Author Topic: Spatial distribution on Kuala Lumpur map  (Read 12530 times)

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Spatial distribution on Kuala Lumpur map
« on: January 06, 2011, 05:40:08 AM »
Hey all! I was assigned to do spatial distribution on the impurities level of the drinking water around Kuala Lumpur. I have the software (ArcGIS 9.3) and the long & latitude coordinates for the locations of each sampling point. Can anyone guide me on how to get the Kuala Lumpur map? And how should I really start it. I just installed the software and have no idea about it.



My supervisor would like to have something like in the picture attached. Please help!
« Last Edit: January 06, 2011, 05:42:00 AM by hhyoong »

MyGIS

  • http://www.facebook.com/groups/mygis
  • Administrator
  • phpBB Newbie
  • *****
  • Posts: 238
  • http://www.facebook.com/groups/mygis
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #1 on: January 06, 2011, 04:14:23 PM »
Hi there,

Have you thought of using the Spatial thoughtsyst Extension to create the contours? This would be enough for you to get that done.

Thanks.

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #2 on: January 07, 2011, 06:02:04 AM »
Thanks for the reply

I have google the information on GIS contour and I found a tutorial for it. But I am stuck in the coordinate system to set your own coordinate. It says "Predefined > Projected Coordinate Systems > UTM > NAD 1927 > NAD 1927 UTM Zone 13N"

http://infohost.nmt.edu/~dbutler/ERTH325/giscontmap.html

How bout Kuala Lumpur map? The main objective is to create Kuala Lumpur map and do the distribution on the it base on the level of concentration. Pls help. Thanks
« Last Edit: January 07, 2011, 06:14:54 AM by hhyoong »

MyGIS

  • http://www.facebook.com/groups/mygis
  • Administrator
  • phpBB Newbie
  • *****
  • Posts: 238
  • http://www.facebook.com/groups/mygis
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #3 on: January 07, 2011, 02:45:18 PM »
Hi,

Your are using NAD? North American Datum? Oops  :o

Never mind, we all went to that. How did you get your data? Is it from a GPS? Can you put in here one sample point? Just to check what kind of coordinates you have.

Thanks.

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #4 on: January 08, 2011, 02:03:28 AM »
Thanks again!

I have no idea what is NAD. That was actually based on the tutorial that I have googled bout "GIS contour map tutorial".

I do not really know how to start it so decided to follow the steps but stuck there.


Ya. From GPS. One of the sampling point coordinate is N03o08.748' & E101o37.927'

« Last Edit: January 08, 2011, 02:10:12 AM by hhyoong »

MyGIS

  • http://www.facebook.com/groups/mygis
  • Administrator
  • phpBB Newbie
  • *****
  • Posts: 238
  • http://www.facebook.com/groups/mygis
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #5 on: January 08, 2011, 02:52:51 AM »
Hi,

That is surely WGS84 you are using with a Degree Decimal Minute setting. You will have some problems in the conversion with the "o" and " ' " characters. You should set your coordinate systems to "World" to achieve your quest.

Cheers.

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #6 on: January 08, 2011, 05:27:53 PM »
Thanks once again!

I will try it out and get back to you.

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #7 on: January 09, 2011, 09:45:05 PM »
Hi!

Can I know is this setting correct?

GCS_WGS_1984
Datum: D_WGS_1984

Also, what is Z-factor? I tried few times but I don't understand the difference.

My Z value range from 0.02~ 0.5. But when I key in 0.1 as my Z-factor, the legend shows from 0.168~0.237. I am using IDW mode

Please guide. Thanks

MyGIS

  • http://www.facebook.com/groups/mygis
  • Administrator
  • phpBB Newbie
  • *****
  • Posts: 238
  • http://www.facebook.com/groups/mygis
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #8 on: January 10, 2011, 02:01:48 AM »
Hi,

Don't bother about the Z value as you would not need to use elevation you will be working on 2D only.

Cheers.

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #9 on: January 10, 2011, 05:32:39 AM »
Thanks for the reply.

How should I show my parameter level on the map?Aren't the colours indicate the level of the impurities? I only see Z values as the thing that shown in the legend which verifies the range of impurity.

Please guide again  :)

MyGIS

  • http://www.facebook.com/groups/mygis
  • Administrator
  • phpBB Newbie
  • *****
  • Posts: 238
  • http://www.facebook.com/groups/mygis
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #10 on: January 11, 2011, 03:44:36 PM »
hi ,

I would assume that you have Spatial thoughtsyst extension. You just need to create contours based on the impurities in the water. If you don't than you should look into setting up a symbology based on the value assigned to each point. Using a point symbology to represent the values would be good - if Spatial thoughtsyst Extension is not available - however you would not be able to create a shaded map like the one you have shown.

If you don't own ArcGIS (?) you could also look into Quantum GIS. It can read ESRI' ShapeFile and has contouring capabilities. They call it interpolation. It is under the plug-in menus.

Good luck.

Noordin

  • phpBB Newbie
  • *
  • Posts: 24
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #11 on: January 14, 2011, 07:08:27 PM »
If you are using GIS to map the distribution of impurities in drinking water, I think the methodology you are using is too simplistic.What you are doing is base on the assumption that the distribution is linear and can be interpolated and/or extrapolated. In actual fact the distribution of the impurities cannot be modeled by IDW or krigging. This is because impurities of one sample point is not directly related to the other sampling point.

But since your main task is for you to use GIS to show "some form" of distribution, inaccuracy is not only on the methodology but also contributed from the maps and data sources. I could see that you have already got some result but for confirmation, probably you could put to us (or summarize) in 5-10 steps how you have achieve the results.

NA

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #12 on: January 17, 2011, 03:30:05 PM »
Sorry for late reply as I just came back from Bali.

I will get back to you guys once I have tried it.

Thanks alot

hhyoong

  • phpBB Newbie
  • *
  • Posts: 18
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #13 on: January 20, 2011, 08:09:50 PM »
@MyGIS

Ya. I'm using ArcGIS and using the Spatial thoughtsyst  extension in the menu. There's a part where I have to key in the Z value >< (Using IDW). So I was wondering is this the one that decides contour levels.

@Noordin

Thanks for guiding me!

The steps are as below:

1. Adding XY data into a Malaysia Map(from website) with WGS 1984 settings

2. Perform the interpolation using Spatial thoughtsyst. Click the second option on General tab under option menu

3. As for thoughtsysis extent is set to Same layer as Coordinates$ Event which is the name for my excel.

4. Interpolate to Raster by selecting IDW (which I don't know which model to be used but follow the tutorial found on the net). I set the Z-value field as ppm which is my level of impurities. The Power, Search Radius type and etc remain unchanged as I have no idea what it is.

What i get is nice colourful contours surrounded by a big rectangle. How should I really fit into Kuala Lumpur Boundary only?

Is IDW the correct model to be used? Please guide on which model to be used to improve the accuracy.

Thanks.

What model would you suggest in order to provide better accuracy? Please guide.
« Last Edit: January 20, 2011, 08:13:17 PM by hhyoong »

Noordin

  • phpBB Newbie
  • *
  • Posts: 24
    • View Profile
Re: Spatial distribution on Kuala Lumpur map
« Reply #14 on: January 21, 2011, 05:58:08 PM »
First of all I have to declare that I am not an expert in modeling water impurities. However from the general knowledge that I have, the impurities in drinking-water originate from many sources such as sewage, agriculture and urban run-off, industrial wastewater, and chemicals used in the water treatment process, and also the piping.

Just a matter of curiosity where does the samples taken from - the source (river) or the water pipes? Anyway both requires you to know the coverage of the water distribution. That is to say each collection/treatment/storage center covers certain region of water distribution. Within this distribution region they will be correlation/relationship. However it will be difficult to establish relationship between the regions.

Of course a detail modeling will require you to have a distribution pattern too. As such this is totally different approach to the use of IDW to interpolate the "general" overview of the impurities content.