Centroids with predictions: Anopheles farauti predicted abundance using hurdle model (Queensland, 5km grid)

Centroids with predictions for Anopheles farauti in Queensland (5 × 5 km grid) This deposit contains an fst file (centroids_5x5_qld_with_predictions_shiny-app.fst) used to reproduce the maps and analyses in manuscript Anopheles farauti abundance in Queensland: a hurdle modelling approach, and to run the companion Shiny dashboard. What the file contains A Queensland-wide 5 km × 5 km grid represented by grid-cell centroids. A spatiotemporal panel with one row per grid cell per date (approx. 2.75 million rows). Core identifiers: grid_id, lon, lat, date, plus calendar fields (year, month, day, week) and season label (season), where season was defined as wet (December–May) and dry (June–November). Environmental / landscape covariates (e.g., elevation, water occurrence, mangrove cover, short-term weather summaries such as tmaxm21, tminm21, ppa21, rhm21). Model outputs from the hurdle framework: Suitability (presence probability) columns: pi_* (e.g., pi_brt, pi_rf, pi_max, pi_glm, pi_gam, pi_ens). Conditional mean abundance columns: mu_* (e.g., mu_rf, mu_brt, mu_glm, mu_gam, mu_ens). Size / structure Rows: ~2,751,960 Columns: ~28 File format: fst (fast on-disk storage for R; efficient to load subsets of columns) How it is used The Shiny app reads this file to: map suitability (pi_*), and map mean predicted abundance (trap-efficiency adjusted) for a chosen threshold τ and model combination, display per-grid time series and predictive intervals. Notes lon/lat are in WGS84 (EPSG:4326). date is stored as an R Date. The file may contain additional intermediate prediction columns used during model development; the dashboard and paper scripts rely primarily on the columns listed above. “Download once; don’t open in Excel; use R + fst; load only needed columns.” “How to load in R” library(fst)dt <- read_fst("centroids_5x5_qld_with_predictions_shiny-app.fst") CitationIf you use this dataset, please cite the associated manuscript and this Zenodo record (DOI 10.5281/zenodo.17984922).