20250916 meeting

Published 2025-09-15 Updated 2025-09-15 1619 words 8 minutes views

Contents

In this study, three real-world datasets are identified for future analysis. After conducting preliminary data preprocessing, one of the datasets will be selected for experiments. Each dataset is first evaluated for missing values to determine its suitability.

In addition to using real datasets as experimental targets, future work should explore other datasets with spatiotemporal covariance structures for further analysis.

NOAA Monthly U.S. Climate Gridded Dataset

The NOAA Monthly U.S. Climate Gridded Dataset (NClimGrid) is provided by the U.S. government. It contains monthly records of maximum temperature, minimum temperature, average temperature, and precipitation across the United States, covering data from 1895 to the present. Hawaii was included in 1991, and Alaska was added in March 2015.

/1140916-meeting/image/NOAA%20%E6%9C%88%E5%BA%A6%E7%BE%8E%E5%9C%8B%E6%B0%A3%E5%80%99%E7%B6%B2%E6%A0%BC%E8%B3%87%E6%96%99%E9%9B%86.png — NOAA Monthly U.S. Climate Gridded Dataset

Website
- https://www.ncei.noaa.gov/access/metadata/landing-page/bin/iso?id=gov.noaa.ncdc%3AC00332
Data format
- .txt
- NetCDF (Version: 4)

A preliminary check revealed that the dataset contains about 43% missing values.

Meteostat

Meteostat is a Python package that collects and integrates global weather data. Its meteorological records are mainly sourced from institutions such as the National Oceanic and Atmospheric Administration (NOAA), the German Weather Service (DWD), and Environment Canada. It provides information from thousands of weather stations worldwide, including both decommissioned and active stations, with historical records traceable as far back as the 19th century. This makes it a highly comprehensive global dataset. However, since most of the data consists of raw station values, it contains many missing entries caused by mechanical or human factors, requiring significant preprocessing before use.

Website
- https://meteostat.net/en/
- GitHub: https://github.com/meteostat/meteostat-python
Data format
- Python package

The station information is as follows:

Execution result reference

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
                    name country region    wmo  icao  latitude  longitude  \
id                                                                          
00FAY        Holden Agdm      CA     AB  71227  CXHD   53.1900  -112.2500   
00TG6        Athabasca 1      CA     AB   <NA>  <NA>   54.7200  -113.2900   
01001          Jan Mayen      NO   <NA>  01001  ENJA   70.9333    -8.6667   
01002           Grahuken      NO     SJ  01002  <NA>   79.7833    14.4667   
01003           Hornsund      NO   <NA>  01003  <NA>   77.0000    15.5000   
01004     New Alesund II      NO     SJ  01004  ENAS   78.9167    11.9333   
01005        Barentsburg      NO     SJ  01005  <NA>   78.0667    13.6333   
01006            Edgeoya      NO     SJ  01006  <NA>   78.2333    22.7833   
01007        New Alesund      NO     SJ  01007  <NA>   78.9167    11.9333   
01008  Svalbard Lufthavn      NO   <NA>  01008  ENSB   78.2500    15.4667   

       elevation             timezone hourly_start hourly_end daily_start  \
id                                                                          
00FAY      688.0     America/Edmonton   2020-01-01 2024-12-07  2002-11-01   
00TG6      515.0     America/Edmonton          NaT        NaT  2000-01-01   
01001       10.0          Europe/Oslo   1931-01-01 2025-03-20  1921-12-31   
01002        0.0          Europe/Oslo   1986-11-09 2025-03-20  2010-10-07   
01003       10.0          Europe/Oslo   1985-06-01 2025-03-20  2009-11-26   
01004        8.0          Europe/Oslo   1973-01-01 2014-05-23  1968-12-31   
01005        9.0  Arctic/Longyearbyen          NaT        NaT         NaT   
01006        0.0          Europe/Oslo   1973-01-01 2025-03-20  2010-10-07   
01007        0.0          Europe/Oslo          NaT        NaT  1973-03-28   
01008        2.0          Europe/Oslo   1975-09-29 2025-09-10  1975-08-01   

       daily_end monthly_start monthly_end  
id                                          
00FAY 2024-03-13    2003-01-01  2022-01-01  
00TG6 2022-07-12    2000-01-01  2010-01-01  
01001 2025-08-24    1922-01-01  2022-01-01  
01002 2020-08-17           NaT         NaT  
01003 2020-08-31    2016-01-01  2017-01-01  
01004 1997-03-01    1969-01-01  1974-01-01  
01005        NaT    1951-01-01  1980-01-01  
01006 2020-08-23           NaT         NaT  
01007 2025-08-24    1974-01-01  2022-01-01  
01008 2025-08-24    1975-01-01  2022-01-01  

A global station map is shown below:

/1140916-meeting/image/%E4%B8%96%E7%95%8C%E7%AB%99%E9%BB%9E%E5%9C%B0%E5%9C%96.png — Global station map

Initially, the plan was to analyze Taiwan stations. However, since the dataset contains too few stations in Taiwan, the analysis instead focuses on U.S. stations.

/1140916-meeting/image/%E8%87%BA%E7%81%A3%E7%AB%99%E9%BB%9E%E5%9C%B0%E5%9C%96.png — Taiwan station map

/1140916-meeting/image/%E7%BE%8E%E5%9C%8B%E7%AB%99%E9%BB%9E%E5%9C%B0%E5%9C%96.png — U.S. station map

After filtering for the time period with the most available stations, the years 2016–2019 were selected for analysis.

Execution result reference

1
2
3
4
                      0     1     2     3     4     5     6     7     8     9     10    11    12    13    14    15    16    17    18    19    20    21    22    23    24    25    26    27    28    29    30    31    32    33    34    35    36    37    38    39    40    41    42    43    44    45    46    47    48    49    50    51    52    53    54    55    56    57    58    59    60    61    62    63    64    65    66    67    68    69    70    71    72    73    74    75    76    77    78    79    80    81    82    83    84    85    86    87    88    89    90    91    92    93    94    95    96    97    98    99    100   101   102   103   104   105   106   107   108   109   110   111   112   113   114   115   116   117   118   119   120   121   122   123   124   125   126   127   128   129   130   131   132   133   134   135   136   137   138   139   140   141   142   143   144   145   146   147   148   149   150   151   152   153   154   155   156   157   158   159   160   161   162   163   164   165
Year                 1857  1858  1859  1860  1861  1862  1863  1864  1865  1866  1867  1868  1869  1870  1871  1872  1873  1874  1875  1876  1877  1878  1879  1880  1881  1882  1883  1884  1885  1886  1887  1888  1889  1890  1891  1892  1893  1894  1895  1896  1897  1898  1899  1900  1901  1902  1903  1904  1905  1906  1907  1908  1909  1910  1911  1912  1913  1914  1915  1916  1917  1918  1919  1920  1921  1922  1923  1924  1925  1926  1927  1928  1929  1930  1931  1932  1933  1934  1935  1936  1937  1938  1939  1940  1941  1942  1943  1944  1945  1946  1947  1948  1949  1950  1951  1952  1953  1954  1955  1956  1957  1958  1959  1960  1961  1962  1963  1964  1965  1966  1967  1968  1969  1970  1971  1972  1973  1974  1975  1976  1977  1978  1979  1980  1981  1982  1983  1984  1985  1986  1987  1988  1989  1990  1991  1992  1993  1994  1995  1996  1997  1998  1999  2000  2001  2002  2003  2004  2005  2006  2007  2008  2009  2010  2011  2012  2013  2014  2015  2016  2017  2018  2019  2020  2021  2022
Monthly_Start_Count     1     0     0     0     0     0     0     0     0     0     0     0     0     0     0     1     0     0     0     0     1     0     0     0     1     0     0     0     1     1     0     1     2     0     1     2     7     3     4     4     3     4     3     3     3     0     1     2     4     1     1     1     4     3     1     0     0     0     3     3     5     3     1     3     0     0     3     1     1     3     0     5     1     6     4     4     2     3     2     7    11    21    20    13    20    56    47    20    31    23     7    65    19     8    14     1    11     5     3     3     3     1     7     2     8     2     4     3     3     3     1     1     2     1     1     4     4     9     5     2     3     1     3     1    24     1     1     2     0     1     2     2     2     2     4     1     1     3     4     4     4    13    27    15    23    10     1     8    68   409    89    73    53    62   108    48    17   124   101    33    30     9     9   308     7     1
Monthly_End_Count       0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     0     1     3     0     0     0     1     0     0     0     2     0     0     0     0     0     1     1     0     1     1     2     0     0     1     0     2     1     2     0     0     2     1     1     1     0     1     0     1     1     1     1     3     1     1     1     1     0     1     1     2     1     2     4     0     2     0     0     1     2     0     0     1     1     3     1     2     3     2     3     3     9     6    12     9     5     7    37   449  1695

/1140916-meeting/image/2016-2019%20%E5%B9%B4%E9%96%93%E5%AD%98%E5%9C%A8%E7%9A%84%E7%BE%8E%E5%9C%8B%E7%AB%99%E9%BB%9E%E5%9C%B0%E5%9C%96.png — Map of U.S. sites from 2016-2019

The number of missing values for each field is as follows:

Execution result reference

1
2
3
4
5
6
7
8
9
Missing values per field:
tavg      6492
tmin     15217
tmax     15172
prcp    113705
wspd    178922
pres     32065
tsun    177826
dtype: int64

Sample data presentation:

Execution result reference

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
            tavg  tmin  tmax  prcp  wspd  pres  tsun
station_id                                          
69015          0    63    63   141   141    35   141
69023          4    10    10    41   167     4   167
70000          0     4     4   162   162    30   162
70026          0     0     0     0   168     0   168
70030          5     3     3    42   148    23   148
70086          1     0     0   133   137   127   137
70104          0    57    57   116   116    48   116
70116          0    41    41    56    56    14    56
70117          0    50    50   118   118    49   118
70121          1    65    65   128   128    29   128
70133          0     0     0     0   168     0   168
70173          0    36    36    89    89    41    89
70174          0     1     1    50   168     0   168
70178          0     0     0     2   168    27   168
70194          0    49    49    75    75    59    75
70200          0     0     0     0   168     0   168
70207          0    29    29   149   149   115   149
70219          0     0     0     0   168     0   168
70222          1    37    37   122   127    43   127
70231          0     1     1     0   168     1   168
70232          0    40    40   134   134   115   134
70235          0    52    52   130   130    56   130
70246         23    25    25    84   161   146   161
...
PATL0          0    42    42   103   103    44   103
PHKO0          2    11    11   167   167     5   167
PHSF0          0    39    39    64    64    40    64
U9ANI        140     0     0    15   140   140   140

Weather2K

Weather2K is a benchmark spatiotemporal meteorological forecasting dataset based on real-time ground station observations. It collects hourly data from 2,130 ground weather stations and provides 20 meteorological variables along with 3 constant location attributes, with a total of 40,896 time steps. This dataset was proposed in a research paper from China.

Paper
- https://arxiv.org/pdf/2302.10493
Website
- https://arxiv.org/abs/2302.10493
- GitHub: https://github.com/Xun-Zhu/weather2k
- Hugging Face: https://huggingface.co/datasets/BUPT-PRIS-727/Weather2K
Data format
- [Not yet specified]

Data preprocessing has not yet been conducted.

Contents

20250916 meeting

NOAA Monthly U.S. Climate Gridded Dataset

Meteostat

Weather2K

Related Posts