ecjoliver / marineheatwaves Goto Github PK

Hello,

There is a small bug when a MHW finishes at end of time series and when its Peak is also at end of time series. There is a test at line 408 to assume decline time equals 1 day :
if tt_peak == T-1:
but it should be
if tt_start + tt_peak == T-1:

Sébastien

Hello!

First, thanks for providing your algorithms as open source.
This is extremely helpful.

Then, in the process of actual use, we encountered the following problem. Is there a better solution.

Any suggestions would be appreciated.

File "/icar/ljj_hujl/marineHeatWaves-master/marineHeatWaves.py", line 388, in detect
mhw['category'].append(categories[np.min([cats[tt_peakCat], 4]).astype(int) - 1])
~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
IndexError: index -5 is out of bounds for axis 0 with size 4

first of all, thank you so much for making this code available on both python and R. I had a query regarding threshold calculation while creating a daily climatology using a user-specified sliding window. As suggested in Hobday et al. (2016) that the users should be cognisant of the biases that might be introduced at the start and end of the base period when calculating threshold exceedances. They also suggested using a bootstrapping procedure such as defined by Zhang et al. (2005) to calculate percentile from subsets of the data.

First, thanks for providing your algorithms as open source.
This is extremely helpful.

Second, are there any key algorithmic differences between this code base and
the R version (aside from the obvious syntax)? I am not keen on learning R at
this stage but wonder if the R version has advanced functionality
and/or bug fixes.

Any guidance you can provide would be most appreciated.

Hi guys,
I am interesting in applying the fantastic Heatwaver package in R for temperature data (+30 years) collected in freshwater systems in western Europe but was hoping to separate and highlight temperature extremities in both the summer and winter months - which would require different threshold metrics. Is this possible with heatwaveR?
Any suggestion would be most appreciated.

Hi,
Thank you for the incredibly helpful scripts. However, I'm following the Net.cdf creation from MHW results script and I need some help.
The shell sets up nicely; however, only the information from the first lon lat is read into the Net cdf document.
e.g.
, , lat = 41.0092277526855

    lon

event_no 1.04260063171387 1.09264658918284 1.14269254665182 1.1927385041208 1.24278446158977 1.29283041905875
1 2.0685 NA NA NA NA NA
2 2.1377 NA NA NA NA NA
3 2.5330 NA NA NA NA NA
4 1.9798 NA NA NA NA NA
5 2.2583 NA NA NA NA NA
6 1.7431 NA NA NA NA NA

I haven't changed any of the script so I'm not to sure why this is happening.