How to prevent pcolor from joining gaps in datetime?

I am using pcolor to create a contour graph where the x-axis is a datetime vector. Data exists in 6 s increments from 09:00:00 to 09:03:54 and then from 09:30:00 to 09:33:54 only:

09:00:00

09:00:06

09:00:12

...

09:03:42

09:03:48

09:03:54

09:30:00

09:30:06

...

When using pcolor, it joins the last sample of the first time span with the first sample of the second time span. See attached image. How do I make it such that the time gap between 09:03:54 and 09:30:00 is not plotted? I would like for this region to just be blank or transparent or otherwise not visible.

Answer 1

[Image Analyst]

Try setting those values to nan.

Or else use imshow() and set those values to 0 (black).

By the way, did you know that pcolor() does not display the last row and alst column of your data?

[David Velasco]

Thanks, but the issue is that there are no values there. If there were, yes, I could easily set them to NaN or 0 or something else. There's a gap in time becase no data was collected there. I suppose I could create datetime vector that would contain time stamps between 09:03:54 and 09:30:00 and fill the data with NaN, but this seems like poor and convoluted solution to the problem.

Yes, I'm aware of that issue with pcolor, but for the types of plots that I need to make and the type of data I work with, pcolor is the best option. Tried working with surf, imagesc, contour and others and eventually settled on pcolor. But thanks for the reminder.

[David Velasco]

Ah, nevermind. I get what you were saying. "those" values you meant the last sample of the first time span. Yes, that does the trick. Not the perfect solution (I'm loosing that sample, which is valid), but it gets me what I need. Thank you.

[Image Analyst]

OK, then could you please "Accept this answer"? Thanks in advance.

[Maria-Eleni Gidarakou]

Can you share that part of your script, because i am having the same problem, i am not sure how I can set this values to nan, because i also don't have that data (they are missing).

[David Velasco]

In my case, data were collected every 6 s, so I looked for the gaps in the datetime vector (Date)...

lastSamp = find(diff(Date) > duration('00:00:07'));

...then set the last sample to nan in my Data:

Data(:,lastSamp) = nan;

[Maria-Eleni Gidarakou]

Thank you very much, it worked perfectly!