Cloud Condensation Nuclei (CCN)—water-soluble particles that droplets condense upon. 10-1000 nm or 0.01-1 µm or 10-6-10-4 cm. (), CCN concentrations vary from 1-105 cm-3 CCN determine cloud droplet (5-50 µm) concentrations, which vary from 1-104 cm-3 CCN are amplified by cloud condensation from 10 to 10,000 nm (0.01 to 10 µm). Thousand in size, million in surface area, billion in volume!

DRI Cloud condensation nuclei (CCN) spectrometers.

Produce a field of supersaturations (S) by thermal diffusion of temperature and water vapor between two parallel plates, where cloud droplets grow on hygroscopic sample particles. More hygroscopic (e.g., larger) particles produce larger cloud droplets. Continuous flow through the cloud chamber (~30s) then into an optical particle counter (OPC). CCN spectrum is deduced from the OPC droplet spectrum. A calibration curve relates OPC droplet size to particle hygroscopicity (critical supersaturation—Sc). Calibration is done with nuclei of known composition (e.g., NaCl) and size (differential mobility analyzer—DMA— electrostatic classifier--EC).

Assumes that all CCN with the same Sc regardless of composition (or size) produce the same droplet sizes. Calibration holds only if all chamber parameters (i.e., flows and temperatures) remain constant. Sc inversely proportional to number of soluble ions. Traditionally CCN plots are cumulative because act cumulatively on the aerosol— all nuclei with Sc < cloud S produce “activated” cloud droplets.

Also previous CCN instruments had too few data points to produce a differential spectrum. DRI CCN spectrometers have enough data points to produce differential spectra. 1% Sc -- 25 nm NaCl

0.1% Sc -- 110 nm NaCl

0.02% Sc – 310 nm NaCl other compositions must have larger sizes for the same Sc. Dec 10, 2004 1226:00-1228:00 AST (1626:00-1628:00) C-130 RICO, Antigua 946 mb CN 194 cm-3 CCN @ 1% 47 cm-3

new spec 100 old spec c

10 dn/dlogS

1 0.01 0.1 1

Sc(%) 1. Simultaneous high time resolution CCN spectra

2. Differential CCN spectra

3. Supersaturation (S) range extended below 0.1%. Extension the traditional CCN (Aitken) range below 0.1% to Large Nuclei, is necessary because a large proportion of CCN have S < 0.1%. This is needed because: • Many clouds have S< 0.1% • Static closure—aerosol and CCN • Dynamic closure—CCN and cloud droplets. • Large nuclei are embryos for . • Droplet spectral width. • Interface with giant nuclei. • CCN sizes. July 19, 2005, 1205-1227 PDT MASE, CCN @ 0.3% S old spectrometer

900

910

920 ascent descent 930

940

950

960

970 pressure (mb) pressure altitude 980

990

1000 0 500 1000 1500 2000 2500 concentration (cm -3) July 15-23, 2005, MASE, CA average within pressure bins CCN 1% for research area 130 15th 120 16th 110 17th 100 18th 90 19th 80 20th 70 22nd 60 23rd 50 cloud top 40 30 20 10 altitude from cloud top altitude from cloud 0 -10 -20 -30 -40 0 500 1000 1500 2000 2500 concentration (cm-3) Average spectra below and just above cloud for the eight flights with unbroken polluted low stratus July 15, 16, 17, 18, 19, 20, 22, and 25 and maritime July 6

below ) 1000 above -3 July 6 below

100 cumulative concentration (cm 10

0.01 0.1 1 S(%) July 15-23 research area; average cumulative spectra just above cloud

1000

15th 16th N(Sc

1000

15th 16th N(Sc

700 )

-3 600

500

400

300

200 cumulative concen. (cm 100

0 10:22 10:23 Time (PDT) August 28, 2007 Christmas Island, PASE 1527-1543 XST

700

750

800

850 pres alt 900

CN 950 CCN

1000

0 200 400 600 800 1000 mean channel number August 28, 2007, new spec C-130 PASE Christmas Island 400

350

300

250

200 CCN

150

100

50

0 11.75 12.00 12.25 12.50 12.75 XST 120 N =0.60CCN-9 1 c 14 100 R = 0.85 Figure 1. Average total (>2.4 µm) FSSP droplet concentrations (Nc) 80 measured during each flight within 2 the 600-900m altitude range and FSSP LWC > 0.1 gm-3 against the 6 average 100m altitude CCN 1112 13 60 15 concentrations at 1% S for each

(cm-3) 18 flight. Data points are plotted as

c 10 8 the flight number (Table 1). All of N the 17 flights considered here except 40 3 9 N =0.61CCN-10 RF17 (J19) had clouds within this 54 c altitude range. The linear regression 7 R = 0.88 line, equation and correlation coefficient 20 (R) is shown as well as the linear for 14 flights regression and R for only the same 14 flights considered by HM7 (excluding 0 RF4 and 11 [D10 and J7]). 0 50 100 150 200 CCN (cm-3) January 16, 2005, RICO, Antigua at 1621-1635 GMT initial full descent 500

600

700

800 CN

pres. alt. (mb) CCN @ ~1% S

900

1000 0 200 400 600 800 1000 1200 1400 1600 1800 particles per cm-3 normalized to sea level pres. January 11, 2005, RICO, Antigua at 2115-2128 GMT final ascent 500

600

700

800 pres. alt. (mb) CN 900 CCN @ ~1% S

1000 0 100 200 300 400 500 600 700 800 900 1000 particles per cm-3 normalized to sea level pres. 4 N = 2.7-0.01CCN R = -0.49 )

-3 3 8 7 Figure 7. As Fig. 4 but for 5 1500-1800m altitude and 35 10 µm diameter. If only RF1, 5, 7, 8, 10 and 18 are 2 considered R is -0.97. R 18 for the other seven flights in this figure is -0.93.

17 1 concentration (cm concentration 129 15 3 1 13 14 0 0 50 100 150 200 CCN (cm-3) 2500

2000 2.4µm 5µm 10µm 1500 15µm 20µm 25µm altitude (m) 30µm 1000 35µm 40µm

500 -1.0 -0.5 0.0 0.5 1.0 R 3000 45µm 55µm 2500 65µm 75µm 85µm 2000 95µm 105µm 115µm 1500 125µm altitude (m)

1000

500 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 R Figure 11. As Fig. 5A but for drops measured by the 260X probe. This includes all flights with data in each altitude band. This mean different numbers of flights and different flights at the various . 600-900m—16 flights (no RF17; same flights as the cloud data at this altitude noted in Fig. 5A caption); 900-1200m—15 flights (no RF13 or 17; same as cloud data noted in Fig. 5A caption); 1200-1500m—15 flights (no RF6 or 17; there was one more data point for the cloud data for this altitude as there was cloud data for RF17); 1500-1800m—13 flights (no RF2, 4, 6 or 17; this differs from the cloud data, which did have RF17 data but not RF11); 1800-2400m—12 flights (no RF2, 4, 6, 9, or 15; this is one more than the cloud data because there is drizzle data at this altitude for RF12); 2400-3000m—6 flights (no RF2, 4, 5, 6, 8, 9, 10, 12, 14, 17 or 18; the same flights as the cloud data at this altitude). 1.0

0.8 600-900m 16 flights 900-1200m 15 flights 0.6 1200-1500m 13 flights 1500-1800m 11 flights 0.4 1800-2400m 10 flights 2400-3000m 6 flights 0.2

0.0 R

-0.2

-0.4

-0.6

-0.8

-1.0 0 50 100 150 200 250 300 threshold diameter (µm)

Figure 12. Correlation coefficients (R) for 1% S CCN at 100m altitude versus cumulative drop(let) concentrations above various size thresholds within the specified altitude ranges. This is the same data displayed in Figs. 5A and 11 but here displayed for each altitude band as a function of drop(let) sizes. As in those other figures the flights considered varied with altitude range and this may cause biases. The number of flights in each altitude range is shown in the legend. 1.0

0.8 0.50-0.55 2.1% 0.45-0.50 2.7% 0.6 0.40-0.45 3.4% 0.35-0.40 4.5% 0.4 0.30-0.35 4.2% 0.25-0.30 6.3% 0.2 0.20-0.25 6.3% 0.15-0.20 8.3%

R 0.0 0.10-0.15 11.4% -0.2 0.05-0.10 16.8% 0.01-0.05 26.9% -0.4

-0.6 -0.8 -1.0 0 1020304050 threshold diameter (µm) Figure 16. As Fig. 14, but for 900-1200m altitude range. There are also eleven flights with data in these LWC intervals. But the flights are different from the flights in Fig. 14. Here RF2, 3, 11, 13, 14 and 17 (D8, D9, J7, J12, J14 and J19) are excluded. With the higher altitude more LWC intervals are available; 7.1% have LWC > 0.55 gm-3. 1.0

0.8

0.6 2µm 0.4 5µm 10µm 0.2 15µm 20µm 0.0 R 25µm -0.2 30µm 35µm -0.4 40µm

-0.6

-0.8

-1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 LWC(gm-3)

Figure 17. As Fig. 15 except that the data are from the 900-1200m altitude range. This is the same data shown in Fig. 16. 1.0

0.8 A 0.75-0.80 4.2% 0.6 0.70-0.75 2.8% 0.4 0.65-0.70 3.9% 0.60-0.65 2.8% 0.2 0.55-0.60 3.6% 0.0 0.50-0.55 4.2% R 0.45-0.50 4.2% -0.2 0.40-0.45 4.1% -0.4 -0.6

-0.8

-1.0 0 1020304050 threshold diameter (µm)

Figure 20. As Fig. 18 but for 2400-3000m altitude. Only four flights; RF1, 11, 13 and 15; 16% exceeded 0.80gm-3. The 0.25-0.30 gm-3 interval with 5.8% of the data is not shown because it did not have data from all four flights. 1.0

0.8 B 0.35-0.40 5.2% 0.30-0.35 5.9% 0.6 0.20-0.25 4.9% 0.4 0.15-0.20 6.4% 0.10-0.15 5.5% 0.2 0.05-0.10 6.3% 0.01-0.05 14.2% 0.0 R -0.2

-0.4 -0.6

-0.8

-1.0 0 1020304050 threshold diameter (µm)

Figure 20B. 1.0

0.8

0.6

0.4 2.4µm 0.2 5µm 0.0 10µm R 15µm -0.2 20µm -0.4

-0.6

-0.8 A

-1.0 0.00.10.20.30.40.50.60.70.8 mean LWC(gm-3)

Figure 21. As Fig. 19 but for 2400-3000m. Same data as in Fig. 20. 1.0

0.8

0.6 B 0.4

0.2 25µm

0.0 30µm R 35µm -0.2 40µm

-0.4

-0.6

-0.8

-1.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 mean LWC(gm-3)

Figure 21B. Adiabaticity

Mean Mean Mean total Adiab. Mn/Ad Altitude altitude LWC LWC LWC LWC LWC range (m) FSSP 260X (gm-3) (gm-3) (m) (gm-3) (gm-3)

2400-3000 2660±150 0.34±0.18 0.22±0.10 0.56 2.69 0.21 1800-2400 2010± 70 0.33±0.16 0.09±0.05 0.42 1.80 0.23 1500-1800 1612± 87 0.29±0.10 0.04±0.03 0.33 1.25 0.26 1200-1500 1358± 74 0.25±0.08 0.02±0.02 0.27 0.96 0.28 900-1200 1049± 77 0.18±0.05 0.01±0.01 0.19 0.57 0.33 600- 900 798± 33 0.10±0.02 0.01±0.01 0.11 0.24 0.46

Table 3. Altitude ranges, means and standard deviations of, measured mean flight-averaged altitudes for LWC > 0.01 gm-3 for flights considered for correlations within the various LWC bins in Figs. 14-21 and text, mean LWC for these flights, adiabatic LWC at the mean altitudes, ratio of mean LWC to adiabatic LWC. 1. Regular ambient measurements. Change integration time (1s) and sample flow rate for better statistics but spectra may suffer 1. CCN measurements invalid in cloud so need to separate data. A. Splashing artifact B. Interstitial measurement—CCN not within droplets, but where is the cutoff? 3. Calibrations—usually done in flight (in cloud) because instrument is very sensitive to changes. Confidence in data. Can be done on ground before or after flight but conditions can be different—esp. Temp. Can bring cal particles in conductive plastic bags. 4. Volatility measurements--~chemistry—NaCl for sure 600-800C. CN/CCN comparison for clean or polluted air—less volatile substances are usually less soluble—ammonium sulfate is both soluble and volatile 200C.

5. Size/Sc measurements—CCN size. Larger CCN are less soluble usually polluted . But PASE measurements may contradict this. 6. May do surface measurements with other instruments.

Continuous, comparisons, Size/Sc, students. July 22, 2005, 1042-1303 High Temperature Processor @ 955-976 mb 1.0

0.8

0.6 CN CCN @ 1% S 0.4

0.2 concentration ratio for HTP to ambient concentration ratio for HTP to 0.0 0 100 200 300 400 500 600 temp C Particle critical Supersaturation (Sc) versus dry particle diameter January 19 and 24, 2005 RICO Eastern Caribbean near Antigua low altitude

1 (%) c

S 0.1 NaCl Amon. Sul. J19 J24a J24b J24c 0.01 10 100 1000 From Hudson 2007 diameter(nm) Particle critical supersaturation (Sc) versus dry diameter July 25, 2005 MASE off the Central California Coast below and above stratus cloud layer

1 (%) c 0.1 S NaCl Amon. Sul. below above below 0.01 101 102 103 From Hudson 2007 diameter(nm) Particle critical Supersaturation (Sc) versus dry particle diameter Nov. 24 and 25 and Dec. 4, 2003 AIRS2 Northeast US

1 (%) c

S 0.1 NaCl Amon. Sul. N24 urban N24 above N25 Lake Huron D4a Lake Huron D4b Lake Huron D4c Lake Huron 0.01 10 100 1000 diameter(nm) Particle critical Supersaturation (Sc) versus dry diameter September 2, 2007 1417-1637 local time PASE near Christmas Island low altitude

1 (%) c

S 0.1

NaCl Amon. Sul. ambient

0.01 10 100 1000 diameter (nm) Particle critical Supersaturation (Sc) versus dry diameter Sep 7, 2007 0913-1507 local time PASE near Christmas Island low altitude

1 (%) c

S 0.1

NaCl Amon. Sul. ambient

0.01 10 100 1000 diameter (nm) 1.6 1.4 1.2 1.0 0.8 B 0.6 0.4 0.2 0.0 102 103 104 CN conc (cm-3)

Figure 4. Average and standard deviation of the solubility

parameter, B, plotted against the simultaneous average CN

concentration during each set of size-Sc measurements. 1. Vertical differences in concentrations—layers above cloud. 2. Is there a low concentration layer right above cloud? 3. Do higher concentrations above get into cloud? 4. Will there be day-to-day variations as in RICO or not as in MASE? 5. CCN-droplet correlations—related to entrainment? 6. Drizzle inverse relationship? 7. Determine cloud S—related inversely to concentration? How related to entrainment? Causes underestimate of S. 8. FSSP missing small cloud droplets when S is low? Haze versus activated droplets? 9. CCN conc. related to cloudiness—cloud scavenging? 10.Volatility

11.Size-Sc, B related to concentration?