-
Notifications
You must be signed in to change notification settings - Fork 36
/
FindThreshold.m
600 lines (510 loc) · 18.1 KB
/
FindThreshold.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
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
function [Sn,thresh_list,xd_list]=FindThreshold(DEM,FD,A,S,num_streams,varargin)
%
% Usage:
% [Sn,thresh_list,xd_list]=FindThreshold(DEM,FD,A,S,num_streams);
% [Sn,thresh_list,xd_list]=FindThreshold(DEM,FD,A,S,num_streams,'name',value,...);
%
% Description:
% Function to interactively select an appopriate threshold area for a given stream
% network. Function will either have you iterate through a number of single streams,
% controlled by the number passed to 'num_streams', extracted from the drainage divide or
% all streams within the provided drainage network if you provide 'all' to 'num_streams'.
% If 'num_streams' is numeric, then the function will use the average of the user selected
% minimum threshold areas to define a new stream network. If 'num_streams' is set to 'all', the
% function will use the user selected minimum threshold areas to define a new stream network for
% each individual stream, i.e. the minimum threshold area will be different for each stream base
% on your selections (though you can force a similar average threshold behavior with the
% 'remake_network' parameter). You can use either chi-elevation or slope-area plots (the default),
% both plots will be displayed regardless of choice, to visually select where channels begin. Function
% also outputs the lists of selected threshold areas and distance from channel head to divide. There is
% now also an 'auto' option which will attempt to find the colluvial-fluvial transition using the
% ischange function on a normalized chi-elevation plot. Using the 'auto' option requires MATLAB 2017a
% or greater. The auto function may be over shoot the threshold area, especially in areas where there
% is a noticeable lower relief portion of a stream, e.g. a stream draining the edge of a plateau,
% one that has recently captured drainage area, etc. You can try to control for this by setting
% the optional 'max_threshold' parameter, which will serve as a ceiling on the auto detected threshold.
%
% Required Inputs:
% DEM - Digital Elevation as a GRIDobj, assumes unconditioned DEM (e.g. DEMoc from ProcessRiverBasins
% or output from MakeStreams)
% FD - Flow direction as FLOWobj
% A - Flow accumulation GRIDobj
% S - Stream network as STREAMobj (doesn't really matter what the threshold area used to calculate
% this was originally)
% num_streams - Number of stream profiles to view and select threshold areas. If you wish to manually
% select threshold areas for all streams in the provided network, provide 'all' instead of a number.
% If you want to instead try an auto detection method for the colluvial to fluvial transtion, provide
% 'auto'.
%
% Optional Inputs;
% ref_concavity [0.50] - refrence concavity used to generate the chi-elevation plot. Also used if
% 'num_streams' is set to 'auto' as the threshold area is determined based on changes in chi-elevation
% relationships
% pick_method ['slope_area']- type of plot you wish to choose the threshold area on, valid options are:
% 'chi' - Choose threshold areas on a chi elevation plot
% 'slope_area' - Choose threshold areas on slope-area plot
% remake_network [false] - logical flag used if 'num_streams' is 'auto' or 'all'. If remake_network is false
% (the default), then each channel head will have a threshold area based on the user selection or the
% auto detection. If remake_network is true, then the behavior will be similar to if a numeric value is
% provided to 'num_streams', where the mean threshold area of all streams will be used to regenerate the
% network. If remake_network is false, this means the total number of channel heads of the output 'Sn'
% STREAMobj will equal that of the input 'S' and thus the drainage density will basically remain unchanged.
% If remake_network is true then there will likely be different numbers of channel heads between Sn and S
% (unless the average threshold area happens to equals the minimum threshold area used to generate S).
% max_threshold [] - a maximum threshold area (in squared map units) to not be exceeded by any stream when the
% 'num_streams' is set to 'auto'.
%
% Outputs:
% Sn - New version of the STREAMobj using the mean threshold area to define streams or the user selected
% minimum threshold areas for each stream if 'num_streams' is set to 'all' or 'auto'
% thresh_list - list of chosen threshold areas
% xd_list - list of chosen distances from channel head to divide
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Function Written by Adam M. Forte - Updated : 02/09/19 %
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Parse Inputs
p = inputParser;
p.FunctionName = 'FitThreshold';
addRequired(p,'DEM',@(x) isa(x,'GRIDobj'));
addRequired(p,'FD',@(x) isa(x,'FLOWobj'));
addRequired(p,'A',@(x) isa(x,'GRIDobj'));
addRequired(p,'S',@(x) isa(x,'STREAMobj'));
addRequired(p,'num_streams',@(x) isnumeric(x) & isscalar(x) || ischar(validatestring(x,{'all','auto'})));
addParameter(p,'pick_method','slope_area',@(x) ischar(validatestring(x,{'chi','slope_area'})));
addParameter(p,'ref_concavity',0.50,@(x) isscalar(x) && isnumeric(x));
addParameter(p,'remake_network',false,@(x) isscalar(x) && islogical(x));
addParameter(p,'max_threshold',[],@(x) isscalar(x) && isnumeric(x) || isempty(x));
parse(p,DEM,FD,A,S,num_streams,varargin{:});
DEM=p.Results.DEM;
FD=p.Results.FD;
A=p.Results.A;
S=p.Results.S;
num_streams=p.Results.num_streams;
pick_method=p.Results.pick_method;
ref_theta=p.Results.ref_concavity;
remake_network=p.Results.remake_network;
mt=p.Results.max_threshold;
% Find channel heads and flow distances
chix=streampoi(S,'channelheads','ix');
FLUS=flowdistance(FD);
FLDS=flowdistance(FD,'downstream');
DA=A.*(DEM.cellsize^2);
% Set bin size based on DEM resolution
bin_size=15*DEM.cellsize;
% Parse type of operation
if isnumeric(num_streams)
op=1;
elseif strcmp(num_streams,'all')
op=2;
elseif strcmp(num_streams,'auto')
op=3;
end
if verLessThan('matlab','9.3') & op==3;
warning('Using the "auto" version of this function requires MATLAB 2017a or later, switching to "all" streams');
op=2;
end
switch op
case 1
% Determine number of channels and compare that to total number of channels
num_ch=numel(chix);
if num_streams>num_ch;
num_streams=num_ch;
if isdeployed
warndlg('Number of streams to fit was greater than number of channel heads, using all streams')
else
warning('Number of streams to fit was greater than number of channel heads, using all streams')
end
end
% Sort channels by length, fitting proceeds from largest to smallest channels
fl=FLUS.Z(chix);
[fl,six]=sort(fl,'descend');
chix=chix(six);
for ii=1:num_streams
chOI=chix(ii);
UP=dependencemap(FD,chOI);
FLDSt=FLUS.*UP;
[~,ix]=max(FLDSt);
IX=influencemap(FD,ix);
St=STREAMobj(FD,IX);
z=mincosthydrocon(St,DEM,'interp',0.1);
C=chiplot(St,z,A,'a0',1,'mn',ref_theta,'plot',false);
[bs,ba,bc,bd,aa,ag,ac]=sa(DEM,St,A,C.chi,bin_size);
f1=figure(1);
set(f1,'Units','normalized','Position',[0.5 0.1 0.45 0.8],'renderer','painters');
clf
colormap(jet);
switch pick_method
case 'chi'
ax2=subplot(2,1,2);
hold on
scatter(aa,ag,5,ac,'+');
scatter(ba,bs,20,bc,'filled','MarkerEdgeColor','k');
xlabel('Log Drainage Area');
ylabel('Log Gradient');
caxis([0 max(C.chi)]);
set(ax2,'YScale','log','XScale','log','XDir','reverse');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax2);
end
hold off
ax1=subplot(2,1,1);
hold on
plot(C.chi,C.elev,'-k');
scatter(C.chi,C.elev,10,C.chi,'filled');
xlabel('\chi');
ylabel('Elevation (m)');
title(['Choose hillslope to channel transition : ' num2str(num_streams-ii) ' streams remaining to pick']);
caxis([0 max(C.chi)]);
ax1.XColor='Red';
ax1.YColor='Red';
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax1);
end
hold off
% Find user selected threshold area
[c,~]=ginput(1);
[~,cix]=min(abs(C.chi-c),[],'omitnan');
a=C.area(cix);
% Find xd
cx=C.x(cix);
cy=C.y(cix);
ccix=coord2ind(DEM,cx,cy);
xd=FLDS.Z(ccix);
case 'slope_area'
ax2=subplot(2,1,2);
hold on
plot(C.chi,C.elev,'-k');
scatter(C.chi,C.elev,10,C.chi,'filled');
xlabel('\chi');
ylabel('Elevation (m)');
caxis([0 max(C.chi)]);
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax2);
end
hold off
ax1=subplot(2,1,1);
hold on
scatter(aa,ag,5,ac,'+');
scatter(ba,bs,20,bc,'filled','MarkerEdgeColor','k');
xlabel('Log Drainage Area');
ylabel('Log Gradient');
title(['Choose hillslope to channel transition : ' num2str(num_streams-ii) ' streams remaining to pick']);
caxis([0 max(C.chi)]);
set(ax1,'YScale','log','XScale','log','XDir','reverse');
ax1.XColor='Red';
ax1.YColor='Red';
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax1);
end
hold off
% Find user selected threshold area
[a,~]=ginput(1);
% Find xd
[~,cix]=min(abs(C.area-a),[],'omitnan');
cx=C.x(cix);
cy=C.y(cix);
ccix=coord2ind(DEM,cx,cy);
xd=FLDS.Z(ccix);
end
close(f1);
thresh_list(ii,1)=a;
xd_list(ii,1)=xd;
end
mean_thresh=mean(thresh_list);
mean_xd=mean(xd_list);
f1=figure(1);
set(f1,'Units','normalized','Position',[0.5 0.1 0.4 0.8],'renderer','painters');
sbplt1=subplot(2,1,1);
hold on
edges=logspace(log10(min(thresh_list)),log10(max(thresh_list)),10);
[N,e]=histcounts(thresh_list,edges);
histogram(thresh_list,edges);
plot([mean_thresh,mean_thresh],[0,max(N)],'-k','LineWidth',2);
xlabel('Picked Threshold Areas (m^{2})');
a_str=sprintf('%0.1e',mean_thresh);
title(['Mean threshold area = ' a_str]);
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt1);
end
hold off
sbplt2=subplot(2,1,2);
hold on
[N,~]=histcounts(xd_list,10);
histogram(xd_list,10);
plot([mean_xd,mean_xd],[0,max(N)],'-k','LineWidth',2);
xlabel('Mean distance from channel head to divide (m)');
title(['Mean xd = ' num2str(round(mean_xd,1))]);
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt2);
end
hold off
Sn=STREAMobj(FD,'minarea',mean_thresh,'unit','mapunits');
case 2
% Sort channels by length, fitting proceeds from largest to smallest channels
fl=FLUS.Z(chix);
[fl,six]=sort(fl,'descend');
chix=chix(six);
% Generate empty outputs
xd_list=zeros(numel(chix),1);
thresh_list=zeros(numel(chix),1);
ix_list=cell(numel(chix),1);
for ii=1:numel(chix)
chOI=chix(ii);
UP=dependencemap(FD,chOI);
FLDSt=FLUS.*UP;
[~,ix]=max(FLDSt);
IX=influencemap(FD,ix);
St=STREAMobj(FD,IX);
z=mincosthydrocon(St,DEM,'interp',0.1);
C=chiplot(St,z,A,'a0',1,'mn',ref_theta,'plot',false);
[bs,ba,bc,bd,aa,ag,ac]=sa(DEM,St,A,C.chi,bin_size);
f1=figure(1);
set(f1,'Units','normalized','Position',[0.5 0.1 0.45 0.8],'renderer','painters');
clf
colormap(jet);
switch pick_method
case 'chi'
ax2=subplot(2,1,2);
hold on
scatter(aa,ag,5,ac,'+');
scatter(ba,bs,20,bc,'filled','MarkerEdgeColor','k');
xlabel('Log Drainage Area');
ylabel('Log Gradient');
caxis([0 max(C.chi)]);
set(ax2,'YScale','log','XScale','log','XDir','reverse');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax2);
end
hold off
ax1=subplot(2,1,1);
hold on
plot(C.chi,C.elev,'-k');
scatter(C.chi,C.elev,10,C.chi,'filled');
xlabel('\chi');
ylabel('Elevation (m)');
title(['Choose hillslope to channel transition : ' num2str(numel(chix)-ii) ' streams remaining to pick']);
caxis([0 max(C.chi)]);
ax1.XColor='Red';
ax1.YColor='Red';
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax1);
end
hold off
% Find user selected threshold area
[c,~]=ginput(1);
[~,cix]=min(abs(C.chi-c),[],'omitnan');
a=C.area(cix);
% Find xd
cx=C.x(cix);
cy=C.y(cix);
ccix=coord2ind(DEM,cx,cy);
xd=FLDS.Z(ccix);
% Find ix for all streams segments
allx=C.x(C.area>=a);
ally=C.y(C.area>=a);
ix_list{ii,1}=coord2ind(DEM,allx,ally);
case 'slope_area'
ax2=subplot(2,1,2);
hold on
plot(C.chi,C.elev,'-k');
scatter(C.chi,C.elev,10,C.chi,'filled');
xlabel('\chi');
ylabel('Elevation (m)');
caxis([0 max(C.chi)]);
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax2);
end
hold off
ax1=subplot(2,1,1);
hold on
scatter(aa,ag,5,ac,'+');
scatter(ba,bs,20,bc,'filled','MarkerEdgeColor','k');
xlabel('Log Drainage Area');
ylabel('Log Gradient');
title(['Choose hillslope to channel transition : ' num2str(numel(chix)-ii) ' streams remaining to pick']);
caxis([0 max(C.chi)]);
set(ax1,'YScale','log','XScale','log','XDir','reverse');
ax1.XColor='Red';
ax1.YColor='Red';
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(ax1);
end
hold off
% Find user selected threshold area
[a,~]=ginput(1);
% Find xd
[~,cix]=min(abs(C.area-a),[],'omitnan');
cx=C.x(cix);
cy=C.y(cix);
ccix=coord2ind(DEM,cx,cy);
xd=FLDS.Z(ccix);
% Find ix for all streams segments
allx=C.x(C.area>=a);
ally=C.y(C.area>=a);
ix_list{ii,1}=coord2ind(DEM,allx,ally);
end
close(f1);
thresh_list(ii,1)=a;
xd_list(ii,1)=xd;
end
f1=figure(1);
set(f1,'Units','normalized','Position',[0.5 0.1 0.4 0.8],'renderer','painters');
sbplt1=subplot(2,1,1);
hold on
edges=logspace(log10(min(thresh_list)),log10(max(thresh_list)),10);
[N,e]=histcounts(thresh_list,edges);
histogram(thresh_list,edges);
xlabel('Picked Threshold Areas (m^{2})');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt1);
end
hold off
sbplt2=subplot(2,1,2);
hold on
[N,~]=histcounts(xd_list,10);
histogram(xd_list,10);
xlabel('Mean distance from channel head to divide (m)');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt2);
end
hold off
if remake_network
mean_thresh=mean(thresh_list);
Sn=STREAMobj(FD,'minarea',mean_thresh,'unit','mapunits');
else
% Collapse ix list and create logical raster
ix_list=vertcat(ix_list{:});
ix_list=unique(ix_list);
W=GRIDobj(DEM,'logical');
W.Z(ix_list)=true;
Sn=STREAMobj(FD,W);
end
case 3
% Generate empty outputs
xd_list=zeros(numel(chix),1);
thresh_list=zeros(numel(chix),1);
ix_list=cell(numel(chix),1);
w1=waitbar(0,'Finding stream thresholds based on chi-elevation inflection...');
for ii=1:numel(chix)
chOI=chix(ii);
UP=dependencemap(FD,chOI);
FLDSt=FLUS.*UP;
[~,ix]=max(FLDSt);
IX=influencemap(FD,ix);
St=STREAMobj(FD,IX);
z=mincosthydrocon(St,DEM,'interp',0.1);
c=chitransform(St,A,'mn',ref_theta,'a0',1);
da=getnal(St,DA);
% ischange expects an ascending ordered list
[cs,six]=sort(c);
zs=z(six);
das=da(six);
% Normalize chi and elevation so consistent threshold can be used
zn=zs-min(zs);
zn=zn./max(zn);
cn=cs./max(cs);
TF=ischange(zn,'linear','SamplePoints',cn,'Threshold',0.05);
chi_ix=find(TF,1,'last');
if isempty(chi_ix)
% Catch for if no change point is detected
ta=DA.Z(chOI);
else
ta=das(chi_ix);
if ~isempty(mt) && ta>mt
ta=mt;
end
end
thresh_list(ii,1)=ta;
[~,loc_ix]=min(abs(da-ta));
% Find xd
cx=St.x(loc_ix);
cy=St.y(loc_ix);
ccix=coord2ind(DEM,cx,cy);
xd_list(ii,1)=FLDS.Z(ccix);
% Find ix for all streams segments
ix_list{ii,1}=St.IXgrid(da>=ta);
waitbar(ii/numel(chix));
end
close(w1);
f1=figure(1);
clf
set(f1,'Units','normalized','Position',[0.5 0.1 0.4 0.8],'renderer','painters');
sbplt1=subplot(2,1,1);
hold on
edges=logspace(log10(min(thresh_list)),log10(max(thresh_list)),10);
[N,e]=histcounts(thresh_list,edges);
histogram(thresh_list,edges);
set(sbplt1,'XScale','log');
xlabel('Picked Threshold Areas (m^{2})');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt1);
end
hold off
sbplt2=subplot(2,1,2);
hold on
[N,~]=histcounts(xd_list,50);
histogram(xd_list,50);
xlabel('Mean distance from channel head to divide (m)');
if ~verLessThan('matlab','9.5')
disableDefaultInteractivity(sbplt2);
end
hold off
if remake_network
mean_thresh=mean(thresh_list);
Sn=STREAMobj(FD,'minarea',mean_thresh,'unit','mapunits');
else
% Collapse ix list and create logical raster
ix_list=vertcat(ix_list{:});
ix_list=unique(ix_list);
W=GRIDobj(DEM,'logical');
W.Z(ix_list)=true;
Sn=STREAMobj(FD,W);
end
end
end
function [bs,ba,bc,bd,a,g,C]=sa(DEM,S,A,C,bin_size)
% Modified slope area function that uses the smooth length to
% to determine the number of bins and uses those same bins
% to find mean values of chi and distance for plotting
% purposes
minX=min(S.distance);
maxX=max(S.distance);
b=[minX:bin_size:maxX+bin_size];
numbins=round(max([numel(b) numel(S.IXgrid)/10]));
an=getnal(S,A.*A.cellsize^2);
z=getnal(S,DEM);
gn=gradient(S,z,'unit','tangent','method','robust','drop',20);
gn=smooth(gn,3);
% Run through STREAMobj2XY so chi and everything else are same size
[~,~,a,g,d]=STREAMobj2XY(S,an,gn,S.distance);
% Remove NaNs
a(isnan(a))=[];
g(isnan(g))=[];
d(isnan(d))=[];
C(isnan(C))=[];
mina=min(a);
maxa=max(a);
edges = logspace(log10(mina-0.1),log10(maxa+1),numbins+1);
try
% histc is deprecated
[ix]=discretize(a,edges);
catch
[~,ix] = histc(a,edges);
end
ba=accumarray(ix,a,[numbins 1],@median,nan);
bs=accumarray(ix,g,[numbins 1],@(x) mean(x(~isnan(x))),nan);
bd=accumarray(ix,d,[numbins 1],@mean,nan);
bc=accumarray(ix,C,[numbins 1],@mean,nan);
% Filter negatives
idx=bs>=0 & ba>=0 & bc>=0 & bd>=0;
bs=bs(idx);
ba=ba(idx);
bc=bc(idx);
bd=bd(idx);
idx=a>=0 & g>=0 & d>=0 & C>=0;
a=a(idx);
g=g(idx);
d=d(idx);
C=C(idx);
end