Questions about Indian station data and daily observation time in Copernicus In-situ Surface Land dataset

Datasets and Usage
1

Hello everyone,

I am working with the Copernicus “In-situ observations – Surface land” dataset from the Climate Data Store (CDS). I have two specific questions about the precipitation station data for India:

  1. Which organisation provides the Indian station precipitation data included in this dataset?
    I want to confirm whether the Indian data are from the India Meteorological Department (IMD) or another provider and how they are documented in the metadata.

  2. What is the official daily observation time reference?
    For example, is the daily precipitation value based on observations up to 08:30 IST (local time) or some other time or UTC standard?

I saw in the documentation (ECMWF Confluence user guide) that contributors include NOAA/NCEI and other organisations — but I want to clarify the source/metadata for the Indian stations too.

If someone can point me to the precise metadata tables or explanations for these points, that would be very helpful.

Thanks in advance!

2

Thank you for your query.

We document the India data up to 1970 as coming from source ID 232 (International collection (non U.S. data received through personal contacts) in the C3S source configuration file, which is consistent with GHCND. I contacted our colleagues at NCEI and they confirmed that the India data were originally derived from the India Meteorological Department (IMD) via NCAR (https://data.ucar.edu/hu/dataset/india-precipitation-daily-monthly-1901-1970) and include observations up to 1970. I have checked the metadata for this dataset but it does not specify the exact observation times. The original data do come from IMD, whose primary morning observation time is 08:30 IST. This is the current standard time for recording the 24-hour cumulative rainfall and the minimum temperature for the day.

However, from 1875 to 1942, observations in India were recorded using Local Time (LT), which varied by location. Beginning on the 1st of January 1943, the observation time was standardized to 08:00 IST to support aviation requirements for synoptic (simultaneous) meteorological observations during World War II, and this was later revised to 08:30 IST in 1949 (Balasubramanian, 2010).

All post-1970 data for India originate from NOAA NCEI’s Global Summary of the Day (GSOD) source ID 166, which is derived from hourly data transmitted over the Global Telecommunications System (GTS) (https://www.ncei.noaa.gov/access/metadata/landing-page/bin/iso?id=gov.noaa.ncdc:C00516). GSOD data are aggregated on a UTC day worldwide, so the effective observation time is 2400 UTC.

Reference:
Balasubramanian, K. V. (2010) Historical evolution of rainfall measurement practices in India. MAUSAM. (Cover Story). National Institute of Science Communication and Information Resources (NISCAIR).

I hope this helps but if you have any further questions please let me know.

Simon Noone.

Copernicus Climate Change Service, C3S2 311 Bis Rescue,

Collection and Processing of in-situ Observations

3

Thank you for the earlier clarifications in this thread. I would like to seek a more specific clarification regarding the daily precipitation data available through GHCND for Indian stations, and also stated that it is converted using hourly data.

So I downloaded GHCNh station data for India from
https://www.ncei.noaa.gov/oa/global-historical-climatology-network/index.html#hourly/access/by-station/
(for example, GHCNh_INI0000VOCI_por.psv). For many Indian stations, I observe that most hourly values are missing (NA), and only a single 24-hour accumulated precipitation value is present for a given day.

As per India Meteorological Department (IMD) practice, daily rainfall is typically measured as a 24-hour accumulation from 08:30 IST to 08:30 IST. In this context, I would appreciate clarification on the following points:

  1. When a single 24-hour precipitation value is reported in GHCNh for Indian stations (with other hourly slots missing), does this value correspond to the local observation time (IST) used by IMD, or has it been converted to UTC by NCEI?

  2. For Indian stations, are the daily precipitation values in GSOD reported according to local time (IST) or are they standardized to a UTC day?

  3. Is the presence of only one daily value (with missing sub-daily values) because the original data were reported by the national agency as once-daily accumulations, rather than true hourly observations?

I ask this because, during validation of my precipitation simulations over India, I obtain significantly better agreement (higher correlation) when comparisons are made assuming IST-based daily accumulation, rather than UTC-based aggregation. I want to ensure that this improvement is physically and temporally consistent with the original observation practices, and not due to a time-reference mismatch.

Clarification on the temporal definition and time-zone handling of Indian station precipitation data in GHCNh and GSOD would be extremely helpful for ensuring correct scientific interpretation.

Thank you very much for your time and support. I look forward to your guidance.