1. What channels does M-COP use?
M-COP uses multiple MSI channels in visible/shortwave infrared and thermal range of the electromagnetic spectrum. These channels help derive cloud optical thickness (M-COT), effective cloud radius (M-REF), and multiple estimates of cloud height (M-CTT, M-CTH, M-CTP) as output products. The algorithm is embedded in the M-CLD processor framework and has been implemented from scratch following a mathematically optimal estimation approach. An overview of the workflow is given in Figure 1. The main idea is to compare simulated reflectance and radiance values for a wide range of possible atmospheric conditions with the measured values from the sensor. The simulations, which build the forward model of the retrieval, are partly created in advance with a radiative transfer model (RTM) and then stored in look-up-tables. An optimal estimation (OE) inversion technique is used to find the best estimate of the cloud properties by an iterative retrieval loop, which compares the forward model result of an assumed state vector with the observation by taking into account the respective uncertainties. The OE technique minimizes a cost function when both the forward model output of the assumed state vector and the observation vector are close enough to meet specific requirements. The OE can also include a-priori knowledge of the to-derived products. The algorithm uses multiple MSI channels to derive cloud properties and incorporates an optimal estimation technique for accurate retrieval.
read more
2. How are MSI L1c product measurements calibrated and geolocated?
The MSI L1c products are calibrated and geolocated instrument data. The measurements for solar channels are given in radiance, while the three terrestrial channels are in brightness temperature. The L1c product is re-sampled onto the grid of a reference TIR channel, ensuring each channel is mapped on the same pixel. This is crucial for the M-CLD processor, which utilizes the combination of all channels. The algorithm calculates the reflectances at the top of the atmosphere in the solar channels using input parameters such as measured radiance and corresponding inband solar irradiance. The reflectances (R(th0, th, ph)) of each channel i are obtained from the measured radiance (L) and the inband solar irradiance (E0,icos(th0)), where i represents the different channels (V IS, N IR, SW IR1, SW IR2). These measurements depend on the solar zenith angle (th0), viewing zenith angle (th), and relative azimuth angle (ph), which is defined by the difference between the sun and instrument viewing azimuth angles.
read more
3. What background information is required for retrieval?
The retrieval requires background information on atmospheric and surface conditions. This includes the long-term albedo climatology from the MODIS science team and the assumption that the white-sky albedo product is best suited for cloudy atmospheres. Over ocean, the surface albedo is assumed to be 0.05 at both 0.6 um and 1.6 um. For the forward model in the longwave channels, land surface emissivity values are used from the MODIS land surface temperature and emissivity products, which provide per-grid temperature and emissivity values.
read more
4. What is the purpose of the forward model operator?
The forward model operator simulates observed radiance at the sensor based on known atmospheric state and auxiliary input parameters. It is channel-based, with independent simulations for each channel. The operator includes uncertainties in auxiliary data and separates the solar spectrum from the thermal emission-dominated spectrum. This model is crucial for accurate atmospheric analysis and remote sensing applications.
read more