<p>
        <span style="font-family:Verdana;font-size:16px;">Hi Dennis,</span>
</p>
<p>
        <br>
</p>
<p>
        <span style="font-family:Verdana;font-size:16px;">The abnormal values are still existent ( at different grid points) when a single year month data are used. </span><span style="font-family:Verdana;font-size:16px;">I have also tried the daily resolution, but it didn't work.</span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;">I could only adjust the calculation order. First calculate the monthly mean tropopause pressure , then set the outliers to missing_value. </span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;">Finally temporal average to yield the climatological mean tropopause pressure without these missing_value.</span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;"><br>
</span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;">Thanks for your reply.</span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;"><br>
</span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;">Best</span></span>
</p>
<p>
        <span style="font-family:Verdana;"><span style="font-size:16px;">Wang Na</span></span>
</p>
<p>
        <br>
</p>
<p>
        <span style="font-family:Verdana;font-size:16px;"><br>
</span>
</p>
<p>
        <br>
</p>
<p>
        <br>
</p>
<blockquote name="replyContent" class="ReferenceQuote" style="padding-left:5px;margin-left:5px;border-left:#b6b6b6 2px solid;margin-right:0;">
        -----原始邮件-----<br>
<b>发件人:</b><span id="rc_from">"Dennis Shea" <shea@ucar.edu></span><br>
<b>发送时间:</b><span id="rc_senttime">2017-09-21 08:25:07 (星期四)</span><br>
<b>收件人:</b> "王娜" <wangna@mail.iap.ac.cn><br>
<b>抄送:</b> "ncl-talk@ucar.edu" <ncl-talk@ucar.edu><br>
<b>主题:</b> Re: [ncl-talk] Question about the built-in function trop_wmo<br>
<br>
        <div dir="ltr">
                <div>
                        <div>
                                <div>
                                        <div>
                                                <div>
                                                        <div>
                                                                <div>
                                                                        <div>
                                                                                Source variable:  air(time, level, lat, lon)<br>
<br>
                                                                        </div>
Then you do<br>
<br>
tf   = addfile(dir + "<a href="http://air.mon.mean.nc">air.mon.mean.nc</a>", "r")<br>
temp = tf->air<br>
temp = temp + 273.15                               ; units conversion (from degC to K) <br>
ta   = dim_avg_n_Wrap(temp, 0)               ; compute climatology<br>
<br>
===<br>
                                                                </div>
Thie climatology is calculated over all months and year yielding<br>
<br>
                                                        </div>
  ta(level,lat,lon)<br>
<br>
===<br>
<br>
                                                </div>
I have *not* run the script or looked at the source  fortran code for the 'trop_wmo' function.<br>
<br>
                                        </div>
The temporal averaging over all years and months has smeared out the vertical profiles at each grid point. <br>
<br>
I *speculate* that the 'vertical change criterion' used by the function to detect the tropopause has been 'smoothed away'.<br>
<br>
                                </div>
You might try a single year month for testing. Keep in mind that the monthly averages are also averages.<br>
<br>
                        </div>
<br>
                </div>
Good luck<br>
                <div>
                        <div>
                                <div>
                                        <div>
                                                <br>
                                        </div>
                                </div>
                        </div>
                </div>
        </div>
        <div class="gmail_extra">
                <br>
                <div class="gmail_quote">
                        On Wed, Sep 20, 2017 at 8:15 AM, 王娜 <span dir="ltr"><<a href="mailto:wangna@mail.iap.ac.cn" target="_blank">wangna@mail.iap.ac.cn</a>></span> wrote:<br>
                        <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
                                <p>
                                        <span style="font-size:14px;font-family:Verdana;">Hi all, </span> 
                                </p>
                                <p>
                                        <br>
                                </p>
                                <p>
                                        <span style="font-size:14px;font-family:Verdana;">I am trying to calculate the tropopause pressure with the built-in function trop_wmo, but got some abnormal values (equal to 85 hPa or -999 hPa) when using the </span><span style="font-family:Verdana;font-size:14px;">monthly temperature from NCEP/NCAR reanalysis 1 (17 levels between 1000 hPa and 10 hPa)</span><span style="font-size:14px;font-family:Verdana;">.  </span><span style="font-family:Verdana;font-size:14px;">It seems like that this problem is related to the vertical resolution of the input temperature. If I use the ERA-interim reanalysis data (32 levels between 1000 hPa and 10 hPa), the abnormal values are non-existent.  Does the function trop_wmo not apply to the renanlysis data with coarse resolution? Any help on how to fix this problem? </span> 
                                </p>
                                <p>
                                        <span style="font-family:Verdana;font-size:14px;">Attached are my ncl script and input data.</span> 
                                </p>
                                <p>
                                        <span style="font-family:Verdana;font-size:14px;"><br>
</span> 
                                </p>
                                <p>
                                        <span style="font-family:Verdana;font-size:14px;">Thanks</span> 
                                </p>
                                <p>
                                        <span style="font-family:Verdana;font-size:14px;">Wang Na</span> 
                                </p>
<br>
<br>
                                <div id="m_2895785064124983044largeAttachmentsBox" style="padding:4px;margin-bottom:15px;background-color:#e3e3e3;width:auto;font-family:Verdana,Arial,Tahoma;font-size:14px;">
                                        <hr style="display:none;">
                                        <div style="padding:6px 0pt 10px 6px;text-align:left;">
                                                <b style="font-size:14px;">从<a href="http://mail.iap.ac.cn" target="_blank">mail.iap.ac.cn</a>发来的超大附件</b>
                                        </div>
                                        <div style="padding:0pt 8px 6px 12px;background:#FFFFFF none repeat scroll 0% 0%;">
                                                <div style="clear:both;">
                                                        <div style="clear:both;">
                                                                <div style="padding:10px 0pt;font-size:12px;">
                                                                        <div style="float:left;margin-right:8px;margin-top:2px;">
                                                                                <b class="m_2895785064124983044bico m_2895785064124983044bfMSWORD"></b>
                                                                        </div>
                                                                        <div class="m_2895785064124983044bigatt_bt" title="">
                                                                                <div class="m_2895785064124983044name_big">
                                                                                        <span><a href="https://mail.cstnet.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIu3sKBT90UTs0mTbKm3VUL3srL3Zt1Sn2LjkKCTyCu3sKBT98dantETs0mTbKm3VUL3srL3Ztdan7EryDuonECzcqpS4CWSeIkasKAwZEXaWCm3VUL3srL3ZtMfeYpzsFWSnKvSbKyTgk8Ss2maUrUUUUrjnCgTuDn0ka6r47ur4kAr47Xw17KFXF6SZ0LaZt1Tsvk1q3Ur7jAUn5U-7jjUefUI7jOU88Ur7jlUekUrJjYUePU2DjOUefU77jjUerUID0mUK1BOg3=&code=4enzgir3" style="color:#000000;text-decoration:none;" target="_blank">air.mon.mean.nc</a> <span style="color:#A0A0A0;"> (305MB, 2017年10月05日 22:15 到期)</span></span>
                                                                                        <div class="m_2895785064124983044down_big">
                                                                                                <a href="https://mail.cstnet.cn/coremail/viewDownloadFile.jsp?key=1U31SsvkjqIu3sKBT90UTs0mTbKm3VUL3srL3Zt1Sn2LjkKCTyCu3sKBT98dantETs0mTbKm3VUL3srL3Ztdan7EryDuonECzcqpS4CWSeIkasKAwZEXaWCm3VUL3srL3ZtMfeYpzsFWSnKvSbKyTgk8Ss2maUrUUUUrjnCgTuDn0ka6r47ur4kAr47Xw17KFXF6SZ0LaZt1Tsvk1q3Ur7jAUn5U-7jjUefUI7jOU88Ur7jlUekUrJjYUePU2DjOUefU77jjUerUID0mUK1BOg3=&code=4enzgir3" target="_blank"><span style="color:#5d5d5d;text-decoration:underline;">查看下载信息</span></a> 
                                                                                        </div>
                                                                                </div>
                                                                        </div>
                                                                </div>
                                                        </div>
                                                </div>
                                        </div>
                                </div>
<br>
______________________________<wbr>_________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/<wbr>mailman/listinfo/ncl-talk</wbr></a><br>
<br>
                        </wbr></blockquote>
                </div>
<br>
        </div>
</blockquote>