MAMU: an R package for GIS-based river terrace mapping, morphostratigraphic evaluation of terrace maps and outcrop data and river long profile modelling

Abstract

River terraces form as a response to regional uplift and climatic fluctuations (Bridgland & Westaway, 2008). While the intensity of climatic oscillations controls trends of aggradation and incision, it is uplift that determines the amplitude of the vertical spacing between consecutive terraces (Demoulin et al., 2017). Glacial and periglacial processes may amplify these trends providing distinct stratigraphic markers on climate and tectonic processes in the fluvial terrace record. Over several glacial-interglacial cycles, terraces can form complex sequences often referred to as a river 'terrace staircase'. Terrace staircases are not necessarily limited to mid- and high latitudes and do often lack (consistent) age models, as accurate radiometric and biostratigraphic time constraints are lacking when records extend into the Mid- and Early Pleistocene. Morphostratigraphy can thereby add valuable information if data on terrace strath (base unconformity) or tread (top unconformity) elevation are available. However, elevation information alone may not be sufficient. Especially for old terraces, that may be patchily preserved, a synopsis of sedimentologic and age data in a morphologic context may be necessary, to uncover stratigraphic questions. We present a GIS based toolset for R, that is designed to i) detect and map potential terrace surfaces from digital elevation data (3D view, mapping), ii) statistically evaluate potential terraces together with additional geological information along 2D profiles (2D view, modelling), iii) transform resulting models in a map view. The toolset allows a semiautomated workflow, optimized, to deliver quick results, enabling mapping and correlation of terraces at mountain range scale. Central part for data evaluation and illustration are 2D profiles. To minimize potential projection artifacts, the profile lines are optimized in a 3D view, by detecting the orientation with best correlation of terrace top elevation data, indicative for paleo-flow and thus ideal local profile line orientation. In the 2D view terrace elevation data is statistically evaluated and models, including error estimation, are fitted to each terrace stratigraphic unit. Additional control is contributed via including outcrop and geologic map information in the profile views. We tested the toolset in the North Alpine Foreland, where more than a century of extensive Quaternary research lead to a vast resource of available geodata and detailed terrace stratigraphic maps. Terraces of up to postulated Early Pleistocene age are partly preserved well and over large areas. However, despite the abundance of data, stratigraphic inconsistencies exist in the current foreland wide terrace stratigraphic model. These need to be addressed, when using local terrace staircases as an archive of geodynamic information. This qualifies the North Alpine Foreland as an ideal test site for our code.