generate heat map based on lat/lon and data value

Hello,

I have a table with Lat/Lon and some error values. I would like to plot the lat/long and create a heatmap using the error value.

lat lon err_d

38.53 -77.40 194.4729

38.75 -77.04 147.9679

39.05 -77.91 340.3928

any help would be greatly appreciated.

Best,

KB

[Walter Roberson]

I wonder if the responses at https://www.mathworks.com/matlabcentral/answers/722079-generate-a-geographical-heat-map?s_tid=srchtitle would help ?

[Ketan Bharucha]

table has 818970 different values of lat/lon and error.

I looked at the link provided but wasn't sure how it was putting everything in a matrix form and I got an error on the reshape command.

rng('default') % for reproducibility
Lat = July1st.lat_t;  
Lon = July1st.lon_t ;
values = July1st.err_d;
valuesMatrix = reshape(values(:),numel(Lat),numel(Lon));   %MARKED
[LatMatix, LonMatrix] = ndgrid(Lat,Lon);
figure()
tiledlayout(2,2)
nexttile
h = geodensityplot(LatMatix(:), LonMatrix(:), valuesMatrix(:),'Radius',3000);
nexttile
h = geodensityplot(LatMatix(:), LonMatrix(:), valuesMatrix(:),'Radius',3000,'FaceColor','interp');
nexttile
h = geoscatter(LatMatix(:), LonMatrix(:), 600, valuesMatrix(:), 'filled','Marker','s','MarkerFaceAlpha',.4);
[latlim, lonlim] = geolimits();
geolimits(latlim+(range(latlim)*.1),lonlim) % 10% lat axis increase

Error using reshape

Number of elements must not change. Use [] as one of the size inputs to automatically calculate the appropriate size for

that dimension.

[Walter Roberson]

It sounds to me as if you might have "scattered" Lat and Lon, instead of effectively having a grid of them?

[Ketan Bharucha]

yes. They are.

[Walter Roberson]

https://www.mathworks.com/matlabcentral/fileexchange/65728-scatter-plot-colored-by-kernel-density-estimate

https://www.mathworks.com/matlabcentral/fileexchange/47165-heatscatter-plot-for-variables-x-and-y

https://www.mathworks.com/matlabcentral/fileexchange/6037-scattercloud

https://www.mathworks.com/matlabcentral/fileexchange/65166-densityplot-x-y-varargin

[Ketan Bharucha]

Thanks for your help.

Answer 1

[KSSV]

Let T be your grid...

x = T.lon ;
y = T.lat ; 
z = T.err_d ; 
% Convert the scattered data into grid 
m = 100 ;   % can be increased 
[X,Y] = meshgrid(linspace(min(x),max(x),m),linspace(min(y),max(y),m)) ; 
Z = griddata(x,y,z,X,Y) ;
pcolor(X,Y,Z)
shading interp 
colorbar 

[Ketan Bharucha]

Thank you, above helpped solve my problem.

Answer 2

[Peter Perkins]

The heatmap function accepts a table as input, this should be a one-liner. If you are looking for a surface plot, KSSV's answer should work.