The anvils of deep convective clouds (DCCs) have an important impact on global radiation balance. While the anvil cloud area feedback to warming temperatures is expected to have a cooling effect, it has the largest uncertainty of any cloud-climate feedback. Differences in anvil structure contribute to this uncertainty due to changes in the proportions of thicker, cooling anvil and thinner, warming anvil cirrus. A lack of long-term observational datasets of both convective and anvil properties of DCCs has limited our understanding of the connections between these processes.

Using a novel cloud tracking algorithm we detect and track the developing cores, thick and thin anvils of DCCs seen in 5 years of GOES-16 imagery, allowing investigations of their properties throughout the DCC lifecycle. Using this dataset, we compare how the amount of thin anvil cirrus changes with the intensity and organisation of observed DCCs. Previous studies of anvil structure have found that the proportion of thin cirrus increases with convective intensity across a range of regimes. We find that the thin anvil proportion increases with convective intensity both in area and lifetime. To the contrary, for more organised DCCs – those with more cores – we find, however, that the thin anvil area and lifetime both decrease as a proportion of the total anvil. While more intense DCCs have shorter growing phases and longer dissipating phases, the opposite is true for more organised DCCs. These differences in lifecycle have an important impact on thin anvil proportion. The contrast in structure and lifecycle between DCCs with increasing intensity and increasing organisation occurs despite both convective processes having positive impacts on the total anvil area, lifetime and temperature. As both the intensity and organisation of DCCs are expected to increase with warming, we may expect differences in anvil cloud area feedback between different regimes depending on the occurrence of isolated or organised DCCs.