scePrecip/conv_identification_algo.R at main · Jbrich95/scePrecip · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
# Applies the convection identified algorithm described in Section 2.1 of Richards et al. (2022b)

#set hyperparameters
g_u = 1.0 # (mm/h) upper threshold
g_l = 0.01 # (mm/h) lower threshold
n_g = 9 # size of neighbourhood
p.star = 0.2 #proportion threshold

lon=lonlat.grid[,,1]
lat=lonlat.grid[,,2]
nlons=dim(Data.grid)[2]
nlats=dim(Data.grid)[1]

conv.precip=array(NA,dim(Data.grid))

# set gradhigh to 1 if high spatial gradient
# set gradlow to 1 if low or high (but non-zero) spatial gradient

for(t in 1:dim(Data.grid)[3]){
  gradlow <- gradhigh <- array(0,dim=c(nlats+1,nlons+1))
  gradlow[nlats+1,nlons+1]=NA
  gradhigh[nlats+1,nlons+1]=NA
  tempDat=Data.grid[,,t]
  for(i in 1:(nlats-1)){
    for(j in 1:nlons){
      if(!is.na(tempDat[i+1,j]) & !is.na(tempDat[i,j])){
        if( tempDat[i+1,j] > tempDat[i,j]+g_u){
          gradhigh[i,j]=1.0
          gradhigh[i+1,j]=1.0
        }

        if(tempDat[i+1,j] > tempDat[i,j]+g_l){
          gradlow[i,j]=1.0
          gradlow[i+1,j]=1.0
        }


      }
    }


  }

  for(i in 2:nlats){
    for(j in 1:nlons){
      if(!is.na(tempDat[i-1,j])&!is.na(tempDat[i,j])){
        if( tempDat[i-1,j] > tempDat[i,j]+g_u){
          gradhigh[i,j]=1.0
          gradhigh[i-1,j]=1.0}

        if(tempDat[i-1,j] > tempDat[i,j]+g_l){
          gradlow[i,j]=1.0
          gradlow[i-1,j]=1.0
        }
      }
    }
  }


  for(i in 1:nlats){
    for(j in 1:(nlons-1)){
      if(!is.na(tempDat[i,j+1])&!is.na(tempDat[i,j])){
        if( tempDat[i,j+1] > tempDat[i,j]+g_u){
          gradhigh[i,j]=1.0
          gradhigh[i,j+1]=1.0}

        if(tempDat[i,j+1] > tempDat[i,j]+g_l){
          gradlow[i,j]=1.0
          gradlow[i,j+1]=1.0
        }
      }
    }
  }
  for(i in 1:nlats){
    for(j in 2:nlons){
      if(!is.na(tempDat[i,j-1])&!is.na(tempDat[i,j])){
        if( tempDat[i,j-1] > tempDat[i,j]+g_u){
          gradhigh[i,j]=1.0
          gradhigh[i,j-1]=1.0}

        if(tempDat[i,j-1] > tempDat[i,j]+g_l){
          gradlow[i,j]=1.0
          gradlow[i,j-1]=1.0
        }

      }
    }}


  nlen = (n_g-1)/2
  for(i in (nlen+1):(nlats-nlen)){
    for(j in (nlen+1):(nlons-nlen)){

      ghdata = gradhigh[(i-nlen):(i+nlen),(j-nlen):(j+nlen)]
      gldata = gradlow[(i-nlen):(i+nlen),(j-nlen):(j+nlen)]
      if(sum(gldata) >= 0.5){
        nfrac = sum(ghdata)/sum(gldata)
        if(nfrac >= p.star){
          conv.precip[i,j,t] = Data.grid[i,j,t]
        }
      }
      if(any(is.na(lon[(i-nlen):(i+nlen),(j-nlen):(j+nlen)]))){
        conv.precip[i,j,t]=NA
      }

      if( sum(ghdata) > sum(gldata)){
        print("ERROR in totals")

      }
    }
  }

  if(t%%100==0){print(t)}
}


conv.precip=conv.precip[(nlen+1):(nlats-nlen),(nlen+1):(nlons-nlen),] #Remove edges
lat=lat[(nlen+1):(nlats-nlen),(nlen+1):(nlons-nlen)]
lon=lon[(nlen+1):(nlats-nlen),(nlen+1):(nlons-nlen)]


#Re-shape to matrices
tmp<-matrix(ncol=prod(dim(conv.precip)[1:2]),nrow=dim(conv.precip)[3])
tmp2<-matrix(nrow=prod(dim(lat)[1:2]),ncol=2)
int<-1
for(i in 1:dim(conv.precip)[1]){
  for(j in 1:dim(conv.precip)[2]){
    tmp[,int]=conv.precip[i,j,]
    tmp2[int,1]=lon[i,j]; tmp2[int,2]=lat[i,j]
    int=int+1
  }
}

coords=tmp2

# Note that tmp currently includes only data identified as convective, with NAs otherwise.
# We take the whole observed field as convective if any convective values are found within a field

conv.t.inds<-which(apply(tmp,1,sum,na.rm=T)>0)

Data<-tmp[conv.t.inds,]
save(Data,coords,file="Data/conv.Rdata")


Data<-tmp[-conv.t.inds,]
save(Data,coords,file="Data/nonconv.Rdata")