|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | Re: [gislist] How to calculate the average annual rainfall from
33rd and 67th percentiles grid data |
| Date: |
08/30/2005 03:25:01 AM |
| From: |
Botanical surveys |
|
|
Leonie,
Am I correct in assuming that the data are collected from a range of
farms and not concentrated in a few locations? If randomly collected
from a number of farms, the calculations used for obtaining a percentile
require that the data is derived from a normal distribution. Given that
you've indicated that the data are skewed, you may have enough
_randomly _derived samples (n>100) to invoke the central limit theorum.
If so, the following formula would provide any percentile required. If
not, you'll need to obtain more data.
Order the data from lowest to highest values and use the formula:
n x p/100
where n = sample size, and p = desired percentile. If the result is not
an integer, round up to next integer. This is the ranked sample that
equates with the percentile.
Hope this helps.
Rgrds
Claire deLacey
Leonie Meiring wrote:
>Dear Listers
>
>How does one calculate the average annual rainfall if only the 33rd
>percentile and 67the percentile is known along with the mean and standard
>deviation?
>
>I have 2 grid annual rainfall datasets, one for the 33rd percentile (dry
>season) and one for the 67th percentile (wet season). I have a list of
>farms for which I need the average annual rainfall. For example:
>
>For one farm the 33rd annual rainfall percentile = 434.3535 mm. There are
>492 296 columns and rows in the whole dataset with a mean of 299.9189 and
>standard deviation of 135.5186.
>The Min value is 60.4778 and the Max value is 981.5317 for the whole
>dataset.
>
>The 67th annual rainfall percentile = 552.0655 mm. There are again 492 296
>columns and rows in the whole dataset with a mean of 433.3182 and standard
>deviation of 173.31943.
>The Min value is 80.0178 and the Max value is 1305.12377 for the whole
>dataset.
>
>>From the histograms it would appear that the data is not symmetrical -
>leaning more towards the left (sorry, I am not a statistician).
>
>Could one still use the following: (67th percentile + 33rd percentile) / 2
>to get the annual average rainfall i.e. 50th percentile which would be in
>the above-mentioned case (552.0655 + 434.3535) / 2 = 493.2095 mm or is there
>another formula one can use?
>
>Your help and advice is much appreciated.
>
>Kind regards
>
>Leonie Meiring
>Department of Agriculture
>
>
>
>
>This email message has been scanned by Symantec Antivirus for the presence of computer viruses.
>_______________________________________________
>gislist mailing list
>gislist@lists.geocomm.com
>http://lists.geocomm.com/mailman/listinfo/gislist
>
>_________________________________
>This list is brought to you by
>The GeoCommunity
>http://www.geocomm.com/
>
>!DSPAM:43140b08149661061112313!
>
>
>
--
Regards
Claire deLacey
Bangalay Botanical Surveys
bangalay@bangalay.com.au
www.bangalay.com.au
ph 02 9544 3941
m 0428 954 439
_______________________________________________
gislist mailing list
gislist@lists.geocomm.com
http://lists.geocomm.com/mailman/listinfo/gislist
_________________________________
This list is brought to you by
The GeoCommunity
http://www.geocomm.com/
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
