Error in interpolating: Actual data does not match with interpolated data
4 views (last 30 days)
Show older comments
Hi,
I am trying to interpolate my gridded data of dimension (longitude x latitude x altitude) at specific latitude and altitude which are not included in the dataset. To check my code and verify if my interpolated data is correct, I tried to interpolate at the point (Lat = 15 Degree & Alt = 100 km) which is included in the original data. By surpise, the plots for interpolated data and original data didnt match. I am not sure what I am doing wrong so requesting if you could help me out.
Below is my matlab code (attached) and link for my data:
Link for my data: https://drive.google.com/drive/folders/195GxDtuUIGRxqMSVsWPZpWeapJ9s2kEc?usp=share_link
%% read data
file1='data_nc\data_10x10\data.nc'; %76x94x54
Temp1_4d = ncread(file1,'Temperature');
Temp1_4d =Temp1_4d(3:38,3:20,:);
Lat1_4d = ncread(file1,'Latitude')*(180/pi);
Lat1_4d = Lat1_4d(3:38,3:20,:);
Long1_4d = ncread(file1, 'Longitude')*(180/pi);
Long1_4d =Long1_4d(3:38,3:20,:);
Alt1_4d = ncread(file1,'Altitude');
Alt1_4d = Alt1_4d(3:38,3:20,:);
Temp1_4d = double(Temp1_4d);% converts single data to double
Temp1a_4d = Temp1_4d(19:36,:,:);Temp1b_4d =Temp1_4d(1:18,:,:);Temp1c_4d = [Temp1a_4d;Temp1b_4d];
Long1_4d =double(Long1_4d);Long1_4d = wrapTo180(Long1_4d);
Long1a_4d = Long1_4d(19:36,:,:);Long1b_4d = Long1_4d(1:18,:,:);Long1c_4d = [Long1a_4d;Long1b_4d];
Lat1_4d = double(Lat1_4d);
Lat1a_4d = Lat1_4d(19:36,:,:);Lat1b_4d = Lat1_4d(1:18,:,:);Lat1c_4d = [Lat1a_4d;Lat1b_4d];
Alt1_4d = double(Alt1_4d);
Alt1a_4d = Alt1_4d(19:36,:,:);Alt1b_4d = Alt1_4d(1:18,:,:);Alt1c_4d = [Alt1a_4d;Alt1b_4d];
Long1 = Long1c_4d(:,1,1); %reads vector of Long
Lat1 = Lat1c_4d(1,:,1); %reads vector of Long
Alt1 = Alt1c_4d(1,1,:);%reads vector of Alt
Alt1 = permute(Alt1,[1 3 2]);
%% interpolate
LUT1=griddedInterpolant({Long1,Lat1,Alt1}, Temp1_4d); %interpolation object 2
TempvLong1=LUT1({Long1,15,100000}); %slices
TempvLong1b=LUT1({Long1,15,100000}); %slices
%%Plot
tiledlayout(1,1)
nexttile
hold on;
plot(Long1(:),TempvLong1(:),'-hexagram','Color','k','MarkerSize',2,'MarkerEdgeColor','k','MarkerFaceColor','k');
plot(Long1c_4d(:,11,3),Temp1c_4d(:,11,3),'-hexagram','Color','b','MarkerSize',2,'MarkerEdgeColor','b','MarkerFaceColor','b');
xlabel 'Longitude';
ylabel 'Temperature'
title 'Lat = 15^o & Alt = 100 km';
%legend ('72x54 cells (5 long x 3.33 lat)','72x90 cells (5 long x 2 lat)','72x180 cells (5 long x 1 lat)','120x90 cells (3 long x 2 lat)');
%legend('location','north','box','on','FontSize',7);
legend('location','north','box','on','FontSize',7);
legend('interpolated data', 'origina data');
xlim([-180 180]);
ax = gca;
set(ax,'Box','on');
ax.XTick = -180:60:180;
hold off;
7 Comments
Accepted Answer
Matt J
on 29 Nov 2022
Edited: Matt J
on 29 Nov 2022
load inputData
i=11; j=3; %query points (Lat and Alt)
%% interpolate
LUT1=griddedInterpolant(Long1c_4d,Lat1c_4d,Alt1c_4d, Temp1c_4d); %interpolation object
[Long1,Lat1,Alt1]=deal(LUT1.GridVectors{:}); %1D versions of the coordinates
TempvLong1=LUT1({Long1,Lat1(i),Alt1(j)}); %slices
%%Plot
tiledlayout(1,1)
nexttile
hold on;
plot(Long1(:),TempvLong1(:),'-hexagram','Color','k','MarkerSize',2,'MarkerEdgeColor','k','MarkerFaceColor','k');
plot(Long1c_4d(:,i,j),Temp1c_4d(:,i,j),'o','Color','r','MarkerSize',8,'MarkerEdgeColor','r');
xlabel 'Longitude';
ylabel 'Temperature'
title(sprintf('Lat = %.2f^o & Alt = %.2f km',Lat1(i),Alt1(j)/1000));
legend('location','northeast','box','on','FontSize',12);
legend('interpolated data', 'original data');
xlim([-180 180]);
ax = gca;
set(ax,'Box','on');
ax.XTick = -180:60:180;
hold off;
More Answers (0)
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!