-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPropConstantCalcsForOptimizer.m
181 lines (161 loc) · 9.02 KB
/
PropConstantCalcsForOptimizer.m
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
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
%% PROPORTIONALITY CONSTANT CALCULATIONS
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Evolutionary model for coral cover (from Baskett et al. 2009) %
% modified by Cheryl Logan ([email protected]) %
% last updated: 8/26/16 %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
[basePath, outputPath, sstPath, SGPath, matPath, n, defaultThreads] = useComputer(3);
% SST DATASET?
Data = 1; % 1=ESM2M_norm
%Data = 2; % 2=HADISST (through 3/16/16)
if Data == 1
dataset = 'ESM2M';
else
dataset = 'HadISST';
end
%% DEFINE CLIMATE CHANGE SCENARIO (from normalized GFDL-ESM2M; J Dunne)
%for i={'rcp26','rcp85'}
%RCP = char(i)
% RCP = 'rcp85'; % options; 'rcp26', 'rcp45', 'rcp60', 'rcp85'
format shortg; c = clock; date = strcat(num2str(c(1)),num2str(c(2)),num2str(c(3))); % today's date stamp
%% LOAD JOHN'S NORMALIZED SSTS FROM EARTH SYSTEM CLIMATE MODEL OR HADISST
%GetSST_norm_GFDL_ESM2M % sub m-file for extracting SSTs for a ALL reef grid cells
% [SST, Reefs_latlon, TIME, startYear] = GetSST_norm_GFDL_ESM2M(sstPath, matPath, Data, RCP);
% matPath is no longer used, 12/2019.
[SST, Reefs_latlon, TIME, startYear] = GetSST_norm_GFDL_ESM2M(sstPath, dataset, RCP);
SST_1861_2000 = SST(:,1:1680);
% never used? SSThist = SST_1861_2000;
%% LOAD OLD SELECTIONAL VARIANCE (psw2)
%load ('~/Dropbox/Matlab/SymbiontGenetics/mat_files/psw2_trials.mat','psw2_var_allv2')
%% Store optimizer inputs from propInputValues with any constant values to be computed.
% expand the array once, including space for averages:
pswInputs(4, 65+ceil(65/4)) = 0;
pswInputs(:,1) = propInputValues';
%% Update 1/13/2020, values for targets 3, 5 and 10% bleaching.
% Now all cases are in sets which include all 4 RCP values, so averaging can be
% done automatically.
% Be sure to skip pswInputs(:, 1) because "1" use used for optimization.
% == Target 3, no advantage ==
pswInputs(:, 2) = [0.025; 1.5; 0.45; 3.1481]; % RCP 2.6, E=0, OA=0
pswInputs(:, 3) = [0.025; 1.5; 0.45; 3.4444]; % RCP 4.5, E=0, OA=0
pswInputs(:, 4) = [0.025; 1.5; 0.45; 3.0972]; % RCP 6, E=0, OA=0
pswInputs(:, 5) = [0.025; 1.5; 0.45; 3.1435]; % RCP 8.5, E=0, OA=0
pswInputs(:, 6) = [0.025; 1.5; 0.45; 3.7685]; % RCP 2.6, E=1, OA=0
pswInputs(:, 7) = [0.025; 1.5; 0.45; 4.0278]; % RCP 4.5, E=1, OA=0
pswInputs(:, 8) = [0.025; 1.5; 0.45; 3.7731]; % RCP 6, E=1, OA=0
pswInputs(:, 9) = [0.025; 1.5; 0.45; 3.6944]; % RCP 8.5, E=1, OA=0
pswInputs(:, 10) = [0.025; 1.5; 0.45; 3.7407]; % RCP 2.6, E=1, OA=1
pswInputs(:, 11) = [0.025; 1.5; 0.45; 4.0046]; % RCP 4.5, E=1, OA=1
pswInputs(:, 12) = [0.025; 1.5; 0.45; 3.7407]; % RCP 6, E=1, OA=1
pswInputs(:, 13) = [0.025; 1.5; 0.45; 3.6713]; % RCP 8.5, E=1, OA=1
% == Target 5, no advantage ==
pswInputs(:, 14) = [0.025; 1.5; 0.45; 4.3426]; % RCP 2.6, E=0, OA=0
pswInputs(:, 15) = [0.025; 1.5; 0.45; 4.4745]; % RCP 4.5, E=0, OA=0
pswInputs(:, 16) = [0.025; 1.5; 0.45; 4.3519]; % RCP 6, E=0, OA=0
pswInputs(:, 17) = [0.025; 1.5; 0.45; 4.339]; % RCP 8.5, E=0, OA=0
pswInputs(:, 18) = [0.025; 1.5; 0.45; 4.6806]; % RCP 2.6, E=1, OA=0
pswInputs(:, 19) = [0.025; 1.5; 0.45; 4.8056]; % RCP 4.5, E=1, OA=0
pswInputs(:, 20) = [0.025; 1.5; 0.45; 4.6898]; % RCP 6, E=1, OA=0
pswInputs(:, 21) = [0.025; 1.5; 0.45; 4.6574]; % RCP 8.5, E=1, OA=0
pswInputs(:, 22) = [0.025; 1.5; 0.45; 4.6528]; % RCP 2.6, E=1, OA=1
pswInputs(:, 23) = [0.025; 1.5; 0.45; 4.8009]; % RCP 4.5, E=1, OA=1
pswInputs(:, 24) = [0.025; 1.5; 0.45; 4.6574]; % RCP 6, E=1, OA=1
pswInputs(:, 25) = [0.025; 1.5; 0.45; 4.6389]; % RCP 8.5, E=1, OA=1
% == Target 5, advantage 1, penalty 0.25 == (only these 4 are at 0.25)
pswInputs(:, 26) = [0.025; 1.5; 0.45; 5.5324]; % RCP 2.6, E=0, OA=0
pswInputs(:, 27) = [0.025; 1.5; 0.45; 5.6667]; % RCP 4.5, E=0, OA=0
pswInputs(:, 28) = [0.025; 1.5; 0.45; 5.5417]; % RCP 6, E=0, OA=0
pswInputs(:, 29) = [0.025; 1.5; 0.45; 5.5417]; % RCP 8.5, E=0, OA=0
% == Target 5, advantage 0.5 ==
pswInputs(:, 30) = [0.025; 1.5; 0.45; 5.4259]; % RCP 2.6, E=0, OA=0
pswInputs(:, 31) = [0.025; 1.5; 0.45; 5.6111]; % RCP 4.5, E=0, OA=0
pswInputs(:, 32) = [0.025; 1.5; 0.45; 5.3009]; % RCP 6, E=0, OA=0
pswInputs(:, 33) = [0.025; 1.5; 0.45; 5.3056]; % RCP 8.5, E=0, OA=0
pswInputs(:, 34) = [0.025; 1.5; 0.45; 5.6574]; % RCP 2.6, E=1, OA=0
pswInputs(:, 35) = [0.025; 1.5; 0.45; 5.8102]; % RCP 4.5, E=1, OA=0
pswInputs(:, 36) = [0.025; 1.5; 0.45; 5.5231]; % RCP 6, E=1, OA=0
pswInputs(:, 37) = [0.025; 1.5; 0.45; 5.5231]; % RCP 8.5, E=1, OA=0
% == Target 5, advantage 1 ==
pswInputs(:, 38) = [0.025; 1.5; 0.45; 5.5509]; % RCP 2.6, E=0, OA=0
pswInputs(:, 39) = [0.025; 1.5; 0.45; 5.6713]; % RCP 4.5, E=0, OA=0
pswInputs(:, 40) = [0.025; 1.5; 0.45; 5.5417]; % RCP 6, E=0, OA=0
pswInputs(:, 41) = [0.025; 1.5; 0.45; 5.537]; % RCP 8.5, E=0, OA=0
pswInputs(:, 42) = [0.025; 1.5; 0.45; 5.8056]; % RCP 2.6, E=1, OA=0
pswInputs(:, 43) = [0.025; 1.5; 0.45; 5.8287]; % RCP 4.5, E=1, OA=0
pswInputs(:, 44) = [0.025; 1.5; 0.45; 5.7454]; % RCP 6, E=1, OA=0
pswInputs(:, 45) = [0.025; 1.5; 0.45; 5.6991]; % RCP 8.5, E=1, OA=0
% == Target 5, advantage 1.5 ==
pswInputs(:, 46) = [0.025; 1.5; 0.45; 5.3843]; % RCP 2.6, E=0, OA=0
pswInputs(:, 47) = [0.025; 1.5; 0.45; 5.3843]; % RCP 4.5, E=0, OA=0
pswInputs(:, 48) = [0.025; 1.5; 0.45; 5.3843]; % RCP 6, E=0, OA=0
pswInputs(:, 49) = [0.025; 1.5; 0.45; 5.2963]; % RCP 8.5, E=0, OA=0
pswInputs(:, 50) = [0.025; 1.5; 0.45; 5.6343]; % RCP 2.6, E=1, OA=0
pswInputs(:, 51) = [0.025; 1.5; 0.45; 5.6204]; % RCP 4.5, E=1, OA=0
pswInputs(:, 52) = [0.025; 1.5; 0.45; 5.6389]; % RCP 6, E=1, OA=0
pswInputs(:, 53) = [0.025; 1.5; 0.45; 5.4861]; % RCP 8.5, E=1, OA=0
% == Target 10, advantage 0 ==
pswInputs(:, 54) = [0.025; 1.5; 0.45; 5.8935]; % RCP 2.6, E=0, OA=0
pswInputs(:, 55) = [0.025; 1.5; 0.45; 5.9537]; % RCP 4.5, E=0, OA=0
pswInputs(:, 56) = [0.025; 1.5; 0.45; 5.8519]; % RCP 6, E=0, OA=0
pswInputs(:, 57) = [0.025; 1.5; 0.45; 5.8426]; % RCP 8.5, E=0, OA=0
pswInputs(:, 58) = [0.025; 1.5; 0.45; 6.3843]; % RCP 2.6, E=1, OA=0
pswInputs(:, 59) = [0.025; 1.5; 0.45; 6.4398]; % RCP 4.5, E=1, OA=0
pswInputs(:, 60) = [0.025; 1.5; 0.45; 6.3472]; % RCP 6, E=1, OA=0
pswInputs(:, 61) = [0.025; 1.5; 0.45; 6.2685]; % RCP 8.5, E=1, OA=0
pswInputs(:, 62) = [0.025; 1.5; 0.45; 6.4213]; % RCP 2.6, E=1, OA=1
pswInputs(:, 63) = [0.025; 1.5; 0.45; 6.4676]; % RCP 4.5, E=1, OA=1
pswInputs(:, 64) = [0.025; 1.5; 0.45; 6.3935]; % RCP 6, E=1, OA=1
pswInputs(:, 65) = [0.025; 1.5; 0.45; 6.2917]; % RCP 8.5, E=1, OA=1
% ===== Inputs below were computed separately =====
% They don't have the same sort order as those above.
% == Target 3, advantage 1 ==
pswInputs(:, 66) = [0.025; 1.5; 0.45; 4.3657]; % RCP 2.6, E=0, OA=0
pswInputs(:, 67) = [0.025; 1.5; 0.45; 4.4491]; % RCP 4.5, E=0, OA=0
pswInputs(:, 68) = [0.025; 1.5; 0.45; 4.3704]; % RCP 6, E=0, OA=0
pswInputs(:, 69) = [0.025; 1.5; 0.45; 4.2963]; % RCP 8.5, E=0, OA=0
pswInputs(:, 70) = [0.025; 1.5; 0.45; 4.6713]; % RCP 2.6, E=1, OA=0
pswInputs(:, 71) = [0.025; 1.5; 0.45; 4.6944]; % RCP 4.5, E=1, OA=0
pswInputs(:, 72) = [0.025; 1.5; 0.45; 4.6991]; % RCP 6, E=1, OA=0
pswInputs(:, 73) = [0.025; 1.5; 0.45; 4.5139]; % RCP 8.5, E=1, OA=0
% == Target 10, advantage 1 ==
pswInputs(:, 74) = [0.025; 1.5; 0.45; 7.3426]; % RCP 2.6, E=0, OA=0
pswInputs(:, 75) = [0.025; 1.5; 0.45; 7.75]; % RCP 4.5, E=0, OA=0
pswInputs(:, 76) = [0.025; 1.5; 0.45; 7.3519]; % RCP 6, E=0, OA=0
pswInputs(:, 77) = [0.025; 1.5; 0.45; 7.6991]; % RCP 8.5, E=0, OA=0
pswInputs(:, 78) = [0.025; 1.5; 0.45; 7.875]; % RCP 2.6, E=1, OA=0
pswInputs(:, 79) = [0.025; 1.5; 0.45; 8.3935]; % RCP 4.5, E=1, OA=0
pswInputs(:, 80) = [0.025; 1.5; 0.45; 7.5602]; % RCP 6, E=1, OA=0
pswInputs(:, 81) = [0.025; 1.5; 0.45; 8.1181]; % RCP 8.5, E=1, OA=0
% Average RCP values for otherwise identical cases in sets of 4.
currentSet = 100;
for i = [2:4:78]
sSum = 0.0;
for j = 0:3
sSum = sSum + pswInputs(4, i+j);
end
% Copy the first 3 values and use the averaged s.
pswInputs(:, currentSet) = [pswInputs(1, i), pswInputs(2, i), pswInputs(3, i), sSum/4];
currentSet = currentSet + 1;
end
[~, pswCount] = size(pswInputs);
%% CALULATE PROP CONSTANT FOR EACH GRID CELL
% Change, Jan 2020.
% Instead of calculating all the cases for future use, at risk of mismatching
% the SST data and the cases, just do case 1 for the optimizer. Only store the
% full list of input values, which will be used to calculate psw2 at the time of
% each simulation.
psw2_new = nan(length(Reefs_latlon),1); % initialize matrix
for reef = 1:length(Reefs_latlon)
SSThistReef = SST_1861_2000(reef,:)'; % extract SSTs for grid cell bn 1861-2000
% E = exp(0.063.*SSThistReef)
% max(pMin,min(pMax,(mean(E)./var(E)).^exponent/divisor))
for j = 1:1 % pswCount
psw2_new(reef,j) = max(pswInputs(1, j),min(pswInputs(2, j),(mean(exp(0.063.*SSThistReef))./var(exp(0.063.*SSThistReef))).^pswInputs(3, j)/pswInputs(4, j))); % John's new eqn 7/25/16
end
clear SSThistReef
end
%% Save new psw2 values
cd(matPath);
save('Optimize_psw2.mat', 'pswInputs'); %% CAL 10-3-16 based on hist SSTs!
%save('Optimize_psw2.mat','psw2_new','Reefs_latlon','pswInputs'); %% CAL 10-3-16 based on hist SSTs!
cd(basePath);