Spatiotemporal shoreline and sea level dynamics along the Kenyan coast (2000–2024): patterns of erosion, accretion, and spatial autocorrelation

Shoreline management in the Western Indian Ocean (WIO) is largely informed by small-scale studies with coarse temporal resolution, potentially obscuring fine spatial patterns and interannual dynamics. This study examines spatiotemporal shoreline change along 400 km of the Kenyan coastline from 2000 to 2024 and tests its association with regional sea-level trend and variability, accounting for spatial dependence. Twenty-five annual Landsat-derived shoreline positions were analyzed using the Digital Shoreline Analysis System (DSAS) along 12,455 transects using Weighted Linear Regression (WLR) weighted by a tide- and registration-based positional uncertainty of approximately 40 to 68 m per epoch. Sea-level predictors were derived from GLORYS12V1 reanalysis at its native grid (47 coastal cells) and tested with Ordinary Least Squares (OLS), a Spatial Lag Model (SLM), and Geographically Weighted Regression (GWR). Results reveal pronounced spatial heterogeneity. The Kilifi and Kwale–Mombasa zones show net accretion (mean WLR 0.57 and 0.51 m yr -1 , respectively), whereas the Tana–Manda zone shows net erosion (mean WLR = −0.92 m yr −1 ), with 36.23% of its transects significantly eroding. The national mean Net Shoreline Movement of +9.45 m obscures this variability. Shoreline change is strongly spatially autocorrelated (Moran’s I = 0.866, p = 0.001) and 12.01% of transects form statistically significant local clusters (6.23% as High–High accretion and 5.57% Low–Low erosion clusters). Neither sea-level trend nor sea-level variability showed a detectable association with shoreline change at the native support of the predictors (OLS R 2 ≤ 0.028, p ≥ 0.26; SLM not significant; GWR adjusted R 2 < 0). At the scales resolved here, shoreline change along the Kenyan coast is therefore not explained by regional sea-level variation and is more plausibly linked to local geomorphic and human factors. National averages must be complemented by spatially explicit, uncertainty-weighted assessments to inform shoreline management policy.

Authors

Institutions

Publication Details

Journal
Frontiers in Marine Science
Published
2026-11-19
DOI
https://doi.org/10.3389/fmars.2026.1952449
Primary Topic
Coastal and Marine Dynamics
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Spatiotemporal shoreline and sea level dynamics along the Kenyan coast (2000–2024): patterns of erosion, accretion, and spatial autocorrelation

James Kamau Mbugua, David O. Obura, Harison Kiplagat Kipkulei, Kennedy Edeye Osuka et al.
Frontiers in Marine Science
Coastal and Marine Dynamics
article

Spatiotemporal shoreline and sea level dynamics along the Kenyan coast (2000–2024): patterns of erosion, accretion, and spatial autocorrelation

James Kamau Mbugua, David O. Obura, Harison Kiplagat Kipkulei, Kennedy Edeye Osuka, George Watene, Obed Ogega
article en

Abstract

Shoreline management in the Western Indian Ocean (WIO) is largely informed by small-scale studies with coarse temporal resolution, potentially obscuring fine spatial patterns and interannual dynamics. This study examines spatiotemporal shoreline change along 400 km of the Kenyan coastline from 2000 to 2024 and tests its association with regional sea-level trend and variability, accounting for spatial dependence. Twenty-five annual Landsat-derived shoreline positions were analyzed using the Digital Shoreline Analysis System (DSAS) along 12,455 transects using Weighted Linear Regression (WLR) weighted by a tide- and registration-based positional uncertainty of approximately 40 to 68 m per epoch. Sea-level predictors were derived from GLORYS12V1 reanalysis at its native grid (47 coastal cells) and tested with Ordinary Least Squares (OLS), a Spatial Lag Model (SLM), and Geographically Weighted Regression (GWR). Results reveal pronounced spatial heterogeneity. The Kilifi and Kwale–Mombasa zones show net accretion (mean WLR 0.57 and 0.51 m yr -1 , respectively), whereas the Tana–Manda zone shows net erosion (mean WLR = −0.92 m yr −1 ), with 36.23% of its transects significantly eroding. The national mean Net Shoreline Movement of +9.45 m obscures this variability. Shoreline change is strongly spatially autocorrelated (Moran’s I = 0.866, p = 0.001) and 12.01% of transects form statistically significant local clusters (6.23% as High–High accretion and 5.57% Low–Low erosion clusters). Neither sea-level trend nor sea-level variability showed a detectable association with shoreline change at the native support of the predictors (OLS R 2 ≤ 0.028, p ≥ 0.26; SLM not significant; GWR adjusted R 2 < 0). At the scales resolved here, shoreline change along the Kenyan coast is therefore not explained by regional sea-level variation and is more plausibly linked to local geomorphic and human factors. National averages must be complemented by spatially explicit, uncertainty-weighted assessments to inform shoreline management policy.

Frontiers in Marine ScienceVol. 13
University of Nairobi (KE), University of Liverpool (GB), University of Augsburg (DE), International Institute for Environment and Development (GB), Jomo Kenyatta University of Agriculture and Technology (KE), 10:10 Climate Action (GB), Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (DE), Coastal Oceans Research and Development in the Indian Ocean (KE)
Coastal and Marine Dynamics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon