The Laboratory of Atmospheric Chemistry The Laboratory of - - PowerPoint PPT Presentation

The Laboratory of Atmospheric Chemistry The Laboratory of Atmospheric Chemistry

Aerosols and Climate Aerosols and Climate CO2 and E t Ecosystems Aerosols NOx, VOC Ozone

Aerosols

Examples:

Ammonium sulfate: ca 0 1 µm

Definition: PM10 = Particles with aerodynamic diameter <10µm Examples:

Diesel soot: ca. 0.1 µm Ammonium sulfate: ca. 0.1 µm Pollen: 10 - 100 µm Sea salt: 0.2 - 10 µm Mineral dust: 0.2 - 10 µm

How big are aerosol particles?

Size relationships

x 20’000 x 20’000 Diesel soot 100 Nanometer (nm) Pin head

2’000’000 nanometer (nm) =

Hot air balloon ( )

= 0.1 micrometer (µm) = 0.0001 millimeter (mm) = 2’000 micrometer (µm) =

2 Millimeter (mm)

40’000’000 micrometer (µm) = 40’000 millimeter (mm) =

40 Meter (m) 40 Meter (m)

In one cubiccentimeter of air: typically 10‘000 particles

Aerosols Aerosols Aerosols Aerosols

• Primary and secondary particles and size distributions
• Instrumentation
• Climate
• Health
• Source identification

Primary particle emissions in 2000 (Tg yr-1)

IPCC, 2001

Formation of secondary aerosol

• Sulfate, homogeneous reaction:

SO2 + •OH + M → HO•SO2 + M HO•SO2 + O2 + M → HOO• + SO3 + M SO3 + H2O → H2SO4

• Sulfate heterogeneous reaction:
• Sulfate, heterogeneous reaction:

SO2(g) ↔ SO2(aq) SO2(aq) + H2O ↔ HSO3

• (aq) + H3O+(aq)

Oxidation with H2O2, Ozon, NO2, …

• Nitrate, homogeneous reaction :

NO2 + •OH → HNO3 NO2 + OH → HNO3 NH3 (g) + HNO3 (g) ↔ NH4NO3 (s)

• Nitrate, heterogeneous reaction :

NO O NO ( ) O NO2 + O3 → NO3 (g) + O2 NO3 (g) + NO2 + M → N2O5 (g) + M N2O5 (g) + H2O (aq) → 2 HNO3 (aq)

• Organics:

VOC + OH, O3, … SOA (secondary organic aerosol)

Source strengths of sulfate and

• rganic carbon (kg m-2 hr-1)

IPCC, 2001

Source strengths of black carbon, mineral dust and sea salt (kg m-2 hr-1)

IPCC, 2001

Size distributions of aerosol particles

Aerosol Size Distribution

40

x103

Nucleation Mode

20 30

Number gDp), cm-3x

Mode

10

N (dN/dlog

Aitken Mode

40

m3/cm3

Droplet Submode Acc m lation

20 30

Volume dlogDp), µm

Condensation Submode Submode Coarse Mode Accumulation Mode

10

(dV/d

0.01 0.1 1 10

Diameter (micrometers)

Traffic emissions: Influence of dilution temperature

1.0E+10

Tair=15C

1.0E+09

m

3

) Tair=25C Tair=35C Tair=42.5C

1 0E+07 1.0E+08

• gDp (#/cm

1 0E+06 1.0E+07

DN/ Dl

1.0E+06 1 10 100 1000

Dp ( nm)

Kittelson et al. (2000)

Size distribution in and around Zürich (day, night): ( y, g )

smallest particles are 10 times more abundant in the city compared to the country side and 100 smallest particles are 10 times more abundant in the city compared to the country side and 100 times more abundant duirng the day than in the night times more abundant duirng the day than in the night

10

6 4

10

5

10

3

10

4

(cm

• 3)

10

1

10

2

N/d(logD)

10 10

1

Urban Area: (Downtown Zьrich) Day (SMPS)

dN

Rural Region: (Zьrcher Oberland) Day (SMPS)

10

• 2

10

• 1

Night (SMPS) Day (OPC) Night (OPC) Night (SMPS) Day (OPC) Night (OPC) 10 100 1000 10000

10

Dp (nm)

10 100 1000 10000

Bukowiecki et al., 2002

Nucleation and Growth in Pittsburgh (August 11 2001) (August 11, 2001)

• S. Pandis

Size distributions in Milan: Evidence for primary and secondary particle formation

Time

Baltensperger et al., 2002

Secondary organic aerosol production Secondary organic aerosol production

Terminology: Terminology: Primary versus secondary

• Primary particles: directly emitted to the atmosphere
• Secondary particles: formed in the atmosphere by condensation
• Secondary particles: formed in the atmosphere by condensation

(nucleation and growth) after chemical transformation

• How about oxidized primary particles ?

aged primary (not secondary as Fuzzi et al. 2006 suggested)

• How about primary particles that evaporate on dilution and condense

after oxidation (Robinson et al., 2007) ? secondary

Measurement Techniques Measurement Techniques

• Number
• Number size distribution
• Mass
• Optical properties
• Aerosol composition (off-line / on-line)

Condensation Particle Counter (CPC) ( )

Example: TSI CPC 3010 Lowest detectable diameter: D = 10 nm Lowest detectable diameter: D = 10 nm Maximum particle concentration: 104 cm-3

Measuring the electrical mobility with a Differential Mobility Analyzers (DMA)

Measurable sizes: D=3-150 nm or 20-900 nm A scan is possible within 60-300 seconds A scan is possible within 60 300 seconds Principle:

• Defined electrical charging of the particles with radiocative source

Defined electrical charging of the particles with radiocative source

• The aerosol flows laminarily through cylindric condensator
• An electrical field force the particles depending on their electrical mobility to

An electrical field force the particles depending on their electrical mobility to move toward the inner electrode.

• Particles of a specific mobility (b=vp/E) are sucked through a gap at the

inner electrode and are detected (typically by a CPC) afterwards.

F t A l Fast Analyzers:

• EEPS

Engine Exhaust Particle

• Spectrometer. Scan time typ. 2 sec.
• FPS

Fast Particle Spectrometer

• FPS

Fast Particle Spectrometer (similar to EEPS)

• FMPS

Fast Mobility Particle Sizer (ambient air version of EEPS) Engine Exhaust Particle Sizer Spectrometer (EEPS, TSI) Fast Particle Spectrometer (DMS500, Cambustion)

Aerodynamic particle sizer

Mobile laboratory

CPC

The PSI mobile laboratory

CPC FMPS MAAP AMS CO2 CO2

Aerosol size distribution (5.6-560 nm) in an Alpine valley

10

6

5x10

6

4

# cm-3 Highway Village

80x10

9

60

m-3 cm-3 Highway Village

3 2

dN/dlogdp / #

40

dV/dlogdp/ nm

1

6 7 8 9

10

2 3 4 5 6 7 8 9

100

2 3

d /

20

6 7 8 9

10

2 3 4 5 6 7 8 9

100

2 3

Alpine Valley

dp / nm dp / nm

Alpine Valley Riviera 2005 Winter

• Consistent picture : Nanoparticle concentrations <30

Consistent picture : Nanoparticle concentrations 30 nanometers very high on highway

• In villages : much lower nanoparticle concentrations, in case of

high wood burning contribution higher volume concentration high wood burning contribution, higher volume concentration

Particulate mass Particulate mass

• Filters and gravimetric analysis

O li B t t TEOM

• On-line : e.g. Betameters, TEOM

Tapered element oscillating microbalance (TEOM)

Optical instruments e.g. light absorption by aethalomete

I0

⎞ ⎜ ⎛ ⎞ ⎜ ⎛ 0 I I x b

abs

e I I

⋅ −

=

“Lambert-Beer’s Law”

Aerosol particles

⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ × = ⎠ ⎞ ⎜ ⎝ ⎛ × =

s r

I I ln 100 I I ln 100 ATN

“Optical attenuation”

⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ − ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ = ∆ ) ( ) ( ln ) ( ) ( ln

1 1 2 2

t I t I t I t I ATN

s r s r

Quartz filter

x I S R For the time interval ∆t=t2-t1 :

⎠ ⎝ ⎠ ⎝ ) ( ) (

1 2 s s I S R The attenuation coefficient (filtered aerosol) The absorption coefficient (airborne aerosol)

t ATN Q A bATN ∆ ∆ =

babs

Correct ion needed, e. g.

The attenuation coefficient (filtered aerosol) The absorption coefficient (airborne aerosol)

• Weingart ner et al. , 2003
• Arnot t et al. , 2005

The absorption exponent α

Absorption exponent α: a measure

babs ∝ λ−α

10.0 12.0

m)

• f the spectral variation in

aerosol light absorption

6 0 8.0

babs (950nm

Example power law fit (λ-2.0) Wood burninga λ-1.8 to -2.2

4.0 6.0

babs(λ) /

g Traffic, diesel soota,b λ-1.0 to -1.1

0.0 2.0 350 450 550 650 750 850 950

a Kirchstetter et al. 2004

b Schnaiter et al. 2003 & 2005 350 450 550 650 750 850 950

λ [nm]

Enhanced UV- absorpt ion due t o t he presence of wood smoke Enhanced UV absorpt on due t o t he presence of wood smoke

Aerosol Chemical Analysis

Offline: Part 1 Offline: Part 2 Online / Direct Sampling Interface Extraction

Chemical Analyzer Chemical Analyzer Analyzer

Data Data

Pump

SO4 = 7.2 µg m-3

2-

Data

SO4

2-

Data Time

Slide courtesy of Jose-Luis Jimenez

Examples of off-line analysis

• Ion chromatography (NH4, NO3, SO4, organic acids)
• X-ray fluorescence, Particle Induced x-ray emission

(PIXE) Inter Coupled Plasma mass spectrometry (ICP (PIXE), Inter Coupled Plasma mass spectrometry (ICP- MS), Neutron activation, Atomic Asorption Spectroscopy (AAS) (elemental analysis: K, S, Pb, Zn, ..) (AAS) (elemental analysis: K, S, Pb, Zn, ..)

• GC- / LC-MS (organic compounds: e g marker

GC / LC MS (organic compounds: e.g. marker compounds hopanes, levoglucosan,)

• IR / UV / proton-NMR- spectroscopy : functional groups
• Mass spectrometry in general : isotope analysis,
• ligomers and more

On Line Analysis On-Line Analysis

• Semi-online: EC-OC (separation of black/elemental

b f i b ) carbon from organic carbon)

Miyazaki et al. , JGR, 2007

On line analysis On-line analysis

• ATOF-MS
• Aerodyne Aerosol mass spectrometer
• ..

Example of Aerosol mass spectrometer measurements together with some black Example of Aerosol mass spectrometer measurements together with some black carbon measurements by an Aethalometer carbon measurements by an Aethalometer

70 60 m

• 3)

Reiden

30 25 m

• 3)

Zurich

50 40 30 20 ss Concentration (µg

Organic Nitrate Ammonium Sulfate BC

20 15 10 ss C oncentration (µg

Organics Nitrate Sulphate Ammonium BC

10 M as 29.01.2006 31.01.2006 02.02.2006 04.02.2006 06.02.2006 08.02.2006 10.02.2006 12.02.2006 Date & Time 100 % 5 M as 08.01.2006 11.01.2006 14.01.2006 17.01.2006 20.01.2006 23.01.2006 Date & Time 100 % 80 60 40 ractional Contribtion 80 60 40 actional Contribution 20 Fr 29.01.2006 31.01.2006 02.02.2006 04.02.2006 06.02.2006 08.02.2006 10.02.2006 12.02.2006 Date & Time 20 Fra 08.01.2006 11.01.2006 14.01.2006 17.01.2006 20.01.2006 23.01.2006 Date & Time

Ti l ti i t d t 6 d t l d t ti li it Time resolution: minutes down to 6 seconds at low detection limits

Aerodyne aerosol mass spectrometer output: size distribution

15 10 Dva (µg m

• 3)

Organics Nitrate Sulphate Ammonium

5 dM/dlogD

4 5 6 7 8 9

100

2 3 4 5 6 7 8 9

1000

2

• Aerosol mass size

distributions

100 1000 dM/dlogDva (µg m

• 3)

Time Resolution Example - Fi Pl f i ft Fire Plumes from aircraft

5000 400 100 4000 400 300 PM1 80 m

• 3)

Altitude AMS total Mass PM1 scattering 3000 ude (ft) 300 Light Scatte 60 Loading (µg rf09_fire2 2000 Altitu 200 ring (Mm

• 1)

40 rosol Mass L rf09_fire1 rf09_fire3 1000 100 20 Ae 8:08 PM 3/23/06 8:10 PM 8:12 PM 8:14 PM 8:16 PM 8:18 PM 8:20 PM UTC UTC

3 Fires in the Yucatan Peninsula - MILAGRO field campaign

Organic Aerosol Analysis in Perspective

HR-AMS

100 alyzed)

Perfect Instrument EC/OC AMS

PBTDMS PILS-OC

HR-AMS Perfect Instrument EC/OC AMS

PBTDMS PILS OC

ion

High

Organic Aerosol Analysis in Perspective 80 60 Mass Ana

PILS-OC FTIR NMR VUV PILS-OC FTIR NMR VUV

Resoluti

40

s (% of M

CI, EA

2D-GCMS

CI, EA

2D-GCMS

and Size

20

eteness

Useless Instrument GC/MS

Tradeoff

Useless Instrument GC/MS

Time a

Low Comple

Class lasses lasses pecies ular ID

Instrument

Tradeoff

Instrument

Correlation btw what is C

One Few C Many C ses to Sp Molecu

Correlation btw what is NOT detected and sources led to missing

Clas

Selectivity SOA for decades

Slide courtesy of Jose-Luis Jimenez

Direct and indirect aerosol effect on climate Direct and indirect aerosol effect on climate

Direct effect: Scattering and absorption of incoming sunlight by aerosol particles Indirect effect: The number concentration of cloud condensation nuclei (CCN) influences the cloud droplet size d th b h th l d and thereby changes the cloud albedo and lifetime

Indirect aerosol effect

Indirect effect

Number of CCN i fl th d l t Large droplets Weak reflection

Small droplets Strong reflection

influences the droplet number and size (Twomey-Effect) and thereby the cloud thereby the cloud albedo and lifetime.

‘Ship tracks’

visualise the indirect effect visualise the indirect effect

Satellitenaufnahme ( Wellenlänge: 3.7 µm)

Indirect aerosol effects

IPCC (2007)

Climate forcing 2000 relative to 1750

IPCC (2001), www.ipcc.ch

Radiative forcing of climate between 1750 and 2005 Radiative forcing of climate between 1750 and 2005

No CO2 time scale is given,as its removal from the atmosphere involves a range of processes that can span long time scales :

IPCC (2007)

Radiative forcing of climate between 1750 and 2005 for different aerosol components 1750 and 2005 for different aerosol components

IPCC (2007)

Aerosol direct radiative forcing – Comparison of different models p

IPCC (2007)

Radiative forcing due to the cloud albedo effect – Comparison of different models cloud albedo effect Comparison of different models

IPCC (2007)

Total Aerosol Optical Depth (MODIS satellite) modelling

IPCC (2007)

Satellite retrieved AOD over Europe: Aerosols show high spatial variability Aerosols show high spatial variability

Polluted Polluted Clean

Robles Gonzales et al., GRL 27, 955 (2000)

Pathways of the Traditional Warm Indirect Aerosol Effect d h Gl i i I di A l Eff

+

and the Glaciation Indirect Aerosol Effect

Cloud albedo + + _ Cloud cover and lifetime + _ Precipitation _ + + + Cloud droplets + + +

Mixed phase cloud hydrometeors

+ + Ice crystals + Aerosol particles Cloud cond. nuclei + + Ice nuclei + Human activity

Lohmann, GRL, 2002

Evolution of particles in cloud: Bergeron-Findeisen process

Cloud Aerosol particle Cloud Ice crystals droplet Snow S o Aerosol particle Rain

Saturation Vapor Pressure (SVP) difference: SVP (ice) < SVP (liquid) ⇒ Flux of water vapor from liquid droplets to ice crystals The WBF mechanism converts many small supercooled drops to only few, large ice crystals, thus changing cloud radiative properties and enhancing precipitation.

Aerosols and Health Aerosols and Health

Aerosols is not a recent problem: The lung of ‘Ötzi‘ black: soot red: soot

Source: www.ecocouncil.dk

Air pollution and mortality during London winter smog in 1952 during London winter smog in 1952.

Adapted from Wilkins (1954)

Increased mortality correlates best with PM2.5 (fine particles)

Dockery et al. (1993)

Epidemiology: clear relationship between PM2.5 (or PM10) and mortality between PM2.5 (or PM10) and mortality

Bold: 1974-1989 (Dockery et al., 1993) ( ) Italic: 1990-1998 (Laden et al., 2006) Laden et al., 2006 CH: 3700 premature deaths per year

Different health effects of PM10

• Well established: Of all air pollutants, PM10 (and even more so PM2.5) show

best correlation with increased mortality best correlation with increased mortality

• Increase of mortality with increase of PM10 by 10 µg/m3 (Laden et al., 2000):
• Traffic:

3.4%

• Coal combustion:

1 1% Coal combustion: 1.1%

• Mineral dust: ~0%

Influence of

• Chemical composition (metals radicals organic compounds acidity)
• Chemical composition (metals, radicals, organic compounds, acidity)
• Biological constituents (allergens, endotoxin)
• Morphology (effect of asbestos)
• Size distribution: smaller particles have greater surface area per unit mass

Size distribution: smaller particles have greater surface area per unit mass

• Number concentration
• …
• Mechanism not well known yet
• Mechanism not well known yet

Loss in life expectancy

attributable to anthropogenic PM2 5 [months] attributable to anthropogenic PM2.5 [months] 2000 2020 2020 C t l i l ti MTFR Current legislation MTFR

Loss in average statistical life expectancy due to identified anthropogenic PM2.5 Calculations for 1997 meteorology Provisional estimates with generic assumption on urban increment of PM

• M. Amann, IIASA

Cytotoxicity of wood smoke from bad combustion: N Klippel Verenum bad combustion: N. Klippel, Verenum

Average composition in Zürich in summer and winter

13% 7% 7%

11% 1% 14%

13%

32% 27% 27%

Black Carbon O i

60%

Zürich (July)

15%

Zürich (January) Organic mass Nitrate Sulfate Ammonium Ammonium

Concentration of solid ammonium nitrate f ti f t t as a function of temperature

7 µg/m3 NH3 µg

3

26.5 µg/m3 HNO3 RH = 30% Gas phase conc.

Chemical composition in Zürich summer and Positive Matrix Factorization of the organic matter

Organic mass O ga c ass

13% 7% 7% 13%

Black Carbon O i

60%

Zürich (July) Organic mass Nitrate Sulfate Ammonium

OOA: Secondary organic aerosol

Lanz et al., ACP (2007) Ammonium

HOA: mostly traffic

8 1.6 wood burning (a=0.6) [µg m

• 3]

Plausibility of solution including 3 factors

29

Roveredo MS: R2 = 0.93 wood burning (modelled) vs CO (measured): R2 = 0.78

6 4 2 1.6 1.4 1.2 1.0 0.8 0.6 wood burning (a=0.6) [µg m

• 3]

CO [ppm] 0.4 0.3 0.2 0.1

• norm. intensity

wood burning aerosol 43 55 60 18 27

chestnut MS: R2 = 0.87

2 09.01.2006 13.01.2006 17.01.2006 21.01.2006 25.01.2006 dat 0.6 0.4 0.2 8 4 OOA (a=0.6) [µg m

• 3]

0.6 44 18 0.1 0.0 140 130 120 110 100 90 80 70 60 50 40 30 20 10 m/z 60 73

G

OOA (modelled) vs. AMS-ammonium (measured) R2 = 0.72

6 4 2 3 2 1 AMS-NH4 [µg m

• 3]

0.6 0.5 0.4 0.3 0.2 0.1

• norm. intensity
• xygenated organic aerosol (OOA, type I)

55 29

culated G culated F

aged rural MS: R2 = 0.93 fulvic acid MS: R2 = 0.87

09.01.2006 13.01.2006 17.01.2006 21.01.2006 25.01.2006 dat

6 5 300 250 HOA (a=0.6) 0.14 0.12 hydrocarbon-like organic aerosol (HOA) 57 55 43 41 0.0 140 130 120 110 100 90 80 70 60 50 40 30 20 10 m/z

cal calc

4 3 2 1 200 150 100 50 NOx [ppb] 0.10 0.08 0.06 0.04 0.02 0.00

• norm. intensity

67 69 71 27 29 83 81 85 91 95

HOA (modelled) vs. NOx (measured): R2 = 0.70

09.01.2006 13.01.2006 17.01.2006 21.01.2006 25.01.2006 dat 0.00 140 130 120 110 100 90 80 70 60 50 40 30 20 10 m/z

Lanz et al., ES&T, accepted

C b ti t i

14C

l i Carbon apportionment using 14C analysis Estimation of fossil and non-fossil SOA contribution

Z_meanPM10

Zürich (mean)

Use of AMS analysis :

38%

Wood burning

ECbb 4% ECfossil

Zürich (mean)

y

• wood burning 38%
• HOA 7%

(biomass burning) 38%

Traffic Secondary (fossil) S d

OM

17%

Assumptions :

• only SOA, HOA and wood

7% 17%

Secondary (non-fossil)

OCfossil 23% OCnonfossil 56%

burning present

• OM/OC=2 for wood

burning and SOA and

19%

EC (fossil)

burning and SOA and OM/OC=1.2 for HOA

EC (fossil) EC (non-fossil)

EC

81%

Lanz et al., accepted in ES&T

Smogchamber results of Carnegie Mellon Smogchamber results of Carnegie Mellon

S h b lt

• Smog chamber results

indicate that wood burning emissions is doubled after

• nly 2 hours of chemistry

Roveredo in an Alpine valley, Januar 2005

ECfossil ECbiomass 9%

06:00 – 14:00

13% OCfossil OCfossil 12% OCnonfossil 66%

ECfossil ECbiomass 11%

Roveredo (GR)

8%

18:00 – 02:00

OCnonfossil 81%

Szidat et al., GRL, 2007

A di l l f th b t i l OC/EC Average diurnal cycle of the carbonaceous material, OC/EC and wood burning versus traffic contributions

Sandradewi et al., ES&T, 2008

Aerosol mass spectra

12 8 4 Levoglucosan 29 44 60 73 167 137

Levoglucosan

12 8 Chestnut, poor burning 29 43 60 167 137

Wood burner (emissions) chestnut, very inefficient burning

4 anics 12 8 Night Filter, March 29 43 73 115 137 167

Night period in Roveredo in March, more than 80% of OC non fossil

% to total orga 4 12 60 73 115 137 167 Averaged nights, Dec 29 43

more than 80% of OC non-fossil Average in Roveredo over the

8 4 12 43 60 73 115 137 167

g whole December

12 8 4 29 44 55 69 81 91 Haerkingen, May

Mass spectra from a Motorway site in May

200 160 120 80 40 m/z

Alfarra et al. ES&T (2007)

SOA formation from TMB (at 50% RH) SOA formation from TMB (at 50% RH) Trimethylbenzene + NOx + light →→ Secondary Organic Aerosol

Observation of SOA oligomerization by Laser Desorption Ionization

1500 1000

mV)

500

Intensity (m

B)

Time after lights on: 3.5 h

1500 1000

mV)

500

Intensity (m

C)

4.5 h

4000 3000

V)

3000 2000 1000

Intensity (mV

D)

6.5 h

800 700 600 500 400

mass m/z

Kalberer et al., Science (2004)

Comparison of smogchamber aerosols with ambient sample

. a.u

trimethylbenzene-SOA

800 700 600 500 400 300

Pattern of Zurich points to α pinene

a.u.

α− pinene-SOA

points to α-pinene rather than TMB (or other anthrop. )

800 700 600 500 400 300

precursors)

a.u.

Downtown Zurich

Baltensperger et al.

800 700 600 500 400 300

p g Faraday Disc. (2005)

Worldwide AMS measurements

• f the chemical composition

Zhang et al., GRL 2007

Models underestimate SOA

Volkamer et al. GRL 2006

SUMMARY SUMMARY

A significant progress was made in last years but many challenges remain: B tt i t t ti ( h ll l

• Better instrumentation (cheaper, smaller, more long-

term, more specific, more precise, … ) is still needed

• Long term chemical composition of aerosols is needed
• Long-term chemical composition of aerosols is needed
• Secondary organic aerosol formation needs to be

understood and implemented in models understood and implemented in models

• Health effects : more specific to size and chemical

composition composition

• Climate : Indirect effects are not quantitatively

understood