Summing elements in an array between uneven date intervals
Afficher commentaires plus anciens
I have a dataset of daily rainfall. I also have a list of 28 days on which environmental sampling was carried out. I'm trying to figure out how to sum the total precipitation that fell between each environmental sampling (represented by "sample_dates")
The environmental samplings are mostly about 30 days apart, but not exactly. I already know how to find the monthly precipitation, but what I want is the exact amount of rainfall that fell between each sampling.
Any advice on what functions to use would be so appreciated. I've read documentation for cumsum, ismember, movsum, and a bunch of previously asked questions but I can't find anything that's helped me come up with a solution.
T = readtable("precip_test.csv", "VariableNamingRule","preserve");
A = table2array(T);
%sequential dates from Sept 2019 to Dec 2022
rain_dates = A(:,1);
%daily precipitation in mm for each day
rain_mm = A(:,2);
%dates that environmental samples were taken
sample_dates = A(:,3);
sample_dates(any(isnan(sample_dates), 2), :) = [];
Réponse acceptée
Star Strider
le 6 Oct 2023
Modifié(e) : Star Strider
le 6 Oct 2023
I am not certain what you want. If I am readiing the file correctly and converting the dates correctly (both 'excel' and 'posixtime' give reasonable results for the dates, however the ‘Date Time’ and ‘Sample Dates’ do not make sense with 'excel') the ‘Sample Dates’ seem to give appropriate results with ‘Precip (mm)’ so it would seem that with that we are finished and nothing further need be done.
T = readtable("precip_test.csv", "VariableNamingRule","preserve")
DateTime = datetime(T.Date, 'ConvertFrom','excel');
SampleDates = datetime(T.('Sample dates'), 'ConvertFrom','excel');
T2 = table(DateTime, T.('Precip (mm)'), SampleDates, 'VariableNames',{'Date Time','Precip (mm)','Sample Dates'})
T2 = rmmissing(T2); % Remove 'NaT' Rows
LastLine = T2(end,:)
VN = T2.Properties.VariableNames;
figure
stairs(T2.('Date Time'), T2.('Precip (mm)'))
grid
xlabel(VN{3})
ylabel(VN{2})
return % Stop Here
A = table2array(T);
%sequential dates from Sept 2019 to Dec 2022
rain_dates = A(:,1);
%daily precipitation in mm for each day
rain_mm = A(:,2);
%dates that environmental samples were taken
sample_dates = A(:,3);
sample_dates(any(isnan(sample_dates), 2), :) = [];
What else do you want to do with these resullts?
.
15 commentaires
Emily
le 6 Oct 2023
Star Strider
le 6 Oct 2023
Modifié(e) : Star Strider
le 6 Oct 2023
O.K. I changed that in my earlier code, so they now convert from 'excel' however now the 'Sample Dates' do not make sense in the context of 'Date Time'.
What do we do with the date information?
T = readtable("precip_test.csv", "VariableNamingRule","preserve")
DateTime = datetime(T.Date, 'ConvertFrom','excel');
SampleDates = datetime(T.('Sample dates'), 'ConvertFrom','excel');
T2 = table(DateTime, T.('Precip (mm)'), SampleDates, 'VariableNames',{'Date Time','Precip (mm)','Sample Dates'})
T2 = rmmissing(T2); % Remove 'NaT' Rows
LastLine = T2(end,:)
VN = T2.Properties.VariableNames;
figure
stairs(T2.('Date Time'), T2.('Precip (mm)'))
grid
xlabel(VN{3})
ylabel(VN{2})
TT = table2timetable(T2(:,[1 2]))
TTrainfall = retime(TT, 'monthly','sum')
% return % Stop Here
%
% A = table2array(T);
%
% %sequential dates from Sept 2019 to Dec 2022
% rain_dates = A(:,1);
%
% %daily precipitation in mm for each day
% rain_mm = A(:,2);
%
% %dates that environmental samples were taken
% sample_dates = A(:,3);
% sample_dates(any(isnan(sample_dates), 2), :) = [];
Since there is only one month, there is only one result, showing that for the data available, September 2019 had 80 mm of liquid precipitation.
.
Emily
le 6 Oct 2023
Star Strider
le 6 Oct 2023
I do not understand the sampling dates. They make no sense with respect to the first column dates.
How are they supposed to be used?
Emily
le 6 Oct 2023
Emily
le 6 Oct 2023
Star Strider
le 6 Oct 2023
The file contains only data for September 2019, and the first sampling interval is betweeen 4 November and 3 December 2019.
That makes absolutely no sense.
Emily
le 6 Oct 2023
Star Strider
le 6 Oct 2023
Modifié(e) : Star Strider
le 7 Oct 2023
The dates are truncated because several of the 'Sample Dates' column were listed as NaT.
Separating out 'Sample Dates' to a separate table permits logical indexing based on its elements.
Try this —
T = readtable("precip_test.csv", "VariableNamingRule","preserve")
DateTime = datetime(T.Date, 'ConvertFrom','excel');
SampleDates = datetime(T.('Sample dates'), 'ConvertFrom','excel');
T2 = table(DateTime, T.('Precip (mm)'), SampleDates, 'VariableNames',{'Date Time','Precip (mm)','Sample Dates'})
VN = T2.Properties.VariableNames;
T3 = T2(:,3); % Separate 'Sample Dates'
T3 = rmmissing(T3)
T2 = T2(:,[1 2])
LastLine = T2(end,:)
PrecipStats = table('Size',[size(T3,1)-1,3], 'VariableTypes',{'datetime','datetime','double'});
for k = 1:size(T3,1)-1
Lv = T2{:,1} >= T3{k,1} & T2{:,1} < T3{k+1,1};
PrecipStats(k,:) = {T3{k,1}, T3{k+1,1}, sum(T2{Lv,2})};
end
PrecipStats.Properties.VariableNames = {'Start Date','End Date','Precip Total'}
LastLine = PrecipStats(end,:)
figure
stairs(T2.('Date Time'), T2.('Precip (mm)'), 'DisplayName',VN{2})
hold on
stairs(PrecipStats.('Start Date'), PrecipStats.('Precip Total'), 'DisplayName','Interim Totals (mm)')
hold off
grid
xlabel(VN{3})
ylabel(VN{2})
legend('Location','NW')
.
Emily
le 7 Oct 2023
My pleasure!
That line creates the logical vector, ‘Lv’. (I could have used find, to determine the numerical indices, and while that would work and in some situations is necessary, it is unnecessary here and inefficient to do the same operation, that being to find the rows in ‘T2’ that meet that specific condition.)
A somewhat simpler and more compact illustration —
T9 = array2table([1:10; randn(1,10)]')
Lv = T9{:,1} >= 3 & T9{:,1} < 7
Out = T9(Lv,:)
That’s essentially all there is to it.
.
Star Strider
le 7 Oct 2023
If my Answer helped you solve your problem, please Accept it!
.
Emily
le 7 Oct 2023
Star Strider
le 7 Oct 2023
As always, my pleasure!
Plus de réponses (0)
Catégories
En savoir plus sur Calendar dans Centre d'aide et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
