A 2014 Landsat analysis of 20 Amazon Basin reaches (Constantine, Dunne, Ahmed, Legleiter and Lazarus, Nature Geoscience 7, 899–903) reported that rivers with higher total suspended sediment (TSS) flux migrate and cut off faster, and inferred that imposed sediment supply modulates floodplain reshaping by meandering rivers. The inference rests on TSS as a proxy for the bed-material load that actually builds point bars, because reach-scale bed-material flux had been measured for only one season in a single Amazonian tributary. Later work has strengthened the pattern globally with modeled sediment fluxes (Greenberg and Ganti 2024), linked supply to sinuosity adjustment via point-bar growth (Ahmed, Constantine and Dunne 2019), and identified floodplain slope as a first-order control on cutoff type basin-wide (Li et al. 2026), while rival analyses identify discharge variability as a primary predictor of river mobility (Wu et al. 2025; Leenman et al. 2025). What remains open is the direct verification: whether measured bed-material supply, not suspended-load proxy or confounding hydrologic controls, drives migration, cutoff, and oxbow-mediated floodplain sediment storage across the Amazon, through what mechanism, and how these responses adjust when dams interrupt the supply.

Background

Meandering rivers wander across their floodplains by eroding the outer (concave) bank of each bend while depositing sediment on an inner-bank point bar — a sediment body attached to the convex bank. Whether the rate of this lateral migration is controlled by the supply of sediment delivered from upstream (an 'imposed' environmental variable set by mountain erosion, climate, land use, or dams), rather than only by channel hydraulics, is a long-standing question in fluvial geomorphology. The distinction matters because the material that builds point bars is the bed-material load — coarse sediment (sand and gravel) transported as bedload on the bed or in suspension near it — whereas most large-river sediment records measure only total suspended sediment (TSS), which is dominated by silt–clay washload that transits the channel without depositing on bars.

The Amazon Basin is the key natural laboratory: its tributaries are largely free of engineering works and span an order-of-magnitude gradient in sediment supply, from Andean rivers delivering on the order of 100 Mt of suspended sediment per year to sediment-starved shield-draining rivers. Constantine, Dunne, Ahmed, Legleiter and Lazarus (2014) digitized channel centrelines from Landsat images (three to five per river, 1985–2013) for 20 reaches spanning the Andes–Foreland Basin, the Central Amazon Trough, and the Guiana/Brazil shield provinces. They found that width-normalized meander migration rate increases with TSS flux per unit channel width (MR=0.043⋅TSS0.28M_R = 0.043 \cdot TSS^{0.28}, with MRM_R the mean annual migration rate in channel widths per year, ch-w yr−1^{-1}, and TSS flux in megatonnes per year per metre of channel width, Mt yr−1^{-1} m−1^{-1}, both normalized by RWR_W), that cutoff rate increases with migration rate (CR≈0.07MR1.29C_R \approx 0.07 M_R^{1.29}, with CRC_R the number of cutoffs per year per channel width of down-valley distance, no. yr−1^{-1} ch-w−1^{-1}, calculated as (no.+1)/NN for no. cutoff events observed over the NN years between recorded images; a cutoff is the abandonment of a bend — by neck cutoff, when the two sides of a bend neck intersect, or by chute cutoff, when a flood carves a new channel across the floodplain), and that the density of oxbow lakes — the abandoned bends left after cutoff, counted as number per channel width of down-valley distance (NL=5.1MR0.98N_L = 5.1 M_R^{0.98}; NL=0.48⋅TSS0.46N_L = 0.48 \cdot TSS^{0.46}) — increases with both. Because cutoffs hold the channel sinuosity TT (channel length divided by valley length) roughly constant while migration continually abandons bends as new oxbow lakes, they argued that the rate of production of valley accommodation space for floodplain sediment storage, SV=CR L RW RDS_V = C_R \, L \, R_W \, R_D, increases with sediment supply. This relation is evaluated with all lengths in channel-width units — mean oxbow length L≈3.0e0.82TL \approx 3.0 e^{0.82T} in ch-w, from a worldwide survey of 911 oxbows (Constantine and Dunne 2008); oxbow depth RD≈0.05RWR_D \approx 0.05 R_W, from Solimões–Amazon floodplain observations (Mertes, Dunne and Martinelli 1996); and RW=1R_W = 1 ch-w, the bankfull channel width — so that SVS_V is in ch-w3^3 yr−1^{-1} per ch-w of down-valley distance, and it holds where a central tendency exists for oxbow lengths and oxbow bankfull width and depth approximate those of the channel at the time of cutoff. Most of that accommodation space is filled by wash load where neck cutoff is commonplace and by variable amounts of bed material where chute cutoff occurs. Supply-rich Andean rivers therefore continually manufacture their own floodplain storage in a plausible steady state of oxbow production and removal.

Three caveats qualify this chain of inference. First, TSS was used as a proxy for bed-material load; when the study was published, direct bed-material flux measurements existed for only one season in a single Amazonian reach, with regional estimates suggesting bedload is under 1% of total load and suspended sand about 20% of TSS. Second, the Landsat-derived migration rates are minima, because channel shift may have reversed direction between the sparsely sampled images. Third, the mechanism linking supply to mobility — point-bar growth steering high-momentum flow toward the outer bank and raising boundary shear stress there — was inferred, not isolated experimentally, and rival controls such as discharge variability and bank erodibility were not partialed out.

Candidate routes to that verification, none of them prescribed, include measuring bed-material flux (bedload plus suspended sand) at reach scale across tributaries spanning the basin's sediment-supply gradient — Andean, Central Amazon Trough, and shield sources — over representative parts of the annual hydrograph and re-testing the relations above against those direct loads; suspended-sand accounting and transport modelling; natural perturbation experiments on the already-dammed Andes–Foreland reaches; point-bar dating; and remote-sensing of bar growth.

Problem Statement

Establish whether imposed sediment supply — specifically the reach-scale bed-material load (bedload plus suspended sand) that builds point bars, not the silt–clay-dominated total suspended sediment used as a proxy — controls meander migration, meander cutoff, and the resulting floodplain sediment-storage capacity of Amazon Basin rivers. Concretely, the open problem comprises three jointly linked and independently refutable sub-claims of the source study's single claim — that sediment supply modulates the reshaping of floodplain environments by meandering rivers in the Amazon:

(1) Whether the planform relations reported from the Landsat analysis are an artefact of the suspended-load proxy and of undersampled imagery — that is, whether they survive replacement of the silt–clay-dominated TSS flux with a direct bed-material load (bedload plus suspended sand), correction of migration-rate minima from sparsely sampled imagery, and partialling out confounding controls beyond the channel gradient and discharge magnitude the source excluded (discharge variability, floodplain slope, vegetation and bank strength). The relations to be tested, in the source's normalizations, are migration rate versus load (MR=0.043⋅TSS0.28M_R = 0.043 \cdot TSS^{0.28}, MRM_R in ch-w yr−1^{-1}, TSS flux in Mt yr−1^{-1} m−1^{-1}), cutoff rate versus migration (CR≈0.07MR1.29C_R \approx 0.07 M_R^{1.29}, CRC_R in no. yr−1^{-1} ch-w−1^{-1} of down-valley distance), oxbow density (NL=5.1MR0.98N_L = 5.1 M_R^{0.98}; NL=0.48⋅TSS0.46N_L = 0.48 \cdot TSS^{0.46}, in no. ch-w−1^{-1}), and accommodation-space production SV=CRLRWRDS_V = C_R L R_W R_D in ch-w3^3 yr−1^{-1} per ch-w of down-valley distance, each evaluated with the bed-material load substituted wherever the source used the TSS proxy.

(2) Whether the mechanism the source asserts operates — that added bed-material load promotes point-bar growth, which advects high-momentum fluid outward, raises boundary shear stress along the outer (concave) bank, and thereby promotes fluvial entrainment of bank material and bar aggradation — rather than an alternative supply pathway such as bank-strength modification, together with whether the source's distinction between the roles of chute and neck cutoff in filling the newly created accommodation space holds (occupied mostly by wash load where neck cutoff is commonplace, and by variable amounts of bed material where chute cutoff occurs).

(3) How migration, cutoff, and oxbow-mediated storage respond when supply is perturbed — in particular by dam interception of Andean sediment — and whether oxbow production and alluviation then re-establish the quasi-steady storage state the source inferred only from the absence of a secular trend in oxbow density. This sub-claim follows the implication the source itself flagged: the great sensitivity of the sediment-rich Andes–Foreland rivers to disruption of sediment supplies by large water-storage reservoirs.

All three sub-claims are to be addressed, and the task is resolved by a written, evidence-based determination that reports, for each, whether the source relation or mechanism is supported, refuted, or still undetermined by the available measurements, models, and remote-sensing records, and that names the decisive measurement or analysis which would settle any sub-claim left undetermined. Sub-claim (1) is the mandatory core, because it alone decides whether the reported relations survive replacement of the suspended-load proxy; sub-claims (2) and (3) frame the mechanism and the perturbation response and may be reported as supported, refuted, or undetermined. No specific field campaign is required: the sub-claims are to be settled from the evidence and methods available.

References

  1. Constantine, J. A., Dunne, T., Ahmed, J., Legleiter, C. & Lazarus, E. D. Sediment supply as a driver of river meandering and floodplain evolution in the Amazon Basin. Nature Geosci. 7, 899–903 (2014). doi:10.1038/ngeo2282, https://doi.org/10.1038/ngeo2282
  2. Constantine, J. A. & Dunne, T. Meander cutoff and the controls on the production of oxbow lakes. Geology 36, 23–26 (2008). doi:10.1130/g24130a.1, https://doi.org/10.1130/g24130a.1
  3. Ahmed, J., Constantine, J. A. & Dunne, T. The role of sediment supply in the adjustment of channel sinuosity across the Amazon Basin. Geology 47, 807–810 (2019). doi:10.1130/g46319.1, https://doi.org/10.1130/g46319.1
  4. Greenberg, E. & Ganti, V. The pace of global river meandering influenced by fluvial sediment supply. Earth Planet. Sci. Lett. 634, 118674 (2024). doi:10.1016/j.epsl.2024.118674, https://doi.org/10.1016/j.epsl.2024.118674
  5. Li, Y., Edmonds, D. A., Konsoer, K. M., Constantine, J. A., Jing, Y., Shoemaker, D. T. & Islam, M. M. Floodplain slope organizes chute and neck meander cutoffs across rivers in the Amazon basin. Science Advances 12, eaef6658 (2026). doi:10.1126/sciadv.aef6658, https://doi.org/10.1126/sciadv.aef6658
  6. Geyman, E. C., Ke, Y., Magyar, J. S., Reahl, J. N., Soldano, V., Brown, N. D., West, A. J., Fischer, W. W. & Lamb, M. P. Scaling laws for sediment storage and turnover in river floodplains. Science Advances 11, eadu8574 (2025). doi:10.1126/sciadv.adu8574, https://doi.org/10.1126/sciadv.adu8574
  7. Greenberg, E., Chadwick, A. J., Li, G. & Ganti, V. Quantifying Channel Mobility and Floodplain Reworking Timescales Across River Planform Morphologies. Geophys. Res. Lett. 51, e2024GL108537 (2024). doi:10.1029/2024GL108537, https://doi.org/10.1029/2024GL108537
  8. Wu, C., Kim, W., Yang, S., Tsai, F. T.-C., Nittrouer, J. A., Dong, T. Y., Keum, D. & Straub, K. M. High variability in flood discharge and stage accelerates river mobility. Science Advances 11, eadv7637 (2025). doi:10.1126/sciadv.adv7637, https://doi.org/10.1126/sciadv.adv7637
  9. Leenman, A., Greenberg, E., Moulds, S., Wortmann, M., Slater, L. & Ganti, V. Accelerated River Mobility Linked to Water Discharge Variability. Geophys. Res. Lett. 52, e2024GL112899 (2025). doi:10.1029/2024GL112899, https://doi.org/10.1029/2024GL112899
  10. Latrubesse, E. M., Arima, E. Y., Dunne, T., Park, E., Baker, V. R., d'Horta, F. M., Wight, C., Wittmann, F., Zuanon, J., Baker, P. A., Ribas, C. C., Norgard, R. B., Filizola, N., Ansar, A., Flyvbjerg, B. & Stevaux, J. C. Damming the rivers of the Amazon basin. Nature 546, 363–369 (2017). doi:10.1038/nature22333, https://doi.org/10.1038/nature22333
  11. Yuan, C., Liu, L., Liu, K., Shi, L., Li, T., Chen, Y. & Song, C. Automatic Extraction of Oxbow Lakes from Landsat Imagery Using Deep Learning Across the Amazon Basin. Remote Sens. 18, 2472 (2026). doi:10.3390/rs18152472, https://doi.org/10.3390/rs18152472
  12. Mertes, L. A. K., Dunne, T. & Martinelli, L. A. Channel-floodplain geomorphology along the Solimões-Amazon River, Brazil. Geol. Soc. Am. Bull. 108, 1089–1107 (1996). doi:10.1130/0016-7606(1996)108<1089:cfgats>2.3.co;2, https://doi.org/10.1130/0016-7606(1996)108<1089:cfgats>2.3.co;2