Investigating Drought–Seismicity Linkages Across Botswana and surrounding areas: Evidence from SPEI and Earthquake Catalogues

Description

This dataset integrates climate and seismic records to investigate potential linkages between drought variability and earthquake occurrence across Botswana and its surrounding areas, with particular emphasis on the Okavango Rift Zone. The central hypothesis is that prolonged drought conditions, as captured by the Standardized Precipitation Evapotranspiration Index (SPEI), may influence seismic activity through hydrological loading and stress redistribution in the crust. SPEI values were derived from reanalysis and observational climate datasets, calculated at multiple timescales, and aggregated into five‑year bins spanning 1900–2024 to capture long‑term drought variability. In parallel, a harmonized earthquake catalog—compiled from International Seismological Centre was grouped into five‑year bins to enable direct temporal comparison with drought indices. The dataset highlights Botswana and its neighboring regions as a critical area for studying climate–tectonic interactions in southern Africa. Preliminary analysis shows that periods of severe and prolonged drought often coincide with elevated seismic activity in the Okavango Rift Zone and adjacent areas, suggesting a coupling between hydrological stress and tectonic processes. Notable findings include clustering of moderate‑magnitude earthquakes (Mw ≥ 2.5) during multi‑year drought episodes, and a tendency for seismicity rates to decline during wetter intervals. These patterns underscore the importance of considering climate variability in regional hazard assessments. Earthquake records include event identifiers, date, time, latitude, longitude, depth, magnitude, and source catalog, while SPEI files present standardized drought indices across multiple timescales. Preprocessing steps involved magnitude homogenization, duplicate removal, and spatial filtering of seismic events, alongside standardized SPEI computation using R-based scripts. The dataset is structured into tabular files (CSV/Excel) with accompanying metadata and scripts to ensure transparency and reproducibility. Potential applications include multi-hazard risk assessment, investigations of climate–tectonic interactions, and educational use in geophysics and climate science. Limitations include variations in seismic catalog completeness and uncertainties in SPEI values, particularly in arid regions.

Steps to reproduce

Methodological Framework The methodological framework adopted in this study integrates SPEI and seismic datasets into a unified analytical process designed to assess hydro tectonic interactions in Botswana and surrounding regions. The workflow consists of the following stages: 1. Data Acquisition o Seismicity: Earthquake records (1980–2023) were obtained from the International Seismological Centre (ISC) catalogue, including event dates, magnitudes, and epicentral locations. 2. Climate: Drought variability was represented using the Standardized Precipitation–Evapotranspiration Index (SPEI), derived from gridded climate archives providing monthly values across southern Africa. 3. Data Preprocessing o Seismic magnitudes were standardized into numeric format, and event dates were parsed into R Date objects. o Monthly SPEI values were aggregated into annual means to capture long term drought variability while minimizing short term fluctuations. o Both datasets were filtered to retain records within Botswana and neighboring countries for spatial consistency. 4. Temporal Aggregation o Data were grouped into five year intervals beginning in 1980. o Breakpoints were defined programmatically, and each earthquake and SPEI record was assigned to its respective interval. o Seismicity was quantified as the number of earthquakes per interval, while drought intensity was represented by mean SPEI values. 5. Data Integration o Aggregated seismic and drought datasets were merged into a unified time series, enabling direct comparison of earthquake frequency and drought variability across the study area. 6. Visualization and Analysis o Implemented in R (version ≥4.3) using reproducible workflows:  readr for data import.  dplyr for data wrangling.  lubridate for date handling.  ggplot2 for visualization.  gridExtra for assembling multi panel figures. o Two visualization approaches were employed:  Dual axis plots: Earthquake counts on the primary y axis and mean SPEI values on the secondary axis, scaled for comparability.  Panel plots: Separate panels for SPEI trends (upper) and earthquake counts (lower), preserving natural units and avoiding distortion. o Figures were exported in high resolution TIFF format (1200 dpi) to ensure publication quality. 7. Framework Characteristics o Modularity: Scripts were structured to allow replication across different SADC countries. o Reproducibility: Consistent temporal binning and aggregation methods ensure comparability across datasets. o Scalability: The framework can be extended to incorporate additional drought indices, groundwater data, or spatially disaggregated seismic catalogues.

Institutions

• University of Botswana

  Gaborone, Gaborone

Categories

Licence