Extensive aircraft CCN measurements were made in the ICE-L project over Colorado and Wyoming during November-December, 2007. In spite of significant differences in altitude, temperature, distances from cloud base, updraft velocity (W), entrainment, etc., correlation coefficients (R) between droplet concentrations (Nc) and CCN concentrations (NCCN) were substantial though not as high as those obtained in warm clouds with less variability of non aerosol influences (e.g., Hudson et al. 2009) (hereafter H9). Although higher generally resulted in higher Nc, increases in Nc were generally less than the increases in NCCN largely because of lower cloud supersaturations when NCCN are higher, which was usually the case at lower altitudes. Figure 1 displays all 143 comparisons of CCN measured near each cloud. Table 1 shows that correlations are better when liquid water content is used as the cloud threshold rather than droplet number concentrations.
LWC/Nc | 0.01 (gm-3) | 0.05 (gm-3) | 0.10 (gm-3) | 0.15 (gm-3) | 0.20 (gm-3) | 1 (cm-3) | 5 (cm-3) | 10 (cm-3) |
clouds | 143 | 121 | 94 | 74 | 59 | 149 | 145 | 145 |
seconds | 11302 | 8755 | 6894 | 5471 | 4252 | 13090 | 12396 | 11954 |
R | 0.69 | 0.75 | 0.78 | 0.74 | 0.72 | 0.56 | 0.63 | 0.64 |
Table 1. Number of clouds, number of seconds within clouds and correlation coefficients (R) using various thresholds to define clouds in terms of liquid water content or droplet number concentration.
Fig. 1 displays the division according to altitude at 3 and 6 km. Table 2 then displays the correlations for each altitude band. R is substantially positive for all but the highest altitude band. But even for that altitude R is positive when the 3 coldest clouds (temperatures < -35 C) (RF12) are not considered. Table 2 also shows that the correlations were higher when only the CCN measurements from below the clouds were considered.
| only below cloud CCN | |||||||
LWC | alt | clouds | sec | R | clouds | sec | R | |
0.01 | < 3 km | 35 | 6331 | 0.53 | 25 | 4758 | 0.43 | |
3-6 km | 87 | 3697 | 0.40 | 62 | 2783 | 0.69 | ||
> 6 km | 21 | 1274 | -0.01 | 11 | 553 | 0.01 | ||
> 6 km EX12 | 18 | 1226 | 0.47 | 9 | 525 | 0.67 | ||
0.10 | < 3 km | 33 | 5090 | 0.64 | 23 | 4014 | 0.68 | |
3-6 km | 50 | 1309 | 0.39 | 31 | 768 | 0.69 | ||
> 6 km | 11 | 495 | 0.31 | 5 | 191 | 0.51 | ||
Table 2. As Table 1 but divided according to altitude and using only two LWC thresholds. The extra row for 0.01 and > 6 km excludes RF12.
Figure 2 shows NCCN -Nc relationships above 6 km. Panel a shows the negative R for all of these clouds. But when the 3 clouds with the lowest Nc are removed R goes to +0.47 (Table 2). These three coldest clouds also have the highest concentrations measured by the 2DC probe (N2DC). This probe measures larger cloud particles, which at low temperatures are usually ice particles. The triangles represent the clouds with the next highest N2DC and when they are also excluded R goes to +0.76. Panel B shows the 12 clouds with the highest N2DC with R of -0.42. Panel C is Panel B without those 3 coldest clouds with the highest N2DC; R = -0.01. Panel D displays the 12 clouds above 6 km with the lowest N2DC; R = 0.82. The clouds with the 9 lowest N2DC have R = 0.81. These results for these high altitude clouds show that the presence of ice disrupts the correlations of NCCN with Nc as should be expected because water tends to diffuse from droplets to ice particles. This disruption of R was only observed above 6 km and for temperatures below -20C.
Figure 3 displays the NCCN-Nc R pattern for ICE-L clouds. This is similar to that uncovered in warm clouds (H9 and Hudson and Noble 2009). This shows R changing from positive for all cloud droplets to negative for slightly larger size thresholds. After reaching a maximum negative R at 15 µm in Fig. 3, R goes to smaller negative and even positive values for even larger size thresholds. Figure 4a shows a lower positive R because of the smaller concentration range of the restricted altitude band. The negative R is due to competition among droplets for condensate, which reduces droplet sizes to a greater extent when NCCN are higher. Figure 4 shows that the negative R occurs just past the mode of the average droplet concentrations as noted by Hudson and Noble 2009. The decreasing negative R or positive R at larger sizes is due to the less competition of the lower concentrations of larger droplets. The R patterns are best explained by Figure 5, which shows predictions of cumulative droplet concentrations of the adiabatic model of Robinson (1984). Droplets are grown under identical conditions on three CCN spectra; two are exact multiples of one that was observed during ICE-L. For cumulative concentrations less than 7 µm in Figure 5 droplet concentrations are in proportion to NCCN (positive R). Between 8.5 and 12.5 µm diameter droplet concentrations are inversely related to the NCCN, thus negative R. Beyond 14.5 µm droplet concentrations are again in proportion to NCCN, thus positive R. This illustrates the clash between proportionality with the aerosol (positive R) and competition among droplets (negative R). This demonstrates that the influence of the aerosol on cloud microphysics is greater than that indicated by R. This demonstrates greater complexity of the indirect aerosol effect.
Hudson, J.G. and S. Noble, 2009: CCN and cloud droplet concentrations at a remote ocean site. Geophys. Res. Let., 36, L13812, doi:10.1029/2009GL038465..
Hudson, J. G., S. Noble, V. Jha, and S. Mishra, 2009: Correlations of small cumuli droplet and drizzle drop concentrations with cloud condensation nuclei concentrations, J. Geophys. Res., 114, D05201, doi:10.1029/2008JD010581.
Robinson, N.F., 1984: The efficient numerical calculation of condensational cloud drop growth. J. Atmos. Sci., 41, 697-700.