Skip to content

Help/ Mathematics

5 messages · Ahmed Attia, Bert Gunter, Jim Lemon +1 more

Ahmed Attia
# 
Hi R users,

I need your help to write a code in r that does the following
calculation from three different datasets;

ac = 1/sum (NPP from date 1 to date 2, dataset=1) * (biomass at date 2
-biomass at date 1, dataset = 2) + (littfall at date 2, dataset=3).

all the dates are in yr-month-day format. Which library or function
Should I use to tell R do these calculations of these variables at
different dates.

I appreciate your help.

Ahmed Attia, Ph.D.
Agronomist & Soil Scientist
Bert Gunter
# 
We expect posters to have made an effort to learn R and show us code
that they have tried. We do not provide free software development. The
"Introduction to R" tutorial that ships with the software is one place
to start, but there are many good web tutorials also available. Just
search.

Cheers,
Bert


Bert Gunter

"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Wed, Jun 21, 2017 at 11:19 AM, Ahmed Attia <ahmedatia80 at gmail.com> wrote:
Jim Lemon
# 
Hi Ahmed,
Your problem appears trivial as you have already specified the form of
the calculation.

Learn how to "extract" specified elements from a data structure:

# first value
sum(dataset1$NPP[dataset1$date >= date1 &
 dataset1$date <= date2])
# second value
dataset2$biomass[dataset2$date == date2] -
 dataset2$biomass[dataset2$date == date1]
# third value
dataset3$littfall[dataset3$date == date2]

Note that you may have to convert character strings to dates to do the
above - see a function like "as.Date". Obviously I do not know the
actual names of your datasets and I am assuming that the variable
names you have given are the actual ones.

Jim
On Thu, Jun 22, 2017 at 4:19 AM, Ahmed Attia <ahmedatia80 at gmail.com> wrote:
David Winsemius
# 
To do this in R you could learn to use the "sqldf"-package since you seem familiar with that syntax. I'm not and so it's rank speculation on my part that such an expression would be meaningful in one or another flavor of SQL.

Or you could use either `match` or `merge`, assuming the dates of the various datasets are congruent.
The "YYYY-MM-DD" format is the standard format in R. You might not even need to convert to the Date-class.
David Winsemius
Alameda, CA, USA
Ahmed Attia
# 
Hi Jim,

Thank you very much, this was so helpful.

GPP_Ahmed13$Date <- as.Date(GPP_Ahmed13$Date, '%Y/%m/%d')
Litterfall_Ahmed97$Date <- as.Date(Litterfall_Ahmed97$Date, '%Y/%m/%d')
leafbiom97$Date <- as.Date( leafbiom97$Date, '%Y/%m/%d')

(leafbiom97$LeafBiog[leafbiom97$Date == "2012-02-12"] -
  leafbiom97$LeafBiog[leafbiom97$Date == "2010-03-15"]+
  Litterfall_Ahmed97$littperiod[Litterfall_Ahmed97$Date =="2011-04-08"])/
  (sum(GPP_Ahmed13$GPP[GPP_Ahmed13$Date >= "2010-03-12" &
  GPP_Ahmed13$Date <= "2012-04-12"])/2)


Best regards

Ahmed
Ahmed Attia, Ph.D.
Agronomist & Soil Scientist
On Thu, Jun 22, 2017 at 12:24 AM, Jim Lemon <drjimlemon at gmail.com> wrote: