Swiss Camp snow height adjustments

[jasonebox]

It's a mess and like other sites, I think we'll get a better result if working from L0. That of course will be a lot of work. But in that effort, we can eliminate some of the noise that comes from sonic signal reflection from drift snow layer. So, I'll post that as a separate issue.

 if nicknames[0]=='SWC':
        t=datetime(1996, 1, 11)
        df['HS2'][t:]+=0.125
        t=datetime(1999, 1, 11)
        df['HS2'][t:]+=0.62
        t=datetime(2000, 12, 31,8)
        value=-0.75
        df['HS1'][t:]+=value
        df['HS2'][t:]+=value
        t=datetime(2001, 5, 12)
        value=-0.33
        df['HS1'][t:]+=value
        df['HS2'][t:]+=value
        t=datetime(2002, 1, 25)
        df['HS1'][t:]+=1.48
        df['HS2'][t:]+=0.35
        t0=datetime(2001, 10, 1) # filter upward spikes
        t1=datetime(2002, 4, 28)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>0]=np.nan
        df['HS2'][t0:t1][df['HS2'][t0:t1]>0]=np.nan
        t0=datetime(2002, 1, 1) # filter upward spikes
        t1=datetime(2002, 2, 28)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>-0.1]=np.nan
        df['HS2'][t0:t1][df['HS2'][t0:t1]>-0.1]=np.nan
        t0=datetime(2003, 1, 19) # filter short term downward shift
        t1=datetime(2003, 1, 24)
        df['HS1'][t0:t1]+=0.59
        t=datetime(2002, 11, 14) # remove strange shift
        df['HS1'][t:]-=0.12
        t=datetime(2003, 1, 25) # remove strange shift
        df['HS1'][t:]-=0.2
        t=datetime(2003, 4, 26) # remove strange shift
        df['HS1'][t:]+=0.17
        df['HS2'][t:]-=0.25
        t0=datetime(2003, 4, 24) # filter downward spikes
        t1=datetime(2003, 4, 28) 
        df['HS1'][t0:t1][df['HS1'][t0:t1]<-1.35]=np.nan
        df['HS2'][t0:t1][df['HS2'][t0:t1]<-1.45]=np.nan
        t=datetime(2004, 1, 21) 
        df['HS2'][t:]+=0.39
        t0=datetime(2008, 10, 30) # filter short term downward shift
        t1=datetime(2008, 12, 31)
        df['HS2'][t0:t1]-=0.5
        t=datetime(2009, 1, 1) 
        df['HS2'][t:]+=0.18
        t=datetime(2009, 5, 1) 
        df['HS1'][t:]-=0.5
        df['HS2'][t:]+=0.3
        t=datetime(2009, 5, 8) 
        df['HS1'][t:]+=0.6
        t0=datetime(2009, 1, 1) # filter downward spikes
        t1=datetime(2009, 6, 1)
        df['HS1'][t0:t1][df['HS1'][t0:t1]<-0.1]=np.nan
        t0=datetime(2009, 1, 1) # filter upward spikes
        t1=datetime(2009, 6, 1)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>0.5]=np.nan 
        t0=datetime(2009, 10, 1) # filter upward spikes
        t1=datetime(2010, 1, 1)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>0.5]=np.nan 
        t0=datetime(2011, 2, 18) # filter short term downward shift
        t1=datetime(2011, 5, 13)
        df['HS2'][t0:t1]-=2.95
        t=datetime(2011, 2, 18)
        value=-0.95
        df['HS1'][t:]+=value
        df['HS2'][t:]+=value
        t0=datetime(2011, 7, 25) # filter bad few data across large gap
        t1=datetime(2012, 5, 20)
        df['HS1'][t0:t1]=np.nan 
        t=datetime(2012, 5, 1) # reset hS 1,2
        value=-0.95
        df['HS1'][t:]+=value
        df['HS2'][t:]+=value
        t=datetime(2012, 5, 1) # reset hS 1,2
        df['HS1'][t:]+=0.2
        t=datetime(2014, 5, 10) # reset hS 1,2
        value=-0.4
        df['HS1'][t:]+=value
        df['HS2'][t:]+=value
        t0=datetime(2014, 5, 11) # filter short term shift
        t1=datetime(2014, 6, 1)
        df['HS1'][t0:t1]+=0.15
        t0=datetime(2013, 10, 1) # filter upward spikes
        t1=datetime(2014, 7, 1)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>-2.35]=np.nan 
        df['HS2'][t0:t1][df['HS2'][t0:t1]>-2.35]=np.nan 
        t0=datetime(2017, 4, 1) # filter upward spikes
        t1=datetime(2017, 9, 1)
        df['HS1'][t0:t1][df['HS1'][t0:t1]>-2.9]=np.nan 
        t0=datetime(2019, 4, 1) # filter upward spikes
        t1=datetime(2019, 7, 1)
        df['HS2'][t0:t1][df['HS2'][t0:t1]>-2.5]=np.nan 

before

after

[BaptisteVandecrux]

Partially solved in the last commit. Leaving open because field notes can help to constrain more annual ablation values.