Cloud properties over the Maritime Continent in ACCESS

Evaluation of cloud properties over the Maritime Continent in ACCESS using CloudSat and CALIPSO simulators

Abstract

This study evaluates cloud properties over the Maritime Continent during the monsoon season from the Australian Community Climate and Earth System Simulator (ACCESS1.3) using the Cloud Feedback Model Intercomparison Project (CFMIP) Observational Simulator Package (COSP). Cloud properties from the model are assessed against observations from CloudSat and CALIPSO cloud fraction, radar reflectivity and lidar scattering ratio. Coupled and uncoupled simulations are compared in order to assess the impact of coupling on the model’s ability to simulate the cloud properties associated with the complex system of islands of the Maritime Continent.

Results from the atmosphere only experiments show that the model reasonably reproduces the total cloud fraction in terms of the land-sea contrast. However, this is mainly due to a better representation of the land-sea contrast from the high and mid-level cloud fractions, although they tend to be overestimated in the upper levels and underestimated below. In contrast, the low-level cloud fraction in the model tends to be significantly underestimated over land.

CloudSat radar reflectivity histograms of hydrometeor occurrence averaged over the region reveal a fairly good agreement between the model and the observations, peaking at the right heights and reflectivities. However the model presents a wide spread of large occurrence of hydrometeors at low-levels, characteristic of drizzle and light rains that does not exist in the observations. In contrast, the occurrence of cloud at mid-levels is significantly underestimated in the model and tends to peak in smaller reflectivity bins than in the observations. This indicates that the ice water content and particle sizes in the clouds are smaller in the model. CALIPSO lidar scattering ratio histograms show that in the model the maxima seen at upper levels are at lower heights, which also suggests that the model underestimates the ice water content in the upper troposphere. At mid-levels in stark contrast with the observations, nearly no occurrences are seen across all bins in the model. These errors in the modeled cloud properties contribute to errors in the vertical heating profiles and surface temperatures over the Maritime Continent.