Hello Friends
I have 425 point locations with different measured attributes (like water quality parameter such as EC or TDS ). I would to do spatial clustering of each parameters but when i am trying using ArcGIS environment (mapping clusters tool box) the attribute should be incident data representing presence or absence . Is there any other software or methods for identify the spatial clustering of measured attributes?
Actually, I don't have idea about your research, but have you tried to transfer your data by using WQI?
You can use GeoDa, a spatial econometric tool of Luc Anselin a.o., University of Chicago. You can choose different clustering techniques, e.g. to determine hotspots, coldspots and non-significant clustered points. It uses the same formats like ArcGIS (shp-files), so you can exchange between them.
Gerlof Rienstra
Thank you for your suggestion GeoDa is a new information for me but i used ArcGIS Cluster and Outlier Analysis (Anselin Local Morans I), which identifies statistically significant hot spots, cold spots, and spatial outliers using the Anselin Local Moran's I statistic.
In GeoDa you have more options, I often use the local G* Cluster Map (under heading Space), using the Getis-Ord Gi* statistic.
The use of Nearest Neighbour Analysis (NNA) could also add meaning to your work.
Hi Achu!
Have you checked your data for CSR (Complete Spatial Randomness), to see what pattern your data follows (random, clustered or dispersed)? If you have established that your data show a clustered pattern, you can move on determining the clusters.
You can find information regarding the theory and implementation of different clustering techniques using ArcGIS here: http://pro.arcgis.com/en/pro-app/tool-reference/spatial-statistics/an-overview-of-the-mapping-clusters-toolset.htm
As I understand you have a point dataset. The following link provides information regarding when you should implement each clustering technique depending on your research objective: http://dusk.geo.orst.edu/ucgis/web/oregon/papers/lu.htm
It is also worth checking out this: http://pro.arcgis.com/en/pro-app/tool-reference/spatial-statistics/multi-distance-spatial-cluster-analysis.htm
Here's also a link on how you can perform spatial point pattern analysis in R: https://cran.r-project.org/web/packages/spatstat/vignettes/getstart.pdf
I hope this helped...
Good Luck!!
Best regards,
Karolina
Hi Achu,
There is plugin in QGIS calling Attribute based clustering.
See in more details about plugin here - http://www.ekazakov.info/projects/abc_tutorial.html
Best
You can cluster your water quality parameters( like EC, pH, COD, etc.) via Cluster Analysis (CA) which is a multivariate technique. You can do the cluster analysis with any of the following software; JMP 13, SPSS version 20 or any higer versions of the above mentioned software.
Hi,
I think the FCM clustering is a good method to cluster water quality data. by use of MATLAB you can do it. in this software there are many methods for clustering.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping