<div dir="ltr">Kyle, I agree. But I think Matt wants a grid of time indices, not just the maximum values. Use dim_max_n to begin, then more work to get the indices.<div><br></div><div><div> dims_3d = dimsizes (data) ; (time,lat,lon)</div><div> ntimes = dims_3d(0)</div><div> max_2d = dim_max_n (data, 0) ; grid of max point values (lat,lon)</div><div> max_3d = conform_dims (dims_3d, max_2d, (/1,2/)) ; expand to all time steps</div><div> mask_3d = (data .eq. max_3d) ; true at all values equal to point maximums</div><div> <br></div><div> inds_1d = ispan (0, ntimes-1, 1) ; make data array of time indices</div><div> inds_3d = conform_dims (dims_3d, inds_1d, 0) ; expand to all lat,lon</div><div> inds_3d@_FillValue = -999</div><div> inds_masked = where (mask_3d, inds_3d, inds_3d@_FillValue)</div><div> ; time indices at all values equal to point maximums</div><div><br></div><div>; There may be two or more identical maximum values for any grid point.</div><div>; Just extract a single time index at each grid point.</div><div>; Grid points with no data are set to missing (_FillValue).</div><div><br></div><div> inds_max_2d = dim_min_n (inds_masked, 0) ; FIRST OCCURRENCE IN TIME</div><div>; inds_max_2d = dim_max_n (inds_masked, 0) ; LAST OCCURRENCE IN TIME</div></div><div><br></div><div>Also, see attached working example, which plots the results.</div><div><br></div><div>--Dave</div><div><br></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Sep 9, 2015 at 9:22 PM, Kyle Griffin <span dir="ltr"><<a href="mailto:ksgriffin2@wisc.edu" target="_blank">ksgriffin2@wisc.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Hi Matt,<div><br></div><div>I think dim_max_n is what you're looking for.</div><div><br></div><div><a href="https://www.ncl.ucar.edu/Document/Functions/Built-in/dim_max_n.shtml" target="_blank">https://www.ncl.ucar.edu/Document/Functions/Built-in/dim_max_n.shtml</a><br></div><div><br></div><div>Not sure if I've used it in this manner before, but if you specify the dimension to find the max over (2nd argument) as 0, I think you should get what you're looking for. As a built-in function, it should be faster than looping over each point.</div><div><br></div><div><br></div><div>Kyle</div></div><div class="gmail_extra"><br clear="all"><div><div><div dir="ltr"><div><div dir="ltr">----------------------------------------<div>Kyle S. Griffin</div><div>Department of Atmospheric and Oceanic Sciences</div><div>University of Wisconsin - Madison</div><div>Room 1407</div><div>1225 W Dayton St, Madison, WI 53706</div><div>Email: <a href="mailto:ksgriffin2@wisc.edu" target="_blank">ksgriffin2@wisc.edu</a></div></div></div></div></div></div>
<br><div class="gmail_quote">On Wed, Sep 9, 2015 at 9:08 PM, Matthew Fearon <span dir="ltr"><<a href="mailto:Matthew.Fearon@dri.edu" target="_blank">Matthew.Fearon@dri.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
<div style="direction:ltr;font-family:Tahoma;color:#000000;font-size:10pt">Dear NCL Users,
<div><br>
</div>
<div>Trying to get the index of the maximum value on a single dimension as part of a multidimensional array. The array is data(time, lat, lon) where nlat x nlon is 400x500 and ntime=24. Would like the maximum over the time dimension for each grid cell or lat/long.
Has anymore performed this without a loop over nlat/nlon ?</div>
<div><br>
</div>
<div>Making the array 1D (and using ind_resolve) allows me to get the max inds per dimension over the whole array, which isn't what I need. </div>
<div><br>
</div>
<div>Any suggestions would be greatly appreciated. </div>
<div>thanks,</div>
<div>Matt</div></div></div></blockquote></div></div></blockquote></div></div></div>