= Solution
For a small fractional change $\epsilon=\Delta N_d/N_d$, use $\Delta A\simeq A(1-A)\epsilon/3$. With $\epsilon=0.1$ the changes are
$$
\boxed{\Delta A_{\rm marine}\simeq0.00833,\qquad
\Delta A_{\rm continental}\simeq0.00700.}
$$
Thus the respective albedos become approximately \b[0.5083 and 0.7070] in the first-order calculation. These are increases of 0.83 and 0.70 percentage points, not 0.83% and 0.70% relative changes. Relative to the initial albedos, the increases are about 1.67% and 1.00%.
The exact finite-change version of the stipulated optical model is $A'=sA/(1-A+sA)$ with $s=(1.1)^{1/3}$, giving $A'_{\rm marine}=0.50794$ and $A'_{\rm continental}=0.70663$. Its slightly smaller changes are consistent with curvature neglected in the differential approximation. For equal fractional number increases, the cleaner, intermediate-albedo marine <cloud> is somewhat more susceptible; for equal absolute number increases its susceptibility would be much greater because of the extra $1/N_d$ factor. The two starting albedos are treated as the supplied separate baselines, not as an exact common-proportionality calibration of the approximate optical-depth law.
\Image[/past-exam-of-the-chemistry-course-of-the-year-4-of-the-natural-sciences-course-of-the-university-of-cambridge/2022/paper-id-cloud-susceptibility.png]
{title=Original finite-change and tangent predictions for the two specified cloud-albedo baselines}
Back to article page
