Juan Felipe Paniagua-Arroyave
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Juan Felipe Paniagua-Arroyave has a total of 44 co-authors. Here are the top co-authors by number of shared publications:
- Jaap H. Nienhuis
Department of Physical Geography Utrecht University Utrecht The Netherlands
7 shared publications - Frances E. Dunn
Utrecht University, Utrecht University, Heidelberglaan 8, Utrecht, Netherlands [NL
5 shared publications Peter N. Adams
Department of Geological Sciences, University of Florida, 241 Williamson Hall, Gainesville, FL 3261, USA
5 shared publications- Torbjörn E. Tornqvist
Department of Earth and Environmental Sciences Tulane University New Orleans LA USA
4 shared publications Arnoldo Valle-Levinson
Department of Civil and Coastal Engineering, University of Florida, 310 Weil Hall, Gainesville, FL 32611, USA
3 shared publications- Sabrina M. Parra
Department of Civil and Coastal Engineering University of Florida Gainesville FL USA
3 shared publications
Latest Publications
- Open Access
- 16 October 2026
- Published by American Geophysical Union (AGU) in AGU Advances
- Vol. 7 (7)
Global River Sediment Retention on Delta Plains
Fluvial sediments have helped form approximately 830,000 km 2 of coastal deltaic land, a large part of which since the Holocene slowdown of global sea‐level rise. The rate at which deltas have retained fluvial sediment to build deltaic land, however, has varied greatly between different deltas. Sources of variability as well as global estimates of sediment retention remain poorly resolved, limiting the assessment of future delta change. Here, we compare delta area and thickness against fluvial sediment supply to derive estimates of delta plain sediment retention for ∼1,700 deltas, representing 93% of the global fluvial sediment supply. We find that 28% ± 14% (standard error of the mean) of the global fluvial sediment discharge is retained on delta plains, totaling 5.8 Gt per year, and equating to roughly 4.3 mm yr −1 as a global average sedimentation rate. The average delta, however, retains much less, about 4% ± 1%, because many smaller deltas do not retain much sediment. We observe a large difference between active and passive margins, with sediment retention much higher along the latter. Tide‐dominated deltas have the most efficient morphology, and retain 11% ± 4% on average but up to 70% when weighted by sediment flux. This is much higher than the 7% ± 2% and 2% ± 1% retained by river‐ and wave‐dominated deltas, respectively. As larger deltas retain a greater proportion of their supply, this study suggests a positive feedback that enables larger deltas to grow faster than smaller ones.
- 16 October 2026
- Published by American Geophysical Union (AGU) in AGU Advances
- Vol. 7 (7)
Open AccessGlobal River Sediment Retention on Delta Plains
Fluvial sediments have helped form approximately 830,000 km 2 of coastal deltaic land, a large part of which since the Holocene slowdown of global sea‐level rise. The rate at which deltas have retained fluvial sediment to build deltaic land, however, has varied greatly between different deltas. Sources of variability as well as global estimates of sediment retention remain poorly resolved, limiting the assessment of future delta change. Here, we compare delta area and thickness against fluvial sediment supply to derive estimates of delta plain sediment retention for ∼1,700 deltas, representing 93% of the global fluvial sediment supply. We find that 28% ± 14% (standard error of the mean) of the global fluvial sediment discharge is retained on delta plains, totaling 5.8 Gt per year, and equating to roughly 4.3 mm yr −1 as a global average sedimentation rate. The average delta, however, retains much less, about 4% ± 1%, because many smaller deltas do not retain much sediment. We observe a large difference between active and passive margins, with sediment retention much higher along the latter. Tide‐dominated deltas have the most efficient morphology, and retain 11% ± 4% on average but up to 70% when weighted by sediment flux. This is much higher than the 7% ± 2% and 2% ± 1% retained by river‐ and wave‐dominated deltas, respectively. As larger deltas retain a greater proportion of their supply, this study suggests a positive feedback that enables larger deltas to grow faster than smaller ones.
- Open Access
- 29 April 2026
- Published by Elsevier in SSRN Electronic Journal
Higher order dynamic mode decomposition of tidal flows over ridge-swale bathymetry
Dimensionality reduction is a widely used tool in oceanography due to the volume of data and the ability to unravel physical patterns. Typically, this reduction relies on commonly used techniques, which have limitations for properly unravelling temporal dynamics. Here, we apply the dynamic mode decomposition to assess their capabilities to discern inner-shelf hydrodynamics at tidal scales. We find that applying the classical dynamic mode decomposition to the data subset yields an inaccurate reconstruction (error of 132%), as expected for a system with more temporal than spatial complexity. Furthermore, by applying the higher-order dynamic mode decomposition, we accurately decomposed the data for a temporal lag of ~1.5 hours and a spatial lag of 1 m (error 19.6%). The similarity in low errors for proportional temporal lags (32-34 and 63-65, error ~20%), in contrast to errors at 31, 35 (error ~85%), and 62, 66 (error ~35%), may indicate an optimal data extent, with enough dynamic content to reproduce spatiotemporal inner-shelf hydrodynamics. Apparently, the higher-order dynamic mode decomposition improves the fidelity of the hydrodynamic decomposition compared with the classical version and ”traditional” methods (e.g., empirical mode decomposition). However, we highlight a shortcoming in the fixed starting point of the temporal dynamics, which obligates the decomposition to include physically spurious modes.
- 29 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessHigher order dynamic mode decomposition of tidal flows over ridge-swale bathymetry
Dimensionality reduction is a widely used tool in oceanography due to the volume of data and the ability to unravel physical patterns. Typically, this reduction relies on commonly used techniques, which have limitations for properly unravelling temporal dynamics. Here, we apply the dynamic mode decomposition to assess their capabilities to discern inner-shelf hydrodynamics at tidal scales. We find that applying the classical dynamic mode decomposition to the data subset yields an inaccurate reconstruction (error of 132%), as expected for a system with more temporal than spatial complexity. Furthermore, by applying the higher-order dynamic mode decomposition, we accurately decomposed the data for a temporal lag of ~1.5 hours and a spatial lag of 1 m (error 19.6%). The similarity in low errors for proportional temporal lags (32-34 and 63-65, error ~20%), in contrast to errors at 31, 35 (error ~85%), and 62, 66 (error ~35%), may indicate an optimal data extent, with enough dynamic content to reproduce spatiotemporal inner-shelf hydrodynamics. Apparently, the higher-order dynamic mode decomposition improves the fidelity of the hydrodynamic decomposition compared with the classical version and ”traditional” methods (e.g., empirical mode decomposition). However, we highlight a shortcoming in the fixed starting point of the temporal dynamics, which obligates the decomposition to include physically spurious modes.
- Open Access
- 15 April 2026
- Published by Elsevier in SSRN Electronic Journal
Stratigraphic and Geotechnical Controls on Wave-Driven Erosion in Tropical Soft Cliffs
Cliff retreat is controlled by morphodynamic feedbacks that relate to the wave’s ability to erode, which decisively depends on rock strength and the degree of beach cover. In soft cliffs, these mechanisms are exacerbated by compressive strengths <1 MPa, which hinder rock response to wave hydraulic forces and enhance wave abrasion by providing coarse sediments after cliff retreat. Here, we explore the role of rock characteristics in wave-driven soft-cliff retreat at Arboletes town in the Northern Andes Caribbean Coast from stratigraphic analyses and mechanical laboratory testing. We performed stratigraphic descriptions of sedimentary sequences, including lithology, textures, fossils, structures, and bioturbation. We also quantified geotechnical index properties and conducted monotonic and cyclic triaxial tests on cliff-base rocks. Our results document the stratigraphy of tropical soft cliffs formed by (friable) brown siltstones overtopping gray mudstones (strength ~0.4 MPa from monotonic triaxial testing). We highlight the presence of sand-conglomerate lenses, which sediments become part of beach cover after cliff retreat and enhance wave-driven abrasion. Index geotechnical properties indicate marked differences in wetting-drying responses (~30% difference in liquid limits and plasticity between sites). Furthermore, cyclic triaxial testing at 0.10 Hz (~10 s) suggests that waves stiffen the cliff base mudstone, highlighting an overlooked effect of waves on soft cliffs. Our contrasting results on rocks typically identified as similar in descriptive studies highlight the need to conduct geological and geotechnical analyses to inform management practices along soft cliffs.
- 15 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessStratigraphic and Geotechnical Controls on Wave-Driven Erosion in Tropical Soft Cliffs
Cliff retreat is controlled by morphodynamic feedbacks that relate to the wave’s ability to erode, which decisively depends on rock strength and the degree of beach cover. In soft cliffs, these mechanisms are exacerbated by compressive strengths <1 MPa, which hinder rock response to wave hydraulic forces and enhance wave abrasion by providing coarse sediments after cliff retreat. Here, we explore the role of rock characteristics in wave-driven soft-cliff retreat at Arboletes town in the Northern Andes Caribbean Coast from stratigraphic analyses and mechanical laboratory testing. We performed stratigraphic descriptions of sedimentary sequences, including lithology, textures, fossils, structures, and bioturbation. We also quantified geotechnical index properties and conducted monotonic and cyclic triaxial tests on cliff-base rocks. Our results document the stratigraphy of tropical soft cliffs formed by (friable) brown siltstones overtopping gray mudstones (strength ~0.4 MPa from monotonic triaxial testing). We highlight the presence of sand-conglomerate lenses, which sediments become part of beach cover after cliff retreat and enhance wave-driven abrasion. Index geotechnical properties indicate marked differences in wetting-drying responses (~30% difference in liquid limits and plasticity between sites). Furthermore, cyclic triaxial testing at 0.10 Hz (~10 s) suggests that waves stiffen the cliff base mudstone, highlighting an overlooked effect of waves on soft cliffs. Our contrasting results on rocks typically identified as similar in descriptive studies highlight the need to conduct geological and geotechnical analyses to inform management practices along soft cliffs.
- Open Access
- 26 November 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Earth Surface
- Vol. 129 (11)
The Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties remain poorly known. Here, we address that gap and show how to quantify delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave‐driven sediment fluxes. Relative uncertainties are greatest for mixed‐process deltas (e.g., Sinu, error of 49%) and tend to decrease for end‐member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (a) delta morphology data, for example, delta slopes that modulate tidal fluxes, (b) data on river sediment flux distribution between individual delta river outlets, and (c) theoretical basis behind fluvial and tidal dominance. Broadly, these methods will help improve delta morphology predictions and assess how natural and anthropogenic forces affect morphologic change.
- 26 November 2024
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Earth Surface
- Vol. 129 (11)
Open AccessThe Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties remain poorly known. Here, we address that gap and show how to quantify delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave‐driven sediment fluxes. Relative uncertainties are greatest for mixed‐process deltas (e.g., Sinu, error of 49%) and tend to decrease for end‐member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (a) delta morphology data, for example, delta slopes that modulate tidal fluxes, (b) data on river sediment flux distribution between individual delta river outlets, and (c) theoretical basis behind fluvial and tidal dominance. Broadly, these methods will help improve delta morphology predictions and assess how natural and anthropogenic forces affect morphologic change.
- Open Access
- 8 July 2024
- Published by Wiley
The Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties have remained poorly known. We address that gap and expand on earlier work that quantified delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave-driven sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river outlets, and (3) theoretical basis behind fluvial and tidal dominance. Future works should address these sources, improving predictions of delta morphology from sediment fluxes that allow us to assess how much climate and global changes affect deltas.
- 8 July 2024
- Published by Wiley
Open AccessThe Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties have remained poorly known. We address that gap and expand on earlier work that quantified delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave-driven sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river outlets, and (3) theoretical basis behind fluvial and tidal dominance. Future works should address these sources, improving predictions of delta morphology from sediment fluxes that allow us to assess how much climate and global changes affect deltas.
- Open Access
- 17 June 2024
- Published by Wiley in ESS Open Archive
The Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties have remained poorly known. We address that gap and expand on earlier work that quantified delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave-driven sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river outlets, and (3) theoretical basis behind fluvial and tidal dominance. Future works should address these sources, improving predictions of delta morphology from sediment fluxes that allow us to assess how much climate and global changes affect deltas.
- 17 June 2024
- Published by Wiley in ESS Open Archive
Open AccessThe Quantified Galloway Ternary Diagram of Delta Morphology
Waves, rivers, and tides shape delta morphology. Recent studies have enabled predictions of their relative influence on deltas globally, but methods and associated uncertainties have remained poorly known. We address that gap and expand on earlier work that quantified delta morphology within the Galloway ternary diagram of river, wave, and tidal sediment fluxes. We assess delta morphology predictions compared to observations for 31 deltas globally and find a median error of 4% (standard deviation of 11%) in the river, tide, or wave-driven sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either wave, tide, or river sediment fluxes dominate (e.g., Fly, error of 0.2%). Prediction uncertainties for delta morphologic metrics are more considerable: the delta shoreline protrusion angles set by wave influence have a median error of 45%, the delta channel widening from tides 25%, and the number of distributary channels 86%. Larger sources of prediction uncertainty are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river outlets, and (3) theoretical basis behind fluvial and tidal dominance. Future works should address these sources, improving predictions of delta morphology from sediment fluxes that allow us to assess how much climate and global changes affect deltas.
- Open Access
- 16 February 2024
- Published by Cambridge University Press (CUP) in Quaternary Research
- Vol. 119, 28-43
Holocene relative sea-level changes along the Caribbean and Pacific coasts of northwestern South America
Predicting coastal change depends upon our knowledge of postglacial relative sea-level variability, partly controlled by glacio-isostatic responses to ice-sheet melting. Here, we reconstruct the postglacial relative sea-level changes along the Caribbean and Pacific coasts of northwestern South America by numerically solving the sea-level equation with two scenarios of mantle viscosity: global standard average and high viscosity. Our results with the standard model (applicable to the Pacific coast) agree with earlier studies by indicating a mid-Northgrippian high stand of ~2 m. The high-viscosity simulation (relevant to the Caribbean coast) shows that the transition from far- to intermediate-field influence of the Laurentide Ice Sheet occurs between Manzanillo del Mar and the Gulf of Morrosquillo. South of this location, the Colombian Caribbean coast has exhibited a still stand with a nearly constant Holocene relative sea level. By analyzing our simulations considering sea-level indicators, we argue that tectonics is more prominent than previously assumed, especially along the Caribbean coast. This influence prevents a simplified view of regional relative sea-level changes on the northwestern South American coast.
- 16 February 2024
- Published by Cambridge University Press (CUP) in Quaternary Research
- Vol. 119, 28-43
Open AccessHolocene relative sea-level changes along the Caribbean and Pacific coasts of northwestern South America
Predicting coastal change depends upon our knowledge of postglacial relative sea-level variability, partly controlled by glacio-isostatic responses to ice-sheet melting. Here, we reconstruct the postglacial relative sea-level changes along the Caribbean and Pacific coasts of northwestern South America by numerically solving the sea-level equation with two scenarios of mantle viscosity: global standard average and high viscosity. Our results with the standard model (applicable to the Pacific coast) agree with earlier studies by indicating a mid-Northgrippian high stand of ~2 m. The high-viscosity simulation (relevant to the Caribbean coast) shows that the transition from far- to intermediate-field influence of the Laurentide Ice Sheet occurs between Manzanillo del Mar and the Gulf of Morrosquillo. South of this location, the Colombian Caribbean coast has exhibited a still stand with a nearly constant Holocene relative sea level. By analyzing our simulations considering sea-level indicators, we argue that tectonics is more prominent than previously assumed, especially along the Caribbean coast. This influence prevents a simplified view of regional relative sea-level changes on the northwestern South American coast.
- Open Access
- 26 December 2022
- Published by Wiley
Global River Delta Morphology: Predictions Versus Observations within the Galloway Ternary Diagram
Waves, rivers, and tides play a leading role in shaping delta morphology. Recent studies have enabled predictions of their relative influence for deltas globally, but methods and associated uncertainties have remained poorly described. Here we aim to address that gap and assess the quality of delta morphology predictions compared to observations for 31 deltas globally. We expand on seminal works that quantified the Galloway ternary diagram from the balance between river, wave, and tidal sediment fluxes. Our data includes uncertainties for delta shoreline protrusion angles set by wave influence (14.1°±12° predicted vs. 20.8°±16.1° observed), channel widening, set by tidal influence (53.5±170.8 predicted vs. 6.5±11.5 observed), and number of distributary channels, set by river influence (55.9±127.5 predicted vs. 21.4±43.0 observed). Within the ternary diagram for delta morphology, we find an average error of 8% (±11%, 1 standard deviation), linked to uncertainties in wave and tide sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either one of wave, tide, or river sediment fluxes dominates (e.g., Fly, error of 0.2%). Large sources of prediction uncertainties are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river mouths, and (3) theoretical basis behind fluvial and tidal dominance. Future work could help address these three sources and improve predictions of delta morphology.
- 26 December 2022
- Published by Wiley
Open AccessGlobal River Delta Morphology: Predictions Versus Observations within the Galloway Ternary Diagram
Waves, rivers, and tides play a leading role in shaping delta morphology. Recent studies have enabled predictions of their relative influence for deltas globally, but methods and associated uncertainties have remained poorly described. Here we aim to address that gap and assess the quality of delta morphology predictions compared to observations for 31 deltas globally. We expand on seminal works that quantified the Galloway ternary diagram from the balance between river, wave, and tidal sediment fluxes. Our data includes uncertainties for delta shoreline protrusion angles set by wave influence (14.1°±12° predicted vs. 20.8°±16.1° observed), channel widening, set by tidal influence (53.5±170.8 predicted vs. 6.5±11.5 observed), and number of distributary channels, set by river influence (55.9±127.5 predicted vs. 21.4±43.0 observed). Within the ternary diagram for delta morphology, we find an average error of 8% (±11%, 1 standard deviation), linked to uncertainties in wave and tide sediment fluxes. Relative uncertainties are greatest for mixed-process deltas (e.g., Sinu, error of 49%) and tend to decrease for end-member morphologies where either one of wave, tide, or river sediment fluxes dominates (e.g., Fly, error of 0.2%). Large sources of prediction uncertainties are (1) delta morphology data, e.g., delta slopes that modulate tidal fluxes, (2) data on river sediment flux distribution between individual delta river mouths, and (3) theoretical basis behind fluvial and tidal dominance. Future work could help address these three sources and improve predictions of delta morphology.