Replication Package for Land Subsidence Susceptibility Mapping and Screening-Level Relative Sea Level Change Scenarios for the Nigerian Coastal Zone Using GIS-Based Multi-Criteria Decision Analysis

John Okwudili Ugwu, Njoku RE, Udo Emmanuel Ahuchaogu, Munachimso Samson Uzoeshi

# Land Subsidence Susceptibility Mapping and Screening-Level Relative Sea Level Change Scenarios for the Nigerian Coastal Zone Analysis code and derived data layers for the manuscript: > Ugwu, O. J., Njoku, R. E., Ahuchaogu, E. U., \& Uzoeshi, S. M. (2026).> \*Land Subsidence Susceptibility Mapping and Screening-Level Relative Sea> Level Change Scenarios for the Nigerian Coastal Zone Using GIS-Based> Multi-Criteria Decision Analysis.\* Remote Sensing Applications: Society> and Environment. Manuscript RSASE-D-26-00979. Corresponding author: Okwudili John Ugwu, Department of Surveying andGeoinformatics, Federal University of Technology Owerri (FUTO), Nigeria.okwudili.ugwu@futo.edu.ng This repository lets a reader reproduce every number, table and figure inthe manuscript from freely available global input datasets. \--- ## What this study is, and is not This is a **screening-level** assessment. It derives a spatially continuousSubsidence Susceptibility Index (SSI) by GIS-based Multi-Criteria DecisionAnalysis, and uses it as a proxy to generate **screening-level relative sealevel change (RSLC) scenarios**. ## Analysis domain The analysis domain covers **six of the eight** Nigerian littoral states —Lagos, Ogun, Ondo, Delta, Bayelsa and Rivers — totalling **65,840 km²** over**5,378,411** valid grid cells at 0.001° (\~110 m). Akwa Ibom and Cross Riverlie east of the iSDAsoil clay layer (F1) extent (\~7.6°E), which is the leastspatially extensive of the six input factors, and are therefore **outside thedomain and not assessed**. \--- ## Repository contents ### Scripts (`04\_Scripts/`) Run in order. Each prints its outputs and where it wrote them. |Script|Produces||-|-||`01\_prepare\_factors.py`|Factor layers F3, F5, F6 base preparation, clipped and normalised (does **not** write F2)||`01b\_fix\_f4\_elevation.py`|F4 inverse-elevation layer from FABDEM||`01c\_fix\_f6\_lithology.py`|F6 lithology score from GLiM||`01f\_revised\_f1\_clay\_content.py`|F1 clay-content layer from iSDAsoil||`01g\_f2\_ogim.py`|F2 oil/gas well proximity from OGIM v2.7 — the **sole** producer of `F2\_oil\_wells\_distance\_norm.tif`||`02\_ahp\_weights\_revised.py`|AHP weights, consistency ratio; writes `ahp\_weights\_revised.json`||`03\_compute\_ssi\_revised.py`|`SSI\_continuous\_revised.tif`, `SSI\_classified\_revised.tif`||`04\_compute\_rslc.py`|`RSLC\_rate\_map.tif` and cumulative projection rasters||`05\_consistency\_checks.py`|Consistency and sensitivity checks C1–C7; writes manuscript Fig. 10||`06\_exposure.py`|Population and urbanised-area exposure; state table||`07\_factor\_correlation.py`|Factor correlation matrix (Supplementary Table S2)||`08\_generate\_maps.py`|Study-area and SSI/RSLC cartographic maps (manuscript Figs 1, 5, 6, 7)||`09\_factor\_and\_exposure\_maps.py`|Factor-layer and exposure figures (manuscript Figs 3, 9)||`10\_workflow\_diagram.py`|Methodology workflow diagram (manuscript Fig. 2)| ### ### Figure files → manuscript figure numbers The script output filenames do **not** all match the manuscript figurenumbers. Use this mapping: |Manuscript|File|Written by||-|-|-||Fig. 1|`Fig01\_study\_area.png`|`08\_generate\_maps.py`||Fig. 2|`Fig02\_methodology\_workflow.png` / `.pdf`|`10\_workflow\_diagram.py`||Fig. 3|`Fig03\_factor\_layers.png`|`09\_factor\_and\_exposure\_maps.py`||Fig. 4|`Fig04\_ahp\_weights.png` / `.pdf`|`02\_ahp\_weights\_revised.py`||Fig. 5|`Fig06\_SSI\_continuous.png`|`08\_generate\_maps.py`||Fig. 6|`Fig04\_SSI\_classified.png`|`08\_generate\_maps.py`||Fig. 7|`Fig05\_RSLC\_rate.png` **or** `Fig07\_rslc\_rate\_map.png`|`08` / `04` (see note)||Fig. 8|`Fig08\_rslc\_projections\_panel.png`|`04\_compute\_rslc.py`||Fig. 9|`Fig09\_exposure.png`|`09\_factor\_and\_exposure\_maps.py`||Fig. 10|`Fig10\_consistency\_sensitivity\_checks.png`|`05\_consistency\_checks.py`| ### Files included in this deposit This repository contains the **analysis code and documentation only**. Theinput rasters and the derived outputs are **not** included — the inputs arethird-party datasets (re-fetch them using `data\_sources.csv`), and the outputsare regenerated by running the scripts. See "Outputs the scripts generate"below. |File|Description||-|-||`01\_prepare\_factors.py` … `10\_workflow\_diagram.py`|The 14 analysis scripts (see the Scripts table above)||`AHP\_matrix.csv`|The full 6×6 AHP pairwise comparison matrix (input to `02`)||`Table\_S2\_factor\_correlations.csv`, `Table\_S2\_factor\_correlations\_spearman.csv`|The factor correlation matrices (Supplementary Table S2), produced by `07` and included here as a convenience||`data\_sources.csv`|Every input dataset: version, resolution, epoch, access date, download link||`environment.yml`|conda environment specification (`nigeria\_insar' or 'nigeria\_flood`)||`README.md`|This file||`LICENSE`|Licence (see below)| ### Outputs the scripts generate (not deposited — regenerated on run) Running the pipeline writes the following into a local `03\_Results/`directory. They are **not** part of this deposit; they are listed here so youknow what a successful run produces. |Output|Written by||-|-||`F1\_clay\_content\_norm.tif` … `F6\_lithology\_score\_norm.tif` (six normalised factor layers)|`01\*`||`SSI\_continuous\_revised.tif`, `SSI\_classified\_revised.tif`|`03`||`RSLC\_rate\_map.tif` and cumulative projection rasters|`04`||`ahp\_weights\_revised.json`|`02`||`consistency\_checks\_results.csv`, `broadened\_sensitivity.csv`, `scale\_factor\_sensitivity.csv`, `rslc\_class\_by\_scale\_factor.csv`, `morans\_i\_results.csv`, `city\_point\_extractions.csv`|`05`||`population\_exposure.csv`, `state\_exposure.csv`, `ssi\_class\_areas.csv`|`06`||`Fig01`–`Fig10` (`.png` / `.pdf`) — all manuscript figures (see mapping above)|`02`, `04`, `05`, `08`, `09`, `10`| \--- ## Input datasets (not redistributed here) The six factors and the population and exposure layers are derived fromthird-party global datasets, which are openly available from their originalproviders. They are **not** redistributed in this repository. Versions,resolutions, acquisition periods, access dates and download links are in`data\_sources.csv`. Summary: |Factor / use|Dataset|Version|Native res.|Period||-|-|-|-|-||F1 clay content|iSDAsoil 0–20 cm clay|v0.13|30 m|2001–2017||F2 oil/gas well proximity|OGIM|v2.7|point|(see note)||F3 groundwater trend|GSFC GRACE/GRACE-FO Mascon|RL06 v2.0|0.5° / 1° mascon|2002-04 – 2025-05||F4 inverse elevation|FABDEM|v1.2|30 m|2020||F5 river proximity|derived from FABDEM (elev < 2 m)|v1.2|30 m|2020||F6 lithology|GLiM|v1.0|0.5°|—||Population / exposure|WorldPop Nigeria, constrained|2020|100 m|2020| \--- ## How to reproduce ```conda env create -f environment.ymlconda activate nigeria\_insar or nigeria\_flood# Download the input datasets listed in data\_sources.csv into# 01\_Factor\_Layers/ and 02\_Study\_Area/ as noted there.python 04\_Scripts/01\_prepare\_factors.pypython 04\_Scripts/01b\_fix\_f4\_elevation.pypython 04\_Scripts/01c\_fix\_f6\_lithology.pypython 04\_Scripts/01f\_revised\_f1\_clay\_content.pypython 04\_Scripts/01g\_f2\_ogim.pypython 04\_Scripts/02\_ahp\_weights\_revised.pypython 04\_Scripts/03\_compute\_ssi\_revised.pypython 04\_Scripts/04\_compute\_rslc.pypython 04\_Scripts/05\_consistency\_checks.pypython 04\_Scripts/06\_exposure.pypython 04\_Scripts/07\_factor\_correlation.pypython 04\_Scripts/08\_generate\_maps.pypython 04\_Scripts/09\_factor\_and\_exposure\_maps.pypython 04\_Scripts/10\_workflow\_diagram.py``` Scripts use an absolute project root (`PROJECT\_ROOT` near the top of eachfile). Set it to wherever you place the repository. ### Reproduction check `05\_consistency\_checks.py` recomputes the SSI from the six factor layers andthe AHP weights and compares it against the `SSI\_continuous\_revised.tif`produced earlier in the same pipeline (by `03`). When the pipeline is run endto end, the maximum absolute difference between the two is \~1.5e-07(single-precision rounding), confirming that the susceptibility surface isreproduced independently within the workflow. Note that this check comparesthe code against its own regenerated output; it requires running the pipelinefrom the input datasets (see `data\_sources.csv`), since neither the inputs northe derived rasters are included in this deposit. ### Key results these scripts reproduce * AHP: Consistency Ratio 0.0125; weights F2 0.397, F4 0.250, F3 0.160, F1 0.097, F5 0.060, F6 0.037* SSI: 0.080 to 0.930 (mean 0.550, SD 0.252) over 5,378,411 cells* RSLC scenario: 4.30 to 17.04 mm/yr (a = 15, no intercept)* Domain area: 65,840 km²; SSI classes: Very High 15,695.6 km² (23.84%) … Very Low 5,808.3 km² (8.82%)* Population: 41,344,645 total; 24,607,733 (59.5%) at RSLC ≥ 10 mm/yr; 8,444,872 (20.4%) at ≥ 15* Urbanised-area proxy (≥ 300 persons/km²): 6,274 km² total; 2,931 km² (46.7%) at RSLC ≥ 10 \--- ## Note on F2 (oil/gas well proximity) F2 is derived from the OGIM v2.7 well database and is produced by`01g\_f2\_ogim.py`, which is the **only** script that writes`F2\_oil\_wells\_distance\_norm.tif`. `01\_prepare\_factors.py` does not write F2.This removes an earlier ambiguity in which two scripts wrote the same filenamefrom different sources. ## Declaration of generative AI in the research process The analysis scripts in this repository were drafted with the assistance of alarge language model and were subsequently reviewed,corrected, and executed by the authors. All numerical results reported in theassociated manuscript were independently recomputed from the source datasets bythe corresponding author and verified against the input rasters. The authorstake full responsibility for the code, the data-processing decisions, and everyresult reported. ## Licence Code in this repository: MIT (see `LICENSE`). The derived data layers, whichare not included in this deposit but are produced by running the code, may bereleased under CC-BY 4.0 if deposited separately.Input datasets retain their original providers