Mahoney Humidity Groups and Thermal Stress Indicators for Ten Indonesian Provincial Capitals Derived from ERA5 Reanalysis (2014 to 2023)
Description
This dataset provides monthly climate records and Mahoney Table classification outputs for ten provincial capitals in Indonesia from January 2014 to December 2023. The cities were Jakarta Pusat, Serang, Bandung, Yogyakarta, Semarang, Surabaya, Jambi, Pekanbaru, Samarinda, and Mataram. Climate variables were retrieved from the Open-Meteo Historical Weather API, which draws on the ERA5 and ERA5-Land reanalysis products of the European Centre for Medium-Range Weather Forecasts. Values were extracted at one coordinate per city and aggregated to monthly values. The file contains 1,200 records, one for each city month. Each record holds the maximum and minimum dry bulb temperature, the maximum and minimum relative humidity, and the monthly precipitation, also prevailing wind directions. It also holds the monthly means of temperature and relative humidity derived from those extremes. Every month was assigned to a Mahoney Humidity Group from its mean relative humidity, following Koenigsberger et al. (1971). A second Humidity Group was assigned from the annual mean of each city year, so the monthly and annual bases can be compared directly. Only the G3 and G4 groups occurred in the record. Each month was also tested against the Mahoney comfort limits. This produced daytime and nocturnal Thermal Stress Indicators coded as hot, comfortable, or cold. The dataset covers only the classification stage of the Mahoney Table. Design recommendations were not computed. No trend tests, period comparisons, or composite scores are included. The input variables sit next to the labels, so users can check each classification or rerun it with other thresholds. All values are reanalysis estimates for grid cells of roughly 9 to 25 kilometers, not station observations. They may differ from local BMKG records, particularly in Bandung and in dense urban cores. The dataset may support research in building climatology, bioclimatic design, and urban climate adaptation that needs a consistent monthly Mahoney classification across several Indonesian cities.
Files
Steps to reproduce
1. Locations. Ten provinces were selected in an associated screening study based on BPS population density and IKLH data. That screening is not part of this dataset. Each province was represented by its capital, the area of highest population concentration. 2. Coordinates. One reference point per capital was taken from the regional government website and validated in Google Maps: Jakarta Pusat (-6.2088, 106.8456); Serang (-6.1153, 106.1542); Bandung (-6.9147, 107.6098); Yogyakarta (-7.7956, 110.3695); Semarang (-6.9704, 110.4127); Surabaya (-7.2506, 112.7508); Jambi (-1.6097, 103.6096); Pekanbaru (0.5054, 101.4477); Samarinda (-0.4947, 117.1519); Mataram (-8.5904, 116.1268). 3. Retrieval. Daily data for 2014 to 2023 were retrieved from the Open-Meteo Historical Weather API (archive endpoint), which serves ERA5 and ERA5-Land reanalysis. Variables were maximum and minimum air temperature and relative humidity at 2 m, and precipitation. Days were defined in local time (UTC+7 for WIB cities, UTC+8 for Samarinda and Mataram). 4. Software. Processing used Rx Climata, an unreleased in-house tool linking the API to a Mahoney Table spreadsheet. All computations are described below so they can be reproduced in any spreadsheet or script. 5. Monthly values. For each month, DBT_Max_C and DBT_Min_C were the monthly means of daily maximum and minimum temperature. RH_Max_Pct and RH_Min_Pct were the monthly means of daily maximum and minimum relative humidity. Rainfall_mm was the monthly sum of daily precipitation. 6. Derived values. Temperature_Avg_C = (DBT_Max_C + DBT_Min_C) / 2. Average_RH_Pct = (RH_Max_Pct + RH_Min_Pct) / 2. DBT_Range_C = DBT_Max_C - DBT_Min_C. 7. Humidity Groups. Each month was classified from Average_RH_Pct: G1 < 30%, G2 30 to 50%, G3 50 to 70% (70.0 included), G4 > 70%. Humidity_Group_Annually applied the same thresholds to the annual mean of Average_RH_Pct per city year. 8. Thermal stress. Annual mean temperature exceeded 20 C in every city year, so the Mahoney limits for that band were used. DBT_Max_C was compared with day limits (G3 23 to 29 C; G4 22 to 27 C) and DBT_Min_C with night limits (G3 17 to 23 C; G4 17 to 21 C). Codes: H above, O within, C below the limits. 9. Scope. Only the classification stage was computed. Mahoney design indicators and recommendations were not. 10. Checks. The file was checked for 120 records per city, 1,200 in total, unique city, year, and month combinations, sequential record numbers, and no missing values.
Institutions
- Binus UniversityJakarta, Jakarta