I am a Postdoctoral Researcher in the Basin Research Group at Bullard Laboratories.

Quantifying the history of uplift of the Earth's surface

This important problem is interesting because a history of uplift constrains the temporal and spatial distribution of tectonic and sub-plate processes (e.g. dynamic support).  

Despite their importance, reliable estimates of surface uplift on tectonically significant timescales and length scales (i.e. 1–100 Ma, 10–1000 km) are difficult to obtain. Denudation is often measured using either radiogenic methods (e.g. cosmogenic surface exposure, fission track analysis) or sedimentary compaction methods (e.g. acoustic velocity calculations). In both cases, there are significant trade-offs between denudation and other parameters (e.g. temperature gradient, initial porosity). Many estimates are spot measurements and have poor spatial resolution, there is a pressing need for methods which allow large-scale patterns of uplift to be identified.

Inverse modeling of longitudinal river profiles 

The goal of my research so far has been to develop inverse algorithms to estimate the history of uplift from observed networks of river profiles. First, a forward model was developed, which calculates river profiles from uplift rate histories. Height variation along a river profile is controlled by uplift rate and moderated by the erosional process. It is assumed that the erosional process can be represented by a combination of advection and diffusion. Secondly, the geologically more interesting problem was posed and solved: which uplift rate history minimises the misfit between observed and calculated river profiles? 

 

Forward modeling of river profiles. (a) Solid circles and line show discretized uplift rate as a function of geologic time, which was uRsed to calculate river profiles; (b) solid line shows cumulative uplift as function of geologic time (i.e., 0t U dt); (c) solid red line is river profile calculated using uplift rate history shown in Figure 3a, where erosive response of river is dominated by knickpoint retreat (i.e., Pe 􏰆 1); (d) erosive response is dominated by a combination of knickpoint retreat and down-wearing (i.e., Pe 􏰄 1); (e)

Figure 1. Forward modeling of river profiles. (a) Solid circles and line show discretized uplift rate as a function of geologic time, which was used to calculate river profiles; (b) solid line shows cumulative uplift as function of geologic time (i.e., ∫ U dt); (c) solid red line is river profile calculated using uplift rate history shown in panel a, where erosive response of river is dominated by knickpoint retreat (i.e., Pe > 1); (d) erosive response is dominated by a combination of knickpoint retreat and down-wearing (i.e., Pe ~ 1); (e) erosive response is dominated by down-wearing (i.e., Pe < 1).

 

 

Africa

River profiles from a series of topographic swells onshore Africa were inverted. Fits between calculated and observed river profiles are excellent. Calculated uplift rate histories suggest that African swells, including the Bie and South African domes, grew rapidly in the last 30–40 million years (Roberts & White, 2010). Uplift histories vary significantly from swell to swell but cumulative uplift histories agree with independent geologic estimates. These independent estimates include sedimentary flux to the coastal ocean around Africa and volcanic activity. 

 

Figure 2. Analysis of Bie Plateau. This topographic swell intersects west African coastline and has a radial drainage pattern. In main plot, dome topography and selected rivers are shown. Labels indicate river profiles inverted in this study (1 indicates Cuvo, 2 indicates Longa, and 3 indicates Cuanza). Each river was inverted 50 times and erosional parameters were assigned random values within bounded ranges of values. (1a) Cuvo river profile analysis. Thick black line and gray envelope show mean uplift rate history and its uncertainty which were calculated by Monte Carlo inverse modeling of river profile. Uplift rate histories within envelope yield acceptable fits to river profile (68% confidence level). (1b) Cumulative uplift history calculated from uplift rate history. (1c) Gray band shows observed river profile; dotted line shows best fitting theoretical river profiles which was calculated by varying uplift rate history. Bar pattern along x axis shows changing river bed lithologies. (2a–2c) Longa river profile analysis. (3a–3c) Cuanza river profile analysis.

 

 

I am also interested the wider problems of coupling mantle convection to the surface processes, the temporal and spatial evolution of mantle convection, geomorphology and quantifying tectonic rates. 

 

Publications

Hartley, R., Roberts, G. G., White, N. & Richardson, C. 2011. Transient convective uplift of an ancient buried landscape. Nature Geoscience, v. 4, doi:10.1038/ngeo1191.

Roberts, G. G. and White, N. 2010. Estimating uplift rate histories from river profiles using African examples. Journal of Geophysical Research, v. 115, B02406, doi:10.1029/2009JB006692.

Pritchard, D., Roberts, G. G., White, N. J. and Richardson, C. N. 2009. Uplift histories from river profiles. Geophysical Research Letters, v. 36, L24301, doi:10.1029/2009GL040928.


Publications: 2006-Present