
def stix_hek_data(start_time=None,end_time=None,lighttime_correct=True,local_copy=False):
    '''Get STIX flare list for given time period. Search HEK/AIA database for flares occuring within +- 5 minutes from specified time. Perform coordinate transformation. 
    To be implemented: Reverse joint visibility calculation (is STIX flare visible from Earth) once CFL locations are returned by stixdcpy'''
    if not start_time:
        start_time=dt(2020,2,10,5,0,0)
    if not end_time:
        end_time=dt.now()
    load_SOLO_SPICE(dt.now()) #SPICE kernels for Solar Orbiter trajectory
    stix_flare_list=get_stix_flares(start_time,end_time,sortedby='time',lighttime_correct=True)
    stix_flare_list['peak_utc']=pd.to_datetime(stix_flare_list.peak_UTC)
    #stix_flare_list['peak_utc']=[d.replace(microsecond=0) for d in stix_flare_list.peak_utc]
    nflares=stix_flare_list.peak_utc.count()
    log.info(f"{nflares} STIX flares found")
    if nflares == 1000:
        log.warning(f"{nflares} STIX flares found, there might be more results in the specified date range!")
    #is there a HEK event at same time?
    hek_events=[hek.HEKEventHandler(f.peak_utc_corrected).df for _,f in stix_flare_list.iterrows()]
    log.info(f"{len(hek_events)} HEK results found")
    hdf=pd.concat(hek_events)
    hdf['event_peaktime']=pd.to_datetime(hdf.event_peaktime)
    #merge STIX results with HEK results
    df=stix_flare_list.merge(hdf,left_on='peak_utc_corrected',right_on='date_obs',how='outer')
    return df

def get_stix_flares(start_time,end_time,sortedby='time',lighttime_correct=True,local_copy=True):
    '''Query STIX data center for flare list. Perform light travel time correction.'''
    flares=jreq.fetch_flare_list(start_time,end_time,sortedby=sortedby) #current limit is 1000, results are returned backwards from end time
    df=pd.DataFrame(flares)
    corrected_times,solo_r=[],[]
    if lighttime_correct:
        for d in df.peak_UTC:
            ctime,solor=spacecraft_to_earth_time(d,solo_r=True)
            corrected_times.append(ctime)
            solo_r.append(solor)
        df['peak_utc_corrected']=corrected_times
        df['solo_r']=solo_r
        df['peak_counts_corrected']=df.peak_counts*(df.solo_r)**2 #correct to counts @ 1AU
    if local_copy:
        df.to_json(f"/home/erica/STIX/data/stix_flares_{dt.strftime(dt.now(),'%Y-%m-%dT%H:%M')}.json")
    return df


def update_stix_hek_data(current_data_file='/home/erica/STIX/data/stix_hek_full.json'):
    '''update the csv file with the latest flares from STIX and AIA'''
    df=pd.read_json(current_data_file)
    df['peak_utc']=pd.to_datetime(df.peak_utc,unit='ms')
    last_date=df.peak_utc.iloc[0] #newest events are added at the beginning
    newdf=stix_hek_data(start_time=last_date)
    udf=pd.concat([newdf,df])
    udf.drop_duplicates(subset=[k for k in udf.keys() if k !='goes'],inplace=True)
    udf.reset_index(inplace=True,drop=True)
    udf.to_json(current_data_file)


df.describe()


tplus=td(minutes=10)
dfs=df.query("visible_from_SOLO==True and STIX_AIA_timedelta_abs < @tplus").dropna(how='all')
#dfs=df.query("visible_from_SOLO==True and STIX_AIA_timedelta_abs < @tplus and r_less_than_50_arcsec").dropna(how='all')


dfgc=dfs.where(dfs.fl_goescls !='').dropna(how='all')
dfgc['GOES_flux']=[goes_class_to_flux(c) for c in dfgc.fl_goescls]


fit=STIX_GOES_fit(dfgc)
fit.do_fit(dfgc)
fit.bin_residuals(dfgc) #gives statistics-based boundaries on predictions - very important!
fit.__dict__

{'start_date': Timestamp('2021-04-24 14:54:07.893000'),
 'end_date': Timestamp('2022-02-02 16:35:08.374000'),
 'nflares': 1165,
 'timestamp': datetime.datetime(2022, 2, 4, 10, 28, 36, 299416),
 'min_counts': 275.8,
 'bins': [2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0],
 'slope': 0.6474373955427877,
 'intercept': -8.017059728369057,
 'rvalue': 0.8404334190665731,
 'pvalue': 1.0143336224873e-311,
 'bin_2.0_sigma': 0.24294403493347283,
 'bin_2.5_sigma': 0.28223799401465166,
 'bin_3.0_sigma': 0.17811255923183106,
 'bin_3.5_sigma': 0.2518162616056952,
 'bin_4.0_sigma': 0.14362285052090484,
 'bin_4.5_sigma': 0.2454768948923288,
 'bin_5.0_sigma': 0.4029749010665078}

	flare_id	total_signal_counts	total_counts	duration	LC0_BKG	_id	GOES_flux	CFL_X_arcsec	CFL_Y_arcsec	peak_counts	...	rotated_y_arcsec	rotated_lon_deg	rotated_lat_deg	rotated_x_px	rotated_y_px	visible_from_SOLO	start_unix	end_unix	STIX_AIA_timedelta	STIX_AIA_timedelta_abs
count	3.423000e+03	3.423000e+03	4.720000e+02	3423.000000	472.000000	3423.000000	4.720000e+02	0.0	0.0	3423.000000	...	1988.000000	1988.000000	1988.000000	1988.000000	1988.000000	2023.000000	2.951000e+03	2.951000e+03	2023	2023
mean	2.125082e+09	4.735241e+05	5.950362e+05	857.280748	279.135593	2305.458662	1.362252e-06	NaN	NaN	3845.376629	...	-93.811257	14.986685	-6.209649	2273.108068	1953.687093	0.867523	1.632102e+09	1.632102e+09	0 days 00:05:20.429510133	0 days 00:20:13.049021749
std	3.496230e+07	5.462422e+06	3.765789e+06	1062.892194	53.964668	991.701731	2.980354e-06	NaN	NaN	27491.654699	...	462.247365	58.579077	34.129521	772.867709	462.252368	0.339092	6.282455e+06	6.282389e+06	0 days 01:15:44.329653006	0 days 01:13:11.063900368
min	2.104240e+09	6.400000e+01	7.929000e+03	80.000000	247.000000	589.000000	0.000000e+00	NaN	NaN	262.466667	...	-1093.814195	-87.407210	-89.177735	836.053013	953.675551	0.000000	1.619226e+09	1.619228e+09	-1 days +08:11:16.071000	0 days 00:00:00.286000
25%	2.108201e+09	4.384000e+03	3.555250e+04	348.000000	247.000000	1444.500000	5.519000e-07	NaN	NaN	337.133333	...	-495.312198	-40.325495	-37.766042	1570.362438	1552.186403	1.000000	1.627829e+09	1.627829e+09	0 days 00:00:50.007500	0 days 00:02:38.810500
50%	2.110041e+09	1.126400e+04	6.639950e+04	584.000000	271.000000	2307.000000	8.603500e-07	NaN	NaN	441.666667	...	-293.295623	22.147903	-25.045989	2354.270866	1754.201461	1.000000	1.632336e+09	1.632337e+09	0 days 00:04:18.543000	0 days 00:05:43.435000
75%	2.112212e+09	4.230800e+04	1.402060e+05	1030.000000	271.000000	3162.500000	1.220925e-06	NaN	NaN	892.866667	...	344.011934	73.784082	28.810110	2930.122439	2391.515049	1.000000	1.636705e+09	1.636708e+09	0 days 00:07:07.060500	0 days 00:08:18.781000
max	2.202022e+09	2.089835e+08	5.447654e+07	35680.000000	463.000000	4028.000000	5.508880e-05	NaN	NaN	753663.000000	...	822.923871	89.111706	54.493623	3613.201130	2870.438193	1.000000	1.642118e+09	1.642119e+09	0 days 14:13:33.318000	0 days 15:48:43.929000

log10 (counts)	1 $\sigma$	N flares
< 3	0.272	830
[3,4)	0.205	264
> 4	0.222	71

Finding a GOES class proxy for STIX¶

Problem description¶

Instrument details¶

Workflow:¶

Data collection & preparation¶

Putting it all together to update results¶

Flare property distributions¶

Total Flares: 3423¶

Jointly Visible: 1498¶

Jointly Visible, with GOES class: 1165¶

Flare Locations¶

Counts- flux relationship¶

Linear fit to log-log data (least squares)¶

Residuals¶

Flares not jointly visible - calculate GOES flux¶