“I calculated the RMM index using ERA5 data (U200/850 + TTR) strictly following the official procedure, but the results are inconsistent with the ‘MJO_2025’ data downloaded from the official website. Could you please advise on the possible reasons for this discrepancy? (It would be highly appreciated if you could also share the official code for the RMM calculation). I would be very grateful if you could share it!”!"
Hello Hailin, thank you for sharing the details of the discrepancy. Joshua is currently on leave, but he will respond upon his return on Monday, 3 August.
Dear Hailin,
Thanks for your message.
Could you please be more specific regarding what differences you are showing?
I re-checked my code and noted I was using a climatology from 1979 to 2001 (this was incorrect). I’ve now updated the data based on the new climatological period. Please check the differences now.
Thanks,
Josh
Dear Joshua,
Thank you for your reply. I would like to provide some additional details regarding my question.
I used the U200, U850, and daily mean TTR data from ERA5 (1979–2026) and followed the MJO index calculation method outlined on the official website [as shown in the figure below, where the climatology is calculated for the period of 1979–2025]. However, the MJO index I calculated does not match the “MJO_DAILY” data provided on the website.
For example, the figure below illustrates the MJO index for July 15, 2025. The first row shows my calculated results, while the second and third rows show the modified and unmodified data from the website, respectively. There is a rather obvious discrepancy, which will impact my subsequent research.
I have attached my MJO calculation code to this email. Could you please take a look and help analyze the potential reasons for this difference? Additionally, I would be extremely grateful if you could share your code for calculating the MJO index, along with any specific data requirements.
Best regards,
Hai Lin
“It appears I cannot attach the code directly here. Please let me know the best way to share it with you.”
Could you drop me an email. Joshua.talib@ecmwf.int
Thank you.I have sent it to you using a different email address (liuh68847@gmail.com). Please kindly check your inbox.
“Each computed PC index is normalised by observed standard deviations of RMM1 and RMM2. Combined EOFs used for MJO projection can be downloaded using the retrieve_MJO_projection_data function.”
“As seen above, the official MJO calculation procedure is somewhat ambiguous regarding the projection step. Could you please share the specific details of how this projection is handled? I would be extremely grateful.”
