Considerable progress has been made in modelling the response of rivers to tectonic perturbation in order to decode the tectonic signals embedded in river long profiles and planform geometry. Whilst studies have showed the importance of rock type on the morphology of rivers responding to tectonics on a local scale, these effects are often not captured in landscape evolution models. In fact, current models of fluvial response to tectonic perturbation such as active faulting require carefully collected data sets to fully constrain or calibrate key parameters, including the effect of bedrock lithology on substrate erodibility and timescales for tectonic signal propagation in bedrock river systems. Here we constrain the role of bedrock in controlling fluvial incision for a 240 km 2 catchment draining into the Gulf of Corinth, which has excellent tectonic constraints and a variety of bedrock lithologies. An active normal fault at the downstream end of the catchment (the East Eliki Fault) is known to have initiated at 0.7 Ma, with average Quaternary uplift and incision rates of 1.00-1.25 mm/yr. The initiation of the East Eliki Fault is recorded in the river as a prominent knickzone 7-16 km upstream of the fault. Detailed field data collected within this catchment at 500 m intervals along the main channel length at this tectonically wellconstrained field site show an order of magnitude increase in river channel slopes (y/x = 0.02 to 0.18) and stream powers (1 kWm -2 to 27 kWm -2 ) at the lithological boundary between weak and resistant bedrock. The weak conglomerates and strong limestones have Schmidt hammer compressive strengths of ca. 30 and 50 respectively, based on in-situ rock strength measurements. Consequent erodibility values of 1.8 ± 0.3 × 10 -14 and 5 -6 ± 2 × 10 -15 ms 2 kg -1 show a factor of three to four decrease in erodibility for a two-fold increase in rock strength. A simple simulation of tectonic signal propagation through the river catchment based on these erodibility values indicates that the observed plan view position of the knickpoint is consistent with our calculated knickpoint positions based on erodibility values derived from incision rate and stream power data. However, the tectonic signal associated with faulting would have propagated completely through the catchment within 1 Ma if it were entirely composed of weak rock for similar climatic and tectonic conditions, indicating that lithology has a major control on landscape response times.Our results help constrain the effect of lithology on river channel geometry and allow erosional parameters to be derived that are crucial for effective modelling of tectonic rates from topography.

Highlights 9• We estimate rock strength, erodibility and drainage divide mobility in the High Atlas 10Mountains 11• The weakest rock-type in the High Atlas is up to two orders of magnitude more erodible than 12 the strongest 13• In gently deformed horizontal strata of the sedimentary cover the drainage divide is mobile 14• Faulted and folded metamorphic sedimentary bedrock coincide with a stable drainage divide 15• Exhumation of crystalline basement forces the drainage divide into the centre of exposed 16 basement 17 Abstract 18 Numerical model simulations and experiments have suggested that when migration of the main 19 drainage divide occurs in a mountain belt, it can lead to the rearrangement of river catchments, 20 rejuvenation of topography, and changes in erosion rates and sediment flux. We assess the 21 progressive mobility of the drainage divide in three lithologically and structurally distinct groups of 22 bedrock in the High Atlas (NW Africa). The geological age of bedrock and its associated tectonic 23 architecture in the mountain belt increases from east to west in the study area, allowing us to 24 track both variations in rock strength and structural configuration which influence drainage 25 mobility during erosion through an exhuming mountain belt. Collection of field derived 26 measurements of rock strength using a Schmidt hammer and computer based extraction of river 27 channel steepness permit estimations of contrasts in fluvial erodibilities of rock types. The 28 resulting difference in fluvial erodibility between the weakest and the strongest lithological unit is 29 up to two orders of magnitude. Published evidence of geomorphic mobility of the drainage divide 30 indicates that such a range in erodibilities in horizontal stratigraphy of the sedimentary cover may 31 lead to changes in erosion rates as rivers erode through strata, leading to drainage divide 32 migration. In contrast, we show that the faulted and folded metamorphic sedimentary rocks in the 33 centre of the mountain belt coincide with a stable drainage divide. Finally, where the strong 34 igneous rocks of the crystalline basement are exposed after erosion of the covering meta-35 sediments, there is a decrease in fluvial erodibility of up to a factor of three, where the drainage 36 divide is mobile towards the centre of the exposed crystalline basement. The mobility of the 37 drainage divide in response to erosion through rock-types and their structural configuration in a 38 mountain belt has implications for the perception of autogenic dynamism of drainage networks 39 and fluvial erosion in mountain belts, and the interpretation of the geomorphology and 40 downstream stratigraphy. 41 42

Numerical model simulations and experiments have suggested that when migration of the main drainage divide occurs in a mountain belt, it can lead to the rearrangement of river catchments, rejuvenation of topography, and changes in erosion rates and sediment flux. We assess the progressive mobility of the drainage divide in three lithologically and structurally distinct groups of bedrock in the High Atlas (NW Africa). The geological age of bedrock and its associated tectonic architecture in the mountain belt increases from east to west in the study area, allowing us to track both variations in rock strength and structural configuration which influence drainage mobility during erosion through an exhuming mountain belt. Collection of field derived measurements of rock strength using a Schmidt hammer and computer based extraction of river channel steepness permit estimations of contrasts in fluvial erodibilities of rock types. The resulting difference in fluvial erodibility between the weakest and the strongest lithological unit is up to two orders of magnitude. Published evidence of geomorphic mobility of the drainage divide indicates that such a range in erodibilities in horizontal stratigraphy of the sedimentary cover may lead to changes in erosion rates as rivers erode through strata, leading to drainage divide migration. In contrast, we show that the faulted and folded metamorphic sedimentary rocks in the centre of the mountain belt coincide with a stable drainage divide. Finally, where the strong igneous rocks of the crystalline basement are exposed after erosion of the covering meta-sediments, there is a decrease in fluvial erodibility of up to a factor of three, where the drainage divide is mobile towards the centre of the exposed crystalline basement. The mobility of the drainage divide in response to erosion through rock-types and their structural configuration in a mountain belt has implications for the perception of autogenic dynamism of drainage networks and fluvial erosion in mountain belts, and the interpretation of the geomorphology and downstream stratigraphy.

No abstract

Much has yet to be learned of the spatial patterning of pre-Columbian people across the Tropical Andes. Using compiled archaeological data and a suite of environmental variables, we generate an ensemble species distribution model (SDM) that incorporates general additive models, random forest models and Maxent models to reconstruct spatial patterns of pre-Columbian people that inhabited the Tropical Andes east of the continental divide, within the modern countries of Bolivia, Peru and Ecuador. Within this region, here referred to as the eastern Andean flank, elevation, mean annual cloud frequency, distance to rivers and precipitation of the driest quarter are the environmental variables most closely related to human occupancy. Our model indicates that 11.04% of our study area (65 368 km 2 ) was likely occupied by pre-Columbian people. Our model shows that 30 of 351 forest inventory plots, which are used to generate ecological understanding of Andean ecosystems, were likely occupied in the pre-Columbian period. In previously occupied sites, successional trajectories may still be shaping forest dynamics, and those forests may still be recovering from the ecological legacy of pre-Columbian impacts. Our ensemble SDM links palaeo- and neo-ecology and can also be used to guide both future archaeological and ecological studies. This article is part of the theme issue ‘Tropical forests in the deep human past’.