrafine particle fraction [20 nm< CN <90 nm]

E2

The ultrafine fraction [20 nm< CN <90 nm] is

low in the PBL, and highest in the upper
tropospheric enriched layers. The highest
values are in layer E2 (youngest aerosol)
Volatility fraction of particles as a function of altitude

The small particles in the enriched

layers are also highly volatile at
250°C, especially again in layer E2
Chemical composition: rBC (SP2)

In the middle and upper troposphere,

rBC is near the detection limit
=> Little transport of PBL aerosol to UT
by convection
 Source apportionment of organic PM
(c)

Suzane Sá, 2017

Source apportionment of organic PM
IEPOX-SOA

MO-OOA

LO-OOA

BBOA

Fac91

HOA

AMS SV-TAG PTR-MS Other

Suzane Sá, 2017

Shifts in PM with anthrop. influences

Time
scale

1h

8 Feb 2014 9 Feb 2014 10 Feb 2014 13 Feb 2014 16 Feb 2014 9 Mar 2014 14 Mar 2014

Suzane Sá, 2017

Shifts in aerosols with anthrop. influences
Shifts in aerosols with anthrop. influences

Time
scale

1h

8 Feb 2014 9 Feb 2014 10 Feb 2014 13 Feb 2014 16 Feb 2014 9 Mar 2014 14 Mar 2014

Figure 5
Shifts in aerosols with anthrop. influences

Time
scale

1h

8 Feb 2014 9 Feb 2014 10 Feb 2014 13 Feb 2014 16 Feb 2014 9 Mar 2014 14 Mar 2014

Figure 5
CO2, CH4 CO,
ATTO, Ascension
and Cape Verde

Monthly averaged daytime

(17:00–20:00 UT) measurements
of CO2, CH4 and CO at the 79m
level of the ATTO tower (blue
line, in comparison with monthly
averaged concentration
measurements from Ascension
Island (brown) and Cape Verde
(green)
3.3. Shifts in PM with anthrop. influences
 SAMBBA Biomass burning emissions
GFED3/4 FINN1.0/1.5 GFAS1.0/1.2
MODIS Burned
Method MODIS TAP MODIS FRP
Area
Spatial 0.5°, 0.25° 1 km2 0.5°, 0.25°
daily daily daily
Temporal 2003 – 2011 2002 – 2012 2001 – 2013
2003 – 2014 2002 – 2015 2001 – now
OC (Tg/yr) 17.6 23 18.2
BC (Tg/yr) 2.2 2.2 2.0
van der Werf et al., Wiedinmyer et al.,
Ref 2010 2011
Kaiser et al., 2012

Injection heights: distributed emissions over six ecosystem-dependent

altitudes between surface and 6 km.

Good agreement between observed and modelled AOD was gained only after
scaling up GFED3 emissions by a factor of 2.0 for HadGEM3 with GLOMAP-mode.
 SV-TAG - semi-volatile thermal desorption aerosol gas
chromatograph with in situ derivatization, which
provides hourly resolved concentrations and gas-particle
partitioning of most common BVOC oxidation tracers

SOAS+GoAmazon: The results of the present study show that the gas-particle partitioning of
approximately 100 known and newly observed oxidation products is not well explained by
environmental factors (e.g., temperature). Compounds having high vapor pressures have higher particle
fractions than expected from absorptive equilibrium partitioning models. Many commonly measured
biogenic oxidation products may be bound in low-volatility mass that decomposes to individual
compounds on analysis

Gabriel Isaacman-VanWertz, A. Godstein EST 2016

Ozone production: G1 data on the Manaus plume

Boxplots para as razões de mistura de O3 Histograma da taxa líquida de

observadas a bordo do avião G1 em transectos produção de O3 calculada com
da pluma urbana a altitudes constantes de base nas observações de superfície
cerca de 700 m, classificadas em função da nos sítios T2 e T3.
distância a Manaus. Os resultados indicam que
a produção de O3 na pluma urbana dá-se
majoritariamente em até 20 km de distância de
sua fonte.
 Vertical transport of free tropospheric small particles – effects on the atmospheric
boundary layer at the T0a (ATTO) site during a precipitation event on March 19, 2014.
Radar reflectivity and
vertical air motion

Reflectivity factor measurements (a, c)

and vertical air motion estimates (b, d)
from the 1290 MHz wind profiler
deployed at the T3 site for a deep
convection system on March 19, 2014
(a, b) and March 20, 2014 (c, d). Strong
surface-driven convective updrafts
reaching altitudes above 12 km were
observed at 8:40 (UTC) on March 19
and 11:30 (UTC) on March 20. Adjacent
to deep convective updraft cores, deep
convective downdrafts from the free
troposphere to the surface were often
observed. Such downward motions
(having maximum velocities up to 10 m
s-1) were consistently observed during
precipitation in the wet season.
 HALO ACRIDICON-CHUVA campaign Sep. 2014
AC09

Low Ozone,
suggesting input of
Not LRT of pollution
BL air
aerosol

High in organic
carbon and NO3-,
low in sulfate

Source: DLR Source: MPIC_PC

- BC from SP-2 and AMS at the HASI Inlet

Slide from Andi Andreae
 Effect of urban pollution on Amazonia cloud properties

Results
Plume decrease
cloud droplet
size and
increase cloud
droplet
concentration.

Micael A. Cecchini, Luiz A.T. Machado et al., ACPD, GoAmazon2014/5 Special Issue
Andi Andreae, 2016
Halo SP2 Black Carbon measurements
Transport of Biomass Burning from Africa
with 30% of sulfate in Amazonia?

28.5% SO4

33.5% SO4
Organic aerosols from ATTO to Tiwa and Manacapuru (with BC)
Wet Dry
T0a

T2

T3
ATTO ACSM Monthly averages
2014-2016
 CAFÉ-Brazil 2020

• Use HALO measurements, in combination with ATTO

data, to study:

• Influence of VOC emissions on oxidation chemistry and

radical cycling at low NOx

• How this chemistry links to particle formation and

concentrations

• How clouds, VOCs and convection controls particle

production in the high troposphere?

• Constrain atmospheric chemistry and climate models

USP Boat to be used as an atmospheric laboratory
in the Madeira River- Western Amazonia
Designed from the beginning to study SOA
formation and trace gas chemistry in the
high Amazonian troposphere with novel
instrumentation and strategies
Sampling Amazonian
Aerosols in Chacaltaya,
Bolivia at 5,240 meters
Chacaltaya: 5.240
meters altitude
observatory in the
Andes, overlooking
Amazonia
Production of IEPOX-derived PM from the
photooxidation of isoprene

Suzane de Sá 2016
IEPOX-SOA fractions of OA
Amazon

W. W. Hu et al., ACP 2015

Suzane Sá backtrajectory analysis, 12 –min resolution

Selected trajectories obey the criteria:

- Cross the background box

- All points to the north of T3 until reaching
the box
- All points after reaching the box are at
higher latitude than T1_lat+0.15 (upper
edge of Manaus box)
- Altitude < 1000m (does not filter any out)

- Cross the Manaus box

- All points in the trajectory are to the east
of T3
- Points in the trajectory until 10h after
reaching the box are consecutively to the
east
- Altitude < 1000m (all obey)
 Suzane: pollution indicators at T3 & trajectories
Filters: (1) Afternoon 12-16 LT
5-95%

Pol: 427 Pol: 444 Pol: 449

Backg: 142 Backg: 149 Backg: 152
 Contours of IEPOX-SOA factor loading for sulfate and NOy concentrations.

Suzane de Sá 2016
The plotted data were recorded during local afternoon
 Simulated rainforest biomass under climate change and different plant trait diversity

Forest height structure recovers with biomass.

Annual biomass over 800 simulation years for 400 ha of Ecuadorian rainforest
from three different versions of the vegetation model LPJmL under a severe
climate change scenario (RCP 8.5 HadGEM2). 1T: annual temperature
difference to the mean temperature of pre-impact time (1971–2000) in K.
Visualization of model output showing 0.5 ha
of the 400 ha of Ecuadorian rainforest in a
selected year during pre-, mid-, and post-
impact time, respectively (top to bottom).
Different crown (stem) colors denote different
SLA (WD) values of individual trees.
 Projected distribution of natural biomes for RCP 2.4, 4.5 and 8.5.
Deforestation scenarios for 20%, 40% and 50% + Fire effect

Nobre et al., PNAS, 2016

• First estimate of diffuse radiation fertilization due to Amazon BBA.
• BBA increases Amazon basin annual mean diffuse radiation by 3.4–
6.8% and NPP by 1.4–2.8%.
• Effect offsets 33–65% of the annual regional carbon emissions from
BBA.

• Estimate 30–60 Tg C a-1 of NPP enhancement

is within woody tissue  accounts for 8–
16% of the observed carbon sink across
mature Amazonian forests.
 Fires increase Amazon forest productivity
through increases in diffuse radiation
Rap et al., 2015

Modeled 1998–2007 mean percentage changes

Amazon basin annual mean NPP enhancement caused by in (a–c) diffuse radiation, (g–i) GPP, and (j–l)
BBA as a function of BBA emissions (black: standard
NPP during the wet (defined here as December
BBA emissions; blue: 3 × BBA emissions; and green: 6
× BBA emissions), for each year during
to May) season, dry (June to November)
season, and August due to BBA emissions.
 Ozone and carbon uptake in Amazonia in the dry season

Ozone exposure reduces carbon uptake at the same order of magnitude as emissions from
deforestation. Potentially doubling the impact of biomass burning on the carbon cycle
Isoprene Chemistry over
CTM prediction
the Amazon Rain Forest
Liu, Artaxo, Martin et al., PNAS

[MVK] + [MACR]
Isoprene

[ISOPOOH]
OH Observations

NO HO2

MVK ISOPOOH
Methyl vinyl ketone Isoprene-derived
hydroperoxides ppb NOy
MACR
Methacrolein OOH Bkgd Anthropogenic Pollution
OH
… …

 Observational constraints of anthropogenic influence.

 Underestimate of the NO pathway for background conditions.
 Cloud Condensation Nuclei (CCN) Activity
Objective:
Investigate the effects of anthropogenic pollution on the CCN activity of particulate
matter over a tropical rainforest
Results:
• Dominated by low-hygroscopicity organic species
• Manaus plume affects the Aitken mode, with greater oxidation further downwind
(T2 compared to T3)

T0: Pöhlker et al., T2: Barbosa et al., T3:

Thalman et al., in prep, GoAmazon2014/5
Special Issue
Radar Profiles of Frequency of Cloud Occurrences

Wet Season:
Low Clouds

Dry season 12 Km

David Troyan, Mike Jensen, Tami Toto, Scott Giangrande and Karen Johnson
 1

MANAUS

3
160
D. Rosenfeld, 2016
 Influence of anthropogenic emissions on isoprene-
derived particulate matter in central Amazonia

 Observational constraints of sulfate as a first

order predictor and NO as a modulator of
IEPOX-derived PM.
 NOy serves as indicator of integrated
exposure of airmass to NO chemistry.
 IEPOX-SOA factor obtained from PMF
analysis of AMS data is a proxy for
IEPOX-derived PM.
 Lower loadings of IEPOX-SOA factor observed
for polluted compared to background conditions

S. de Sá, L. Alexander, S. Martin et al., in prep.

Plant physiological functions in the tropics
Objective: Leaf response curves of photosynthesis with isoprene in Amazonia.
Results:
• Supports function of isoprene through excess photosynthetic energy consumption
• a protective role of isoprene for photosynthesis during high temperature extremes
regularly experienced in secondary rainforest ecosystems.
Light (PAR) Leaf Temperature

As light increases, non-linear increase in Photosynthesis and isoprene emissions

isoprene emissions and photosynthesis. uncouple at high leaf temperature
Jardine, K. J., Jardine, A. B., et al., Atmos. Chem. Phys. Discuss., 2016, GoAmazon2014/5 Special Issue.
 More Secondary Aerosol Potential than Expected
Objective:
Test through in situ data sets the potential for secondary aerosol production and compare to
state of understanding with respect to known species and their reactive yields

IOP1 Wet Season IOP2 Dry Season

Measured Formation, mg m-3

Measured Formation, mg m-3

UTC UTC

Predicted Formation, mg m-3 Predicted Formation, mg m-3

Results:
• Secondary aerosol produced from oxidation by OH of ambient air
• Production much greater than predicted from modeled yields of measured ambient
precursors
• Suggests that production is dominated by unmeasured species
Palm, Jimenez, et al., in prep. for GoAmazon2014/5 Special Issue
From VOCs, SOA, CCN to the canopy processes in Amazonia

Fuentes et al., BAMS 2016

CCN Enhancement at high altitudes
CCN are also enhanced in UT, but
not as strongly as CN: smaller
particles and/or less hygroscopic

Enhancements in UT similar over

clean area and polluted area:
Aerosol is not result of pollution,
but likely product of particle
SS=0.5%
production from clean BL air – BVOC
AC12 oxidation products
AC09

Andi Andreae, Mira Pöhlker, 2016

Ozone, Terpenes and Toluene Concentrations

500 m, 11 AM local, 13 March 2014

Data Source: John Shilling, IARA Experiment, DOE AAF G1 Platform

 Isoprene Photochemistry in Transition?
O O
OH
OH
Isoprene ISOPOO

Isomerization RO2
NO HO2
Products Products
methyl vinyl ketone
(MVK, C4H6O) OOH
ISOPOOHs
methacrolein (C5H10O3) OH
(MACR, C4H6O) …
…

(Tuazon et al., 1990; Paulot et al., 2009; Surratt et al., 2010; Crounse et al., 2011,
Peeters et al., 2009; 2010; 2014; Fuchs et al., 2013…)
Harvard University Environmental Chemistry Group
 HALO ACRIDICON-CHUVA campaign Sep. 2014
F- AC09

Source: DLR, LMU

 G1 Cloud droplet number concentrations
Cloud influenced by plume Cloud outside plume

Cloud droplet number concentrations (CDNC) are greatly increased

in the pollution plume compared to the natural conditions outside
of the plume, as observed by the G1 aircraft.
Slide prepared by Jiwen Fan, PNNL, IARA experiment, DOE AAF G1 platform.
 Conceptual overview of
terrestrial carbon cycle –
chemistry – climate interactions

Arneth et al., 2011

Kulmala et al, 2013

T0z: Scattering
versus organics
 12 years of remote sensing of aerosols and clouds
AOD (550 nm)

Aerosol Optical Depth (550 nm) Upwind

MODIS-Terra - Dry Season Downwind
0.50
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00

03 0 04 0 05 0 06 0 07 0 08 00 9 010 0 11 0 12 0 13 0 14 0 15
20 2 2 2 2 2 2 2 2 2 2 2 2

Effe cti ve Radius (μm )

Cloud Effective Radius Liquid Phase Upwind
MODIS-Aqua - Dry Season Downwind
Cloud Effective Radius Liquid Phase Upwind
21
MODIS-Terra - Wet Season Downwind
20 18.0
Effective Radius (μm)

17.5
19
17.0
18 16.5

17 16.0

15.5
16
3 4 5 6 7 8 9 15.0
0 0 0 0 0 0 0 10 11 12 13 14 15
20 20 20 20 20 20 20 20 2 0 2 0 2 0 20 20
 HALO ACRIDICON-CHUVA campaign Sep. 2014
AC09

Low Ozone,
suggesting input of
Not LRT of pollution
BL air
aerosol

High in organic
carbon and NO3-,
low in sulfate

Source: DLR Source: MPIC_PC

- BC from SP-2 and AMS at the HASI Inlet

Slide from Andi Andreae
Consistent aerosol size distribution over 14 years

Alta Floresta 2000-2014 Cuiabá 2000-2014

P.S.: Thanks to Brent, Joel and Fernando

 T3 – Light Scattering versus T3 – Light Absorption versus
wind direction and speed wind direction and speed

T3 Jan-Dez 2014 T3 Jan-Dez 2014

Higher scattering and absorption observed from

Manaus wind direction and at low wind speeds
Plots from Glauber Cirino
G5 HALO plane - “High
Altitude and Long Range
Research Aircraft” at the
“ACRIDICON: Aerosol,
Cloud, Precipitation, and
Radiation Interactions
and Dynamics of
CONvective Cloud
Systems”.
 T0-ATTO Hygroscopicity parameter kappa vs.
midpoint activation diameter
T0a - ATTO March 2014-Feb 2015
Accumulation

Aitken

- highest kappa values and lowest particle in the rainy season

- the different kappa value between Aitken mode and accumulation mode results from
different composition of the two modes. From Mira Kruger and H. Barbosa
 One year of CCN Measurements at T3
Larger Kappa Values Correspond to Greater CCN Activity of
Individual Particles

Data Source: R. Thalman, J. Wang, et al., BNL.

 Hovmoller diagram of monthly
rainfall from 1951 to 2010 for
southern Amazonia.
Units are in mm/month. The 100 mm/month
isohyet is marked in bold and is an indicator of
dry season.
 
The onset of the Amazon rainy season shows a
large temporal and spatial variability, delays on
the data of the onset may have strong impacts
on local agriculture, hydroelectric power
generation as well as on the hydrology of large
rivers.

Two “once-on-a-century” droughts occurred in

2005 and 2010, and it was shown that on
those events the rainy season started later
than normal, and also that on the last 10 years
the dry season has increased in length by
about one month.

Marengo et al., 2011

How external (Biomass Burning, Sahara
dust, pollution) and internal emissions
(primary organic aerosol particles and
SOA) interact chemically and physically
altering aerosol and cloud properties.

2 years continuous measurements at

seven ground based stations and two
aircraft flying dry and wet seasons
Pöschl, et al., Science, 2010

Image Source: Jerome Fast

 GoAmazon Experiment 2014-2015
4 ground sites (before at and after Manaus plume)
DoE G1 plane and the German G5 HALO plane for large scale

Study of the interactions of the urban plume of Manaus with the

forest, producing secondary organic aerosols, ozone and others
Aerosol and cloud lifecycles
Amazon Basin has strong coupling between terrestrial ecosystem and the
hydrologic cycle: The linkages among carbon cycle, aerosol life cycle,
and cloud life cycle need to be understood and quantified.

Susceptibility and
expected reaction to
stresses of global
climate change as well
as pollution introduced
by future regional
economic development
are not known or
quantified at present
time.

Source: Barth et al., “Coupling between Land Ecosystems and the Atmospheric
Hydrologic Cycle through Biogenic Aerosol  Particles,” BAMS, 86, 1738-1742, 2005.
 Transverse Transects of Urban Plume
500 m, 11 AM local, 13 March 2014

Data Source: Mei Fan, Stephen Springston, IARA Experiment, DOE AAF G1 Platform
 Transverse Transects of Urban Plume
500 m, 11 AM local, 13 March 2014

Data Sources: Mei Fan, Stephen Springston, IARA Experiment, DOE AAF G1 Platform
Simple question: How do atmospheric composition
looks like at T3 compared to T0a, T0z and T2?

re ctio n
d di
Win
The problem: Ozone at mid day 12:00-13:00 LT for each site

Is O3 higher at T3 compared to T2? Lower? When?

 Simple Method: comparing MEDIANS of MONTHLY values for each site
Example: Carbon Monoxide - CO
T0z_CO ppb T2_CO_ppb T3_CO_ppb
T0z T0z T0z T0z T0z T2 T2 T2 T2 T2 T3 T3 T3 T3 T3
Month Mean Median StDev %Coverage Month Mean Median StDev %Coverage Month Mean Median StDev %Coverage
1 203 135 0 0.0 1 155 148 40 4.1 1 139 128 59 10.6
2       0.0 2 151 139 45 89.2 2 124 121 19 33.3
3       0.0 3 150 136 47 74.4 3 125 123 19 34.7
4       0.0 4 130 116 48 51.1 4 124 117 29 34.7
5 113 112 9 80.7 5 129 111 58 91.7 5 107 102 22 34.7
6 118 117 10 99.8 6 150 128 65 93.0 6 117 114 27 34.7
7 126 125 15 98.2 7 152 132 52 99.5 7 127 124 24 34.7
8 181 152 78 99.0 8 222 186 114 80.9 8 158 148 49 19.0
9 165 160 22 98.0 9 218 191 82 89.5 9 209 186 91 32.3
10 168 165 29 86.8 10 190 175 57 94.5 10 202 175 96 31.3
11 168 163 25 82.9 11 213 191 77 98.8 11 183 171 52 99.3
12 190 184 45 98.3 12 211 198 59 91.3 12 174 167 42 99.3

Issues:
Different instruments, different calibrations;
Instrument down for a while
Duplicate measurements at each site (CO at T3 with at least 3 instruments) (BC: 5 instruments
Need to do the medians with Level 3 data
 SO2 (ppb)
SO2 concentrations at T2 - Medians
SO2 higher in the dry 0.30
T2_SO2
season because of 0.25

biomass burning 0.20

0.15
emissions? Or less
0.10
deposition in the dry
0.05
season? Or change in
0.00
trajectories? 1 2 3 4 5 6 7 8 9 10 11 12

Month of 2014

NO2 (ppb) NO2 concentrations at T2 - Medians

3.0
T2_NO2 T3_NOy
2.5

2.0

1.5

NOy and NO2 1.0

0.5

0.0
1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014
 Isoprene at T2 and T3 - Medians

Isoprene (ppb)
PTR-MS 1.4
T2 T3 T3_isoprene
1.2
measurements 1.0

0.8

0.6

0.4

0.2

Excellent 0.0
1 2 3 4 5 6 7 8 9 10 11 12

consistency Month of 2014

M o n o terp en es (p p b )

between T2 and
T3 and between
Monoterpenes at T2 and T3 - Medians
the several 0.35

0.30 T2 T3 T3_Monoterpenes

instruments 0.25

0.20

0.15

0.10

0.05

0.00
1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014
PTR-MS comparison: Benzene, Toluene Acetone and Acetonitrile
Benzene (ppb)

Toluene (ppb)
Benzene at T2 and T3 - Medians Toluene at T2 and T3 - Medians
0.18 0.8
0.16 T2 T3 T3_Benzene 0.7 T2 T3 T3_Toluene
0.14 0.6
0.12
0.5
0.10
0.4
0.08
0.3
0.06
0.2
0.04
0.02 0.1
0.00 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014 Month of 2014

A c et o n it rile (p p b )
Acetone (ppb)

Acetone at T2 and T3 - Medians

2.5
T2 T3 T3_Acetone Acetonitrile at T2 - Medians
2.0 0.25
T2 T3_Acetonitrile
0.20
1.5
0.15
1.0
0.10
0.5
0.05

0.0 0.00
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014 Month of 2014
A c e t a ld e h y d e (p p b ) PTR-MS at T2: Acetaldehyde and Methanol

Acetaldehyde at T2 - Medians
2.0
1.8 T2
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0 1 2 3 4 5 6 7 8 9 10 11 12
M e th a n o l (p p b )

Month of 2014

Methanol at T2 - Medians
3.5
3.0 T2
2.5
2.0
1.5
1.0
0.5
0.0
1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014
 ACSM and AMS: Organics, Sulfate, Nitrates and NH4
O r g a n ic s A C S M (u g / m ³)

S u lfa t e A C S M (u g / m ³)
Organics ACSM at T0z T2 T3 - Medians Sulfate ACSM at T0z T2 T3 - Medians
14.0 1.6
T0z_Organics T2_organics T3_Organic 1.4 T0z_Sulfate T2_Sulfate T3_Sulfate
12.0
1.2
10.0
1.0
8.0
0.8
6.0
0.6
4.0 0.4
2.0 0.2

A m m o n iu m A C S M ( u g / m ³)
0.0 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
N it r a t e s A C S M (u g / m ³)

Month of 2014 Month of 2014

Nitrates ACSM at T0z T2 T3 - Medians

0.5
NH4 ACSM at T0z T2 T3 - Medians
0.4 T0z_Nitrate T2_Nitrate T3_NO3 0.6
0.4 T0z_NH4 T2_NH4 T3_NH4
0.5
0.3
0.4
0.3
0.2 0.3
0.2 0.2
0.1
0.1
0.1
0.0 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month of 2014 Month of 2014

SO4, Organics: very similar T2 and T3: No aerosol formation? No evidence of pasture NH4
 O r g a n ic s A C S M (u g / m
ACSM T0z: 2013 versus 2014
Organics ACSM at T0z 2013-2014 Medians
7.0

Very similar concentrations 6.0

T0z_Org_2013 T0z_Organics 2014

5.0
for 2013 and 2014, also 4.0

similar seasonality 3.0

2.0
1.0
0.0
1 2 3 4 5 6 7 8 9 10 11 12
Month of 2013-2014

S u lfa t e A C S M (u g / m ³)
N it r a t e A C S M (u g / m ³)

Nitrate ACSM at T0z 2013-2014 Medians Sulfate ACSM at T0z 2013-2014 Medians
0.3 1.0
T0z_NO3_2013 T0z_Nitrate 2014 0.9 T0z_SO4_2013 T0z_Sulfate 2014
0.2 0.8
0.7
0.2 0.6
0.5
0.1 0.4
0.3
0.1 0.2
0.1
0.0 0.0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month of 2013-2014 Month of 2013-2014

See poster of Rafael Stern

 Trace elements for the 3 sites at wet and dry seasons
Trace elements fine mode WET season Trace elements coarse mode WET season
1000.00 1000.00
T0z-ZF2 T2 Tiwa T3 Manacapuru T0z-ZF2 T2 Tiwa T3 Manacapuru
100.00
100.00
Concentrations ng/m³

Concentrations ng/m³
10.00
10.00
1.00

1.00
0.10

0.01 0.10
F F F F _F F F F _F _F _F _F _G _G G G G G G _G _G _G _G _G
a_ Al
_ P_ Cl
_
Ca V_ n_ Ni
_
Zn Br Sr Sb Al P_ Cl
_ _ V_ n_ Ni Br Sr
N M Na Ca M Zn Sb

Trace elements fine mode DRY season Trace elements coarse mode DRY season
1000.00 1000.00
T0z-ZF2 T2 Tiwa T3 Manacapuru T0z-ZF2 T2 Tiwa T3 Manacapuru

100.00 100.00
Concentrations ng/m³

10.00 Concentrations ng/m³ 10.00

1.00 1.00

0.10 0.10
F _F F F _F F F F _F _F _F _F G G G G
_
Al P_ Cl
_ V_ n_ Ni
_
Br Sr _G _ G
P_ _ G _ V_ n_ _G _G _G _G b_G
Na Ca M Zn Sb Na Al Cl Ca M Ni Zn Br Sr S

T3 and T0 very similar also in elemental composition for both fine and coarse mode
Thank you!
SSA Diurnal
Variability for each
site and season
Black Carbon
Diurnal Variability for
each site and season
AAE – Absorption
Angstrom Exponent
Diurnal Variability for
each site and season
SAE Scattering Angstrom Exponent
Scattering coefficient at 635 nm
 Filters OC/EC analysis: a changing story…
Analysis by
Andrea Arana and Ana L. Loureiro
OC (ng/m³)

OC/EC analysis Berkeley filters Day / Night

EC (ng/m³)
1400 1400

1200
OC EC 1200

1000 1000

800 800

600 600

400 400

200 200

0 0

Night
Day
T2: Jan-Dec 2014 - Aerosol size distribution and SO2 concentration
M a s s C o n c e n t ra ti o n s (µ

ZF2 T0z 2008-2015 Fine and Coarse mode aerosol mass concentration (µg m-3)
20
Fine
18 Coars
16
e

14

12

10

Compilation by Andrea Arana

B C C o nce ntrati o ns (ng / m ³)

900 ZF2 T0z 2008-2015 Black Carbon (ng

800 m-3) Fine
700 Coars
600 e
500
400
300
200
100
0
12-Feb-0819-Aug-08 19-Feb-09 10-Sep-09 8-Apr-10 15-Nov-10 26-Jun-11 17-Jun-12 6-Nov-14

Compilation by Andrea Arana

AOD and deforestation in Amazonia 2000-2015