Cosmogenic nuclide and solute flux data from central Cuban rivers emphasize the importance of both physical and chemical mass loss from tropical landscapes

Abstract

Abstract. We use 25 new measurements of in situ produced cosmogenic 26Al and 10Be in river sand, paired with estimates of dissolved load flux in river water, to characterize the processes and pace of landscape change in central Cuba. Long-term erosion rates inferred from 10Be concentrations in quartz extracted from central Cuban river sand range from 3.4–189 Mg km−2 yr−1 (mean 59, median 45). Dissolved loads (10–176 Mg km−2 yr−1; mean 92, median 97), calculated from stream solute concentrations and modeled runoff, exceed measured cosmogenic-10Be-derived erosion rates in 18 of 23 basins. This disparity mandates that in this environment landscape-scale mass loss is not fully represented by the cosmogenic nuclide measurements. The 26Al / 10Be ratios are lower than expected for steady-state exposure or erosion in 16 of 24 samples. Depressed 26Al / 10Be ratios occur in many of the basins that have the greatest disparity between dissolved loads (high) and erosion rates inferred from cosmogenic nuclide concentrations (low). Depressed 26Al / 10Be ratios are consistent with the presence of a deep, mixed, regolith layer providing extended storage times on slopes and/or burial and extended storage during fluvial transport. River water chemical analyses indicate that many basins with lower 26Al / 10Be ratios and high 10Be concentrations are underlain at least in part by evaporitic rocks that rapidly dissolve. Our data show that when assessing mass loss in humid tropical landscapes, accounting for the contribution of rock dissolution at depth is particularly important. In such warm, wet climates, mineral dissolution can occur many meters below the surface, beyond the penetration depth of most cosmic rays and thus the production of most cosmogenic nuclides. Our data suggest the importance of estimating solute fluxes and measuring paired cosmogenic nuclides to better understand the processes and rates of mass transfer at a basin scale.

The slow downwearing of Madagascar: Inferring patterns and controls on long‐term basin‐averaged erosion rates from in situ10Be at the catchment and regional level

“…We calculate a chemical weathering flux of 113 (50-194) t km À2 year À1 for that year, suggesting chemical denudation rates of 42 (19-72) mm kyr À1 when assuming that weathering is not iso-volumetric (Text S1.4). These values for the central highlands of Madagascar are in the range of the estimates obtained by Campbell et al (2022) but higher than the rates obtained for Sri Lanka. Even though this estimate entails large uncertainties, it suggests that mass loss through chemical weathering is probably a major (and hitherto unacknowledged) factor for regional denudation in Madagascar.…”

“…(2016) estimated that deep chemical weathering accounts for 60% of the total denudation in a thick, highly weathered, regolith on the Cameroon craton. A recent study from Cuba furthermore indicated significant landscape‐scale mass loss that was not reflected in in situ 10 Be cosmogenic nuclide measurements (4–180 t km −2 year −1 corresponding to 2–69 mm kyr −1 when = 2.6 t m 3 ), with rock dissolution rates (24–154 t km −2 year −1 corresponding to 9–59 mm kyr −1 when = 2.6 t m 3 ) exceeding the latter in 15 of 22 basins (Campbell et al., 2022).…”

“…In situ 10 Be concentrations give an indication of the total denudation (physical erosion and chemical weathering) of the upper metres of regolith, but do not reflect fluxes related to deep chemical weathering beneath the cosmic ray attenuation depth (Bierman & Steig, 1996; Dixon et al., 2009; von Blanckenburg, 2005; Riebe & Granger, 2013). These deep fluxes can be considerable in tropical regoliths (Campbell et al., 2022; Dixon et al., 2009; Riebe & Granger, 2013), such as are found in Madagascar. We therefore refer to ‘ 10 Be erosion rates’ throughout this work to make it explicit that our data do not account for potential mass loss by dissolution deep within the regolith.…”

“…As the thick regolith shields deeper layers (>2 m) from cosmogenic nuclide buildup, material that gets lost by chemical weathering processes below this depth will not be reflected in denudation rates inferred from cosmogenic nuclides. This can result in an underestimation of more than 100% of the actual landscape denudation rates (Dixon et al, 2009;Ott et al, 2022;Riebe & Granger, 2013) corresponding to 9-59 mm kyr À1 when ρ = 2.6 t m 3 ) exceeding the latter in 15 of 22 basins (Campbell et al, 2022).…”

Shaping the Huara Intrusive Complex in the Hyperarid Atacama Desert—Erosional Near‐Stasis Contrasting High Topographic Gradients

“…Basin‐averaged erosion rate estimates from TCNs use an idealized model of erosional processes, which requires a set of assumptions to be fulfilled (Granger et al., 2001; Von Blanckenburg, 2005). However, in extreme environments, such as the Atacama Desert, some of these assumptions might be violated, which, however, can be identified using paired cosmogenic nuclides (Campbell et al., 2022). The following section will consider the sensitivity of TCN‐derived erosion rates to processes particular to hyperarid environments.…”