Uncertainties in crustal thickness models for data sparse environments: A review for South America and Africa

With the recently available high resolution gravity data from the GOCE satellite a whole range of crustal thickness models have been derived. The added value of GOCE is that it provides data globally, including regions that are poorly covered by seismological studies, like large parts of Africa and South America. Potentially these models can provide new insight in crustal structure for these data poor regions. We compare different models of crustal thickness for South America and Africa and attempt to assess the quality of different modelling techniques and the impact of different data sources. We introduce one new global crustal thickness model based on gravity data, DMM-1, and use seven additional, recently published, continental or global crustal thickness models based on gravity or seismological data. All models use different modelling techniques, and either gravity (four models) or seismological data (four models). We will show that significant differences exist between the models but that these cannot be directly related to the used data. Choices made in the selection and parametrization of the various modelling techniques have more impact than using different data sources including data sources of supposed higher quality. The significant differences, up to 28 km, between models can have a major influence on geodynamical analysis for the two continents. We propose that future work should focus on developing a standard for modelling in data sparse environments, and expanding seismological efforts in those regions that are most different between the shown models to verify the actual crustal thickness. Furthermore, the contribution and inclusion of (satellite) gravity data in crustal thickness models should be further explored. (C) 2014 Elsevier Ltd. All rights reserved.