Jennifer Adams

Validation of satellite-based retrievals of land surface albedo using in-situ measurements is essential to identify differences between them, to improve retrieval algorithms and to assess conformity to accuracy requirements. Differences between in-situ and satellite-based retrievals depend on the actual difference and their associated uncertainties, where it is crucial that the uncertainties of both can be computed to properly understand potential differences. This study introduces a model-based framework for assessing the quality of in-situ albedo measurements. A 3D Monte Carlo Ray Tracing (MCRT) radiative transfer model is used to simulate field measurements of surface albedo, and is able to identify and quantify potential sources of error in the field measurement. Compliance with the World Meteorological Organisation (WMO) requirement for 3% accuracy is tested.

Keywords: Quality assessment Model-based framework Radiative transfer Monte Carlo ray tracing In situ Albedo a b s t r a c t Satellite-based retrievals of land surface albedo are essential for climate and environmental modelling communities. To be of use, satellite-retrievals are required to comply to given accuracy requirements, mainly achieved through comparison with in situ measurements. Differences between in situ and satellite-based retrievals depend on their actual difference and their associated uncertainties. It is essential that these uncertainties can be computed to properly understand the differences between satellite-based and in situ measurements of albedo, however quantifying the individual contributions of uncertainty is difficult. This study introduces a model-based framework for assessing the quality of in situ albedo measurements. A 3D Monte Carlo Ray Tracing (MCRT) radiative transfer model is used to simulate field measurements of surface albedo, and is able to identify and quantify potential sources of error in the field measurement. Compliance with the World Meteorological Organisation (WMO) requirement for 3% accuracy is tested. 8 scenarios were investigated, covering a range of ecosystem types and canopy structures, seasons, illumination angles and tree heights. Results indicate that height of measurement above the canopy is the controlling factor in accuracy, with each canopy scenario reaching the WMO requirement at different heights. Increasing canopy hetero-geneity and tree height noticeably reduces the accuracy, whereas changing seasonality from summer to winter in a deciduous forest increases accuracy. For canopies with a row structure, illumination angle can significantly impact accuracy as a result of shadowing effects. Tests were made on the potential use of multiple in situ measurements, indicating considerably increased accuracy if two or more in situ measurements can be made.

—Land surface albedo defines the fraction of shortwave radiation reflected by the Earth's surface and controls the surface energy balance; thus, it is important for environmental and climate scientific communities. Remote sensing is the only means to globally map land surface albedo, however for it to be of use to the aforementioned communities, it must be accurate with respect to Global Climate Observing System (GCOS) requirements. Sources of error are introduced in each step of the provision of land surface albedo products, whereby this letter intends to investigate sources of error introduced by the narrow-band-to-broad-band conversion formula step. The radiative transfer modeling of vegetation is used to simulate spectral albedo over complex 3-D vegetation canopies; then narrow-band-to-broad-band conversion formulas for numerous sensors are applied on the spectral albedo to compute the broad-band albedo (BBA), and the accuracy of formulas is investigated. Results indicate that the effectiveness of conversion formulas is determined by the sensor, depending on the placement and number of the sensor wavebands, the ecosystem complexity, and the broad-band range of the BBA.

Ecosystem changes can have fundamental impacts on climate, therefore determining how much shortwave radiation that is absorbed and reflected from vegetation canopies can help understand and predict near-surface climate. The shortwave radiation budget is a function of the coupling between canopy structure and leaf biochemistry. In order to interpret a remotely sensed signal from vegetation canopies, scattering must be modelled at both the leaf and canopy level, complicating the retrieval of leaf biochemistry. This study empirically tests a new approach to model both canopy and leaf scattering using a Directional Area Scattering Factor (DASF) to correct reflectance (BRF) data modelled by Monte Carlo Ray Tracing simulations of different canopies for structural effects to predict total canopy scattering.

It was shown that both DASF and total canopy scattering could be accurately extracted under idealised conditions of directional-hemispherical reflectance, equal leaf asymmetry and sufficiently dense canopies with black soil. Additionally it was proven that under these conditions total canopy scattering could be predicted using information from the 710-790nm region and given no prior knowledge of leaf optical
properties. However, the latter highlighted important consequences of no prior knowledge of leaf optical properties; if the leaf single scattering albedo is not known, the recollision probability cannot be found, and vice versa. This is significant since recollision probability is widely used in canopy reflectance modelling.

Departures from these idealised conditions; varying view geometry, bi-directional reflectance, LAI and soil effects, were tested. Canopy scattering was extracted stably under conditions of varying view geometry and bi-directional reflectance, but LAI and soil effects were proven to influence the accuracy with which canopy scattering can be modelled using this approach. Canopy architecture, described by homogeneous and heterogeneous canopies has important influences over DASF and consequently the accuracy of retrieval of total canopy scattering.