Upload folder using huggingface_hub

README.md

@@ -0,0 +1,116 @@
+---
+license: cc-by-4.0
+task_categories:
+  - tabular-classification
+  - tabular-regression
+language:
+  - en
+tags:
+  - climate-health
+  - sea-level-rise
+  - coastal-health
+  - flooding
+  - health-infrastructure
+  - synthetic
+  - sub-saharan-africa
+pretty_name: Sea-Level Rise & Coastal Health Infrastructure (SSA)
+size_categories:
+  - 10K<n<100K
+configs:
+  - config_name: west_africa_delta
+    data_files: data/slr_west_africa_delta.csv
+    default: true
+  - config_name: east_africa_island
+    data_files: data/slr_east_africa_island.csv
+  - config_name: southern_coastal_city
+    data_files: data/slr_southern_coastal_city.csv
+---
+
+# Sea-Level Rise & Coastal Health Infrastructure in Sub-Saharan Africa
+
+## Abstract
+
+A synthetic dataset modelling sea-level rise impacts on coastal health infrastructure and disease outcomes across three coastal zone scenarios in SSA. Each record represents a location-year observation describing flood risk, facility damage, supply chain disruption, and disease burden. IPCC AR6 projects 0.3-1.0m global mean SLR by 2100, with African coasts among the most vulnerable due to low-lying deltas, limited adaptation, and high population density.
+
+### Scenarios
+
+- **West Africa Delta**: Low-lying Niger/Volta delta zones with high flood and erosion risk.
+- **East Africa Island**: Indian Ocean island and coastal communities with storm surge exposure.
+- **Southern Coastal City**: Urbanized coastal cities with higher baseline infrastructure but growing risk.
+
+## Dataset Structure
+
+Each scenario contains 10,000 records (30,000 total). Key columns:
+
+- `elevation_m`, `slr_rate_mm_per_yr`, `cumulative_slr_cm`
+- `storm_surge`, `surge_height_m`, `coastal_erosion_m_per_yr`
+- `flood_risk`, `flooded`, `saltwater_intrusion`
+- `facility_at_risk`, `facility_damaged`, `facility_inaccessible`
+- `supply_chain_disrupted`, `essential_medicine_stockout`
+- `displaced`, `population_exposed`
+- `waterborne_disease`, `malaria_case`, `cholera_outbreak`, `mental_health_impact`
+- `health_system_resilience`, `adaptation_score`
+
+## Parameterization Evidence
+
+| Parameter | Value Used | Source | Year |
+| --- | --- | --- | --- |
+| Global mean SLR 3-4 mm/yr | SLR rate baseline | IPCC AR6 WG1 | 2021 |
+| West African coastline eroding 1-5 m/yr | Erosion rates | Appeaning Addo et al. | 2011 |
+| 30% of African health facilities in climate-vulnerable zones | Facility risk | WHO-UNFCCC | 2021 |
+| Flooding amplifies waterborne disease and malaria | Disease linkage | Lancet Countdown Africa | 2022 |
+
+## Validation Summary
+
+The 8-panel validation report (`validation_report.png`) confirms:
+
+1. **Flood risk**: West Africa Delta has highest flood risk; elevation inversely correlated.
+2. **Health impacts**: Waterborne disease and malaria highest in delta scenario.
+3. **Facility risk**: Facility damage and inaccessibility track flood exposure.
+4. **Elevation gradient**: Lower elevation strongly predicts higher flood risk.
+5. **Supply chain**: Disruption rates highest in delta and island scenarios.
+6. **Resilience**: Southern coastal cities show highest health system resilience.
+7. **SLR trend**: Cumulative SLR increases over time across all scenarios.
+8. **Adaptation**: Urban settings show modestly higher adaptation scores.
+
+![Validation Report](validation_report.png)
+
+## Usage
+
+```python
+from datasets import load_dataset
+
+ds = load_dataset("electricsheepafrica/sea-level-rise-coastal-health", name="west_africa_delta")
+df = ds["train"].to_pandas()
+
+print(df.groupby("flooded")[["waterborne_disease", "malaria_case"]].mean())
+```
+
+## Limitations
+
+- **Synthetic data**: Not from tide gauges, satellite altimetry, or health facility surveys.
+- **No spatial geocoding**: Scenarios proxy geography, not precise coordinates.
+- **Simplified SLR model**: Linear trend; actual SLR is nonlinear with regional variation.
+
+## References
+
+1. IPCC. AR6 Working Group I: The Physical Science Basis. 2021.
+2. Appeaning Addo K, et al. Shoreline change in Ghana's coastal zone. *Afr Geogr Rev*, 2011.
+3. WHO-UNFCCC. Health and Climate Change Country Profiles. 2021.
+4. Lancet Countdown on Health and Climate Change: Africa Report. 2022.
+
+## Citation
+
+```bibtex
+@dataset{electricsheepafrica_sea_level_rise_coastal_health_2025,
+  title={Sea-Level Rise and Coastal Health Infrastructure in Sub-Saharan Africa},
+  author={Electric Sheep Africa},
+  year={2025},
+  publisher={HuggingFace},
+  url={https://huggingface.co/datasets/electricsheepafrica/sea-level-rise-coastal-health}
+}
+```
+
+## License
+
+CC-BY-4.0

generate_dataset.py

@@ -0,0 +1,197 @@
+"""Generate synthetic sea-level rise & coastal health infrastructure dataset for SSA."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import numpy as np
+import pandas as pd
+
+SEED = 42
+N_PER_SCENARIO = 10_000
+
+YEAR_RANGE = np.arange(2010, 2025)
+YEAR_WEIGHTS = np.linspace(0.85, 1.3, len(YEAR_RANGE))
+YEAR_WEIGHTS = YEAR_WEIGHTS / YEAR_WEIGHTS.sum()
+
+SCENARIOS = {
+    "west_africa_delta": {
+        "slr_mm_per_yr_mean": 3.5,
+        "slr_mm_per_yr_sd": 0.8,
+        "storm_surge_freq": 0.25,
+        "coastal_erosion_m_per_yr": 2.5,
+        "elevation_mean": 3.0,
+        "elevation_sd": 2.0,
+        "flood_risk_base": 0.35,
+        "saltwater_intrusion_pct": 0.40,
+        "facility_at_risk_pct": 0.30,
+        "facility_damaged_pct": 0.12,
+        "population_density_mean": 450,
+        "displacement_pct": 0.20,
+        "waterborne_disease_rate": 120,
+        "malaria_rate_per1k": 180,
+        "cholera_risk": 0.15,
+        "mental_health_impact": 0.25,
+        "setting_probs": {"coastal_urban": 0.35, "coastal_rural": 0.40, "delta": 0.25},
+    },
+    "east_africa_island": {
+        "slr_mm_per_yr_mean": 4.0,
+        "slr_mm_per_yr_sd": 1.0,
+        "storm_surge_freq": 0.30,
+        "coastal_erosion_m_per_yr": 1.8,
+        "elevation_mean": 4.5,
+        "elevation_sd": 2.5,
+        "flood_risk_base": 0.28,
+        "saltwater_intrusion_pct": 0.35,
+        "facility_at_risk_pct": 0.25,
+        "facility_damaged_pct": 0.10,
+        "population_density_mean": 300,
+        "displacement_pct": 0.15,
+        "waterborne_disease_rate": 95,
+        "malaria_rate_per1k": 140,
+        "cholera_risk": 0.10,
+        "mental_health_impact": 0.22,
+        "setting_probs": {"coastal_urban": 0.30, "coastal_rural": 0.35, "island": 0.35},
+    },
+    "southern_coastal_city": {
+        "slr_mm_per_yr_mean": 3.0,
+        "slr_mm_per_yr_sd": 0.6,
+        "storm_surge_freq": 0.20,
+        "coastal_erosion_m_per_yr": 1.5,
+        "elevation_mean": 6.0,
+        "elevation_sd": 3.0,
+        "flood_risk_base": 0.22,
+        "saltwater_intrusion_pct": 0.25,
+        "facility_at_risk_pct": 0.18,
+        "facility_damaged_pct": 0.07,
+        "population_density_mean": 600,
+        "displacement_pct": 0.10,
+        "waterborne_disease_rate": 65,
+        "malaria_rate_per1k": 80,
+        "cholera_risk": 0.06,
+        "mental_health_impact": 0.18,
+        "setting_probs": {"coastal_urban": 0.55, "coastal_rural": 0.25, "peri_urban": 0.20},
+    },
+}
+
+SCENARIO_FILES = {
+    "west_africa_delta": "slr_west_africa_delta.csv",
+    "east_africa_island": "slr_east_africa_island.csv",
+    "southern_coastal_city": "slr_southern_coastal_city.csv",
+}
+
+
+def _choice(rng, prob_map):
+    keys = list(prob_map.keys())
+    weights = np.array(list(prob_map.values()), dtype=float)
+    weights = weights / weights.sum()
+    return rng.choice(keys, p=weights)
+
+
+def _simulate_scenario(name, params, seed):
+    rng = np.random.default_rng(seed)
+    records = []
+
+    for idx in range(N_PER_SCENARIO):
+        year = int(rng.choice(YEAR_RANGE, p=YEAR_WEIGHTS))
+        setting = _choice(rng, params["setting_probs"])
+
+        elevation_m = float(np.clip(rng.normal(params["elevation_mean"], params["elevation_sd"]), 0.2, 20))
+        slr_rate = float(np.clip(rng.normal(params["slr_mm_per_yr_mean"], params["slr_mm_per_yr_sd"]), 1, 8))
+        cumulative_slr_cm = float(slr_rate * (year - 2000) / 10)
+
+        storm_surge = int(rng.random() < params["storm_surge_freq"])
+        surge_height_m = float(np.clip(rng.exponential(0.8), 0, 3.5)) if storm_surge else 0.0
+        coastal_erosion = float(np.clip(
+            rng.normal(params["coastal_erosion_m_per_yr"], 0.8), 0, 8
+        ))
+
+        flood_risk = float(np.clip(
+            params["flood_risk_base"] + (cumulative_slr_cm / 15) * 0.2 - (elevation_m / 10) * 0.15,
+            0, 1,
+        ))
+        flooded = int(rng.random() < flood_risk)
+        saltwater_intrusion = int(rng.random() < params["saltwater_intrusion_pct"] + cumulative_slr_cm * 0.01)
+
+        facility_at_risk = int(rng.random() < params["facility_at_risk_pct"])
+        facility_damaged = int(facility_at_risk and rng.random() < params["facility_damaged_pct"] / params["facility_at_risk_pct"])
+        facility_inaccessible = int((flooded or facility_damaged) and rng.random() < 0.4)
+
+        supply_chain_disrupted = int((flooded or storm_surge) and rng.random() < 0.35)
+        essential_medicine_stockout = int(supply_chain_disrupted and rng.random() < 0.45)
+
+        displaced = int(rng.random() < params["displacement_pct"] * (1 + flooded * 0.5))
+        population_exposed = int(np.clip(
+            rng.normal(params["population_density_mean"], 120) * (1 if elevation_m < 5 else 0.5),
+            50, 2000,
+        ))
+
+        wb_disease_rate = params["waterborne_disease_rate"] * (1 + flooded * 0.6 + saltwater_intrusion * 0.3)
+        waterborne_disease = int(rng.random() < wb_disease_rate / 1000)
+
+        malaria_rate = params["malaria_rate_per1k"] * (1 + flooded * 0.4)
+        malaria_case = int(rng.random() < malaria_rate / 1000)
+
+        cholera_outbreak = int(rng.random() < params["cholera_risk"] * (1 + flooded * 1.5 + saltwater_intrusion * 0.5))
+
+        mental_health_impact = int(rng.random() < params["mental_health_impact"] * (1 + displaced * 0.5))
+
+        health_system_resilience = float(np.clip(
+            0.5
+            - facility_damaged * 0.15
+            - supply_chain_disrupted * 0.1
+            - facility_inaccessible * 0.1
+            + (0.1 if "urban" in setting else 0)
+            + rng.normal(0, 0.05),
+            0, 1,
+        ))
+
+        adaptation_score = float(np.clip(
+            rng.normal(0.3, 0.12) + (0.1 if "urban" in setting else 0),
+            0, 1,
+        ))
+
+        record = {
+            "record_id": f"{name[:3].upper()}-{idx:05d}",
+            "scenario": name,
+            "year": year,
+            "setting": setting,
+            "elevation_m": round(elevation_m, 1),
+            "slr_rate_mm_per_yr": round(slr_rate, 1),
+            "cumulative_slr_cm": round(cumulative_slr_cm, 1),
+            "storm_surge": storm_surge,
+            "surge_height_m": round(surge_height_m, 1),
+            "coastal_erosion_m_per_yr": round(coastal_erosion, 1),
+            "flood_risk": round(flood_risk, 2),
+            "flooded": flooded,
+            "saltwater_intrusion": saltwater_intrusion,
+            "facility_at_risk": facility_at_risk,
+            "facility_damaged": facility_damaged,
+            "facility_inaccessible": facility_inaccessible,
+            "supply_chain_disrupted": supply_chain_disrupted,
+            "essential_medicine_stockout": essential_medicine_stockout,
+            "displaced": displaced,
+            "population_exposed": population_exposed,
+            "waterborne_disease": waterborne_disease,
+            "malaria_case": malaria_case,
+            "cholera_outbreak": cholera_outbreak,
+            "mental_health_impact": mental_health_impact,
+            "health_system_resilience": round(health_system_resilience, 2),
+            "adaptation_score": round(adaptation_score, 2),
+        }
+        records.append(record)
+
+    return pd.DataFrame(records)
+
+
+def main():
+    output_dir = Path("data")
+    output_dir.mkdir(parents=True, exist_ok=True)
+    for idx, (name, params) in enumerate(SCENARIOS.items()):
+        df = _simulate_scenario(name, params, SEED + idx * 211)
+        df.to_csv(output_dir / SCENARIO_FILES[name], index=False)
+        print(f"Saved {name} -> {SCENARIO_FILES[name]}")
+
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,3 @@
+numpy>=1.24
+pandas>=2.0
+matplotlib>=3.7

validate_dataset.py

@@ -0,0 +1,117 @@
+"""Validate synthetic sea-level rise & coastal health infrastructure dataset."""
+
+from __future__ import annotations
+
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+import pandas as pd
+
+SCENARIO_FILES = {
+    "west_africa_delta": "slr_west_africa_delta.csv",
+    "east_africa_island": "slr_east_africa_island.csv",
+    "southern_coastal_city": "slr_southern_coastal_city.csv",
+}
+
+COLORS = {
+    "west_africa_delta": "#e6550d",
+    "east_africa_island": "#3182bd",
+    "southern_coastal_city": "#31a354",
+}
+
+
+def load_data() -> pd.DataFrame:
+    frames = []
+    for scenario, filename in SCENARIO_FILES.items():
+        df = pd.read_csv(Path("data") / filename)
+        frames.append(df)
+    return pd.concat(frames, ignore_index=True)
+
+
+def plot_validation(df: pd.DataFrame, output_path: Path) -> None:
+    fig, axes = plt.subplots(4, 2, figsize=(14, 16))
+    axes = axes.flatten()
+
+    # Panel 1: Flood risk by scenario
+    for s in SCENARIO_FILES:
+        subset = df[df["scenario"] == s]
+        axes[0].hist(subset["flood_risk"], bins=25, alpha=0.5, color=COLORS[s], label=s)
+    axes[0].set_title("Flood Risk Distribution by Scenario")
+    axes[0].set_xlabel("Flood risk")
+    axes[0].legend(fontsize=7)
+
+    # Panel 2: Health impacts
+    health_cols = ["waterborne_disease", "malaria_case", "cholera_outbreak", "mental_health_impact"]
+    rates = df.groupby("scenario")[health_cols].mean() * 100
+    rates.plot(kind="bar", ax=axes[1])
+    axes[1].set_title("Health Impact Rates (%)")
+    axes[1].set_ylabel("Percent")
+    axes[1].legend(fontsize=7)
+
+    # Panel 3: Facility damage & inaccessibility
+    fac_cols = ["facility_at_risk", "facility_damaged", "facility_inaccessible"]
+    fac_rates = df.groupby("scenario")[fac_cols].mean() * 100
+    fac_rates.plot(kind="bar", ax=axes[2])
+    axes[2].set_title("Health Facility Risk (%)")
+    axes[2].set_ylabel("Percent")
+    axes[2].legend(fontsize=7)
+
+    # Panel 4: Elevation vs flood risk
+    for s in SCENARIO_FILES:
+        subset = df[df["scenario"] == s]
+        axes[3].scatter(subset["elevation_m"], subset["flood_risk"], s=6, alpha=0.15, color=COLORS[s], label=s)
+    axes[3].set_title("Elevation vs Flood Risk")
+    axes[3].set_xlabel("Elevation (m)")
+    axes[3].set_ylabel("Flood risk")
+    axes[3].legend(fontsize=7)
+
+    # Panel 5: Supply chain disruption
+    sc_cols = ["supply_chain_disrupted", "essential_medicine_stockout"]
+    sc_rates = df.groupby("scenario")[sc_cols].mean() * 100
+    sc_rates.plot(kind="bar", ax=axes[4])
+    axes[4].set_title("Supply Chain Disruption (%)")
+    axes[4].set_ylabel("Percent")
+    axes[4].legend(fontsize=7)
+
+    # Panel 6: Health system resilience
+    res_data = [df[df["scenario"] == s]["health_system_resilience"] for s in SCENARIO_FILES]
+    bp = axes[5].boxplot(res_data, tick_labels=SCENARIO_FILES.keys(), patch_artist=True)
+    for patch, s in zip(bp["boxes"], SCENARIO_FILES.keys()):
+        patch.set_facecolor(COLORS[s])
+        patch.set_alpha(0.6)
+    axes[5].set_title("Health System Resilience Score")
+    axes[5].set_ylabel("Score")
+
+    # Panel 7: Cumulative SLR trend
+    yearly = df.groupby(["scenario", "year"])["cumulative_slr_cm"].mean().reset_index()
+    for s in SCENARIO_FILES:
+        sub = yearly[yearly["scenario"] == s]
+        axes[6].plot(sub["year"], sub["cumulative_slr_cm"], marker="o", color=COLORS[s], label=s)
+    axes[6].set_title("Cumulative Sea-Level Rise Trend")
+    axes[6].set_xlabel("Year")
+    axes[6].set_ylabel("Cumulative SLR (cm)")
+    axes[6].legend(fontsize=7)
+
+    # Panel 8: Adaptation score
+    adapt_data = [df[df["scenario"] == s]["adaptation_score"] for s in SCENARIO_FILES]
+    bp2 = axes[7].boxplot(adapt_data, tick_labels=SCENARIO_FILES.keys(), patch_artist=True)
+    for patch, s in zip(bp2["boxes"], SCENARIO_FILES.keys()):
+        patch.set_facecolor(COLORS[s])
+        patch.set_alpha(0.6)
+    axes[7].set_title("Adaptation Score by Scenario")
+    axes[7].set_ylabel("Score")
+
+    plt.tight_layout()
+    fig.savefig(output_path, dpi=200)
+    plt.close(fig)
+
+
+def main() -> None:
+    df = load_data()
+    output_path = Path("validation_report.png")
+    plot_validation(df, output_path)
+    print(f"Saved {output_path}")
+
+
+if __name__ == "__main__":
+    main()