ThaiGovData

Rainfall Telemetry Station Clustering Analysis Results

Methodology for clustering rainfall telemetry stations of the Water Resources Information Center (WRIC). 1. Calculate the monthly average rainfall for each WRIC station for the period 2014–2023 to be used as Training Data, and use the remaining rainfall data as Test Data. 2. Perform Dimensionality Reduction using Principal Component Analysis (PCA) with the following variables: - Latitude - Longitude - Scaled Rainfall 3. Define data clusters using Euclidean Distance between the centroids of each cluster and determine the optimal number of clusters for this dataset. The clustering of stations resulted in a total of 12 clusters. Additional Explanations 1. stationcode refers to the station code. 2. latitude, longitude are the station coordinates. 3. cluster refers to the group of stations clustered according to the rainfall telemetry station clustering methodology.

Agency:
hii
Licence:
Creative Commons Attribution Non-Commercial
Category:
ข้อมูลสาธารณะ
Geographic coverage:
พิกัด
Update frequency:
ปี
Source portal:
data_go_th
Updated:
25/08/2026

English title machine-translated from the Thai original

Record updated 17 days ago, within its stated annual cycle

The same record appears 1 more times on: gdcatalog — counted once here

clusteringrainfalltelemetry stationการจัดกลุ่มปริมาณฝนสถานีโทรมาตร

What is in the file

First 10 of 1,302 rows · 4 columns

stationcodetextlatitudenumericlongitudenumericclusternumeric
BTGH5.775882101.0924452
MOU4545.785542101.055862
MOU1425.8074923101.8451841
SKN0025.8275533101.752351
SKN0015.8356757101.734851
BUKT5.848724101.880312
MOU4355.884526101.022082
FOP0715.887238101.695751

From the PDF

Extracted from the PDF's own text layer, not by OCR. The original file is the record; this is an extract of it.

Hierarchical clustering of HII rainfall stations1) Calculate the average monthly rainfall for each HII station from 2014-2023 as training dataand remaining rainfall is test data2) Reduce dimensionality using PCA with 'latitude', 'longitude', 'scaled_rain'3) Determine clusters using the Euclidean distance between cluster centroid and identify theoptimal distance for the number of clustersResults:HII Training data : 2014-2023 HII Test data : 2024Method: centroid : 𝑑(𝑠, 𝑡) = ‖𝑐௦ + 𝑐௧‖ଶ where 𝑐௦ and 𝑐௧ are the centroids of clusters 𝑠 and 𝑡 ,respectively. When two clusters are combined into a new cluster 𝑢 , the new centroid is computed overall the original objects in clusters. The distance then becomes the Euclidean distance between thecentroid of and the centroid of a remaining cluster in the forest.Optimal distance of Euclidean distance is 2.8

Data files (2)

This dataset carries no explicit licence, so we link to the publisher’s file rather than redistributing a copy. Check the terms of use with the publishing agency.

  • CSVข้อมูลการจัดกลุ่มสถานีโทรมาตรวัดปริมาณฝน สสน.37 KB

    ข้อมูลการจัดกลุ่มสถานีโทรมาตรวัดปริมาณฝน สสน.

    View at source →
  • PDFวิธีการจัดกลุ่มของสถานีโทรมาตรปริมาณฝน สสน.755 KB

    วิธีการจัดกลุ่มของสถานีโทรมาตรปริมาณฝน สสน.

    View at source →

More from this agency

All datasets from this agency