Rainy Ottoman Days: Rescuing and Analysing Rainfall Data (1846–1917) in Constantinople (Istanbul, Türkiye)
ABSTRACT This study focuses on rescuing and analysing historical monthly rainfall data in Istanbul from 1846 to 1917 for the first time. Rainfall records from various stations, collected by foreign scientists, engineers and officials during the last century of the Ottoman Empire, were digitised in a...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2025-04-01
|
| Series: | Geoscience Data Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/gdj3.70002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | ABSTRACT This study focuses on rescuing and analysing historical monthly rainfall data in Istanbul from 1846 to 1917 for the first time. Rainfall records from various stations, collected by foreign scientists, engineers and officials during the last century of the Ottoman Empire, were digitised in accordance with the Guidelines on Best Practices for Climate Data Rescue by the World Meteorological Organisation and assessed for homogeneity. The Pettitt test was employed to identify and address inhomogeneities and detect any potential change points at each station. Monthly and annual rainfall time series (1846–2023) were reconstructed, and long‐term trends were analysed by using the Hamed and Rao Modified Mann–Kendall test to evaluate changes in Istanbul's rainfall patterns over time. Comparisons between historical data (1846–1923) and recent data (1946–2023) reveal a shift towards increased early‐year rainfall and decreased late‐year rainfall in recent decades. The study also identifies significant variability in observation data prior to 1937 compared to 20th CRv3 reanalysis data, attributed to the limited early records, with improved consistency in recent years. An analysis of reconstructed rainfall data from 1846 to 2023 revealed a significant annual decrease, with decreasing trends in August, September and November. In contrast, the 20thCRv3 reanalysis data indicates a significant annual increase with increasing trends in October over the same period. |
|---|---|
| ISSN: | 2049-6060 |