DEVELOPING TRANSFERABLE SPATIAL PREDICTION MODELS: A CASE STUDY OF SATELLITE BASED LANDCOVER MAPPING

Air temperature is an important baseline parameter for terrestrial Antarctica in the context of patterns and processes in climatology, hydrology or ecology. In this paper, we present AntAir, a new dataset of gridded air temperatures in 1 km spatial and daily temporal resolution that is available since 2003. AntAir was created by modelling daily air temperature from MODIS satellite-based land surface temperature using machine learning algorithms and measurements from 70 weather stations.

An emissions inventory is a compilation of the mass of pollutants released by different sources. The quantification of vehicular emissions is difficult because these sources are in movement across streets. Also, emissions processes are multiple and complex. In this paper, we present an open-source software for calculating spatial vehicular emissions, including exhaust, evaporation and wear, named VEIN. The software is an R package available at https://github.com/atmoschem/vein.

A spatially explicit mapping of rainfall is required for southern Africa but obtaining accurate estimates is still a challenging task. We estimated hourly rainfall based on optical satellite data and neural networks. The results indicated that the majority of rainfall events could be captured by the model, but with a clear tendency to overestimate rainfall. Despite being a comparably simple approach, the presented rainfall retrieval could outperform a complex global rainfall product.