-
Notifications
You must be signed in to change notification settings - Fork 6
/
Copy pathLakeCat_functions.py
1290 lines (1165 loc) · 56.5 KB
/
LakeCat_functions.py
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
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
# -*- coding: utf-8 -*-
"""
Created on Tue May 31 15:22:43 2016
@author: Rdebbout
"""
import os
import re
import sys
import arcpy
import struct
import decimal
import warnings
import datetime
import itertools
import numpy as np
import pysal as ps
import pandas as pd
import rasterio as rs
from osgeo import gdal
from Tkinter import Tk
import geopandas as gpd
# we run out of memory when trying to convert the basin rasters to features
# so are now using Arc's RasterToPolygon
#from rasterio.crs import CRS
#from rasterio import features
from arcpy.sa import Watershed
from geopandas.tools import sjoin
from arcpy import PolygonToRaster_conversion as p2r
from arcpy import RasterToPolygon_conversion as r2p
from collections import deque, OrderedDict, defaultdict
from tkFileDialog import askdirectory
arcpy.CheckOutExtension("spatial")
warnings.filterwarnings("ignore", category=FutureWarning)
Tk().withdraw()
##############################################################################
inputs = OrderedDict([('06', 'MS'),
('05', 'MS'),
('10U', 'MS'),
('10L', 'MS'),
('07', 'MS'),
('11', 'MS'),
('14', 'CO'),
('01', 'NE'),
('17', 'PN'),
('16', 'GB'),
('15', 'CO'),
('13', 'RG'),
('12', 'TX'),
('09', 'SR'),
('02', 'MA'),
('08', 'MS'),
('04', 'GL'),
('03W', 'SA'),
('03S', 'SA'),
('03N', 'SA'),
('18', 'CA')])
rpus = OrderedDict([(u'01', ['01a']),
(u'02', ['02a', '02b']),
(u'03N', ['03a', '03b']),
(u'03S', ['03c', '03d']),
(u'03W', ['03e', '03f']),
(u'04', ['04a', '04b', '04c', '04d']),
(u'05', ['05a', '05b', '05c', '05d']),
(u'06', ['06a']),
(u'07', ['07a', '07b', '07c']),
(u'08', ['03g', '08a', '08b']),
(u'09', ['09a']),
(u'10L', ['10a', '10b', '10c', '10d']),
(u'10U', ['10e', '10f', '10g', '10h', '10i']),
(u'11', ['11a', '11b', '11c', '11d']),
(u'12', ['12a', '12b', '12c', '12d']),
(u'13', ['13a', '13b', '13c', '13d']),
(u'14', ['14a', '14b']),
(u'15', ['15a', '15b']),
(u'16', ['16a', '16b']),
(u'17', ['17a', '17b', '17c', '17d']),
(u'18', ['18a', '18b', '18c'])])
# this is the same projection as the 2 below, but this matches the name of the proj
# from the NHD fdr files..
fiftyseventy = "PROJCS['NAD_1983_Albers',\
GEOGCS['GCS_North_American_1983',\
DATUM['D_North_American_1983',\
SPHEROID['GRS_1980',6378137.0,298.257222101]],\
PRIMEM['Greenwich',0.0],\
UNIT['Degree',0.0174532925199433]],\
PROJECTION['Albers'],PARAMETER['False_Easting',0.0],\
PARAMETER['False_Northing',0.0],\
PARAMETER['Central_Meridian',-96.0],\
PARAMETER['Standard_Parallel_1',29.5],\
PARAMETER['Standard_Parallel_2',45.5],\
PARAMETER['Latitude_Of_Origin',23.0],\
UNIT['Meter',1.0]],\
VERTCS['Unknown VCS',VDATUM['Unknown'],\
PARAMETER['Vertical_Shift',0.0],\
PARAMETER['Direction',1.0],\
UNIT['User_Defined_Unit',0.01]]"
# this is the projection that will be written out if set to 5070 in QGIS..
#"PROJCS['NAD83_Conus_Albers',\
#GEOGCS['GCS_North_American_1983',\
#DATUM['D_North_American_1983',\
#SPHEROID['GRS_1980',6378137.0,298.257222101]],\
#PRIMEM['Greenwich',0.0],\
#UNIT['Degree',0.0174532925199433]],\
#PROJECTION['Albers'],\
#PARAMETER['false_easting',0.0],\
#PARAMETER['false_northing',0.0],\
#PARAMETER['central_meridian',-96.0],\
#PARAMETER['standard_parallel_1',29.5],\
#PARAMETER['standard_parallel_2',45.5],\
#PARAMETER['latitude_of_origin',23.0],\
#UNIT['Meter',1.0]]"
# this is just another that we had been using, it matches others
#"PROJCS['NAD_1983_Contiguous_USA_Albers',\
#GEOGCS['GCS_North_American_1983',\
#DATUM['D_North_American_1983',\
#SPHEROID['GRS_1980',6378137.0,298.257222101]],\
#PRIMEM['Greenwich',0.0],\
#UNIT['Degree',0.0174532925199433]],\
#PROJECTION['Albers'],\
#PARAMETER['false_easting',0.0],\
#PARAMETER['false_northing',0.0],\
#PARAMETER['central_meridian',-96.0],\
#PARAMETER['standard_parallel_1',29.5],\
#PARAMETER['standard_parallel_2',45.5],\
#PARAMETER['latitude_of_origin',23.0],\
#UNIT['Meter',1.0]]"
##############################################################################
def dbfreader(f):
"""Returns an iterator over records in a Xbase DBF file.
The first row returned contains the field names.
The second row contains field specs: (type, size, decimal places).
Subsequent rows contain the data records.
If a record is marked as deleted, it is skipped.
File should be opened for binary reads.
"""
# See DBF format spec at:
# http://www.pgts.com.au/download/public/xbase.htm#DBF_STRUCT
numrec, lenheader = struct.unpack('<xxxxLH22x', f.read(32))
numfields = (lenheader - 33) // 32
fields = []
for fieldno in xrange(numfields):
name, typ, size, deci = struct.unpack('<11sc4xBB14x', f.read(32))
name = name.replace('\0', '') # eliminate NULs from string
fields.append((name, typ, size, deci))
yield [field[0] for field in fields]
yield [tuple(field[1:]) for field in fields]
terminator = f.read(1)
assert terminator == '\r'
fields.insert(0, ('DeletionFlag', 'C', 1, 0))
fmt = ''.join(['%ds' % fieldinfo[2] for fieldinfo in fields])
fmtsiz = struct.calcsize(fmt)
for i in xrange(numrec):
record = struct.unpack(fmt, f.read(fmtsiz))
if record[0] != ' ':
continue # deleted record
result = []
for (name, typ, size, deci), value in itertools.izip(fields, record):
if name == 'DeletionFlag':
continue
if typ == "N":
value = value.replace('\0', '').lstrip()
if value == '':
value = 0
elif deci:
value = decimal.Decimal(value)
else:
value = int(value)
elif typ == 'C':
value = value.rstrip()
elif typ == 'D':
try:
y, m, d = int(value[:4]), int(value[4:6]), int(value[6:8])
value = datetime.date(y, m, d)
except:
value = None
elif typ == 'L':
value = (value in 'YyTt' and 'T') or (value in 'NnFf' and 'F') or '?'
elif typ == 'F':
value = float(value)
result.append(value)
yield result
def dbf2DF(f, upper=True):
data = list(dbfreader(open(f, 'rb')))
if upper == False:
return pd.DataFrame(data[2:], columns=data[0])
else:
return pd.DataFrame(data[2:], columns=map(str.upper,data[0]))
##############################################################################
def Accumulation(arr, COMIDs, lengths, upStream, tbl_type, icol):
'''
__author__ = "Marc Weber <[email protected]>"
"Ryan Hill <[email protected]>"
Uses the 'Cat' and 'UpCat' columns to caluculate watershed values and returns those values in 'Cat' columns
so they can be appended to 'CatResult' tables in other zones before accumulation.
Arguments
---------
arr : table containing watershed values
COMIDs : numpy array of all zones COMIDs
lengths : numpy array with lengths of upstream COMIDs
upstream : numpy array of all upstream arrays for each COMID
tbl_type : string value of table metrics to be returned
'''
coms = np.array(arr[icol]) #Read in COMIDs
indices = swapper(coms, upStream) #Get indices that will be used to map values
del upStream # a and indices are big - clean up to minimize RAM
cols = arr.columns[1:] #Get column names that will be accumulated
z = np.zeros(COMIDs.shape) #Make empty vector for placing values
outT = np.zeros((len(COMIDs), len(arr.columns))) #Make empty array for placing final values
outT[:,0] = COMIDs #Define first column as comids
#Loop and accumulate values
for k in range(0,len(cols)):
col = cols[k]
c = np.array(arr[col]) # arr[col].fillna(0) keep out zeros where no data!
d = c[indices] #Make final vector from desired data (c)
if 'PctFull' in col:
area = np.array(arr.ix[:, 1])
ar = area[indices]
x = 0
for i in range(0, len(lengths)):
# using nan_to_num in average function to treat NA's as zeros when summing
z[i] = np.ma.average(np.nan_to_num(d[x:x + lengths[i]]), weights=ar[x:x + lengths[i]])
x = x + lengths[i]
else:
x = 0
for i in range(0, len(lengths)):
z[i] = np.nansum(d[x:x + lengths[i]])
x = x + lengths[i]
outT[:,k+1] = z #np.nan_to_num() -used to convert to zeros here, now done above in the np.ma.average()
outT = outT[np.in1d(outT[:,0], coms),:] #Remove the extra COMIDs
outDF = pd.DataFrame(outT)
if tbl_type == 'Ws':
outDF.columns = np.append(icol, map(lambda x : x.replace('Cat', 'Ws'),cols.values))
elif tbl_type == 'UpCat':
outDF.columns = np.append(icol, 'Up' + cols.values)
else:
outDF.columns = [icol] + cols.tolist()
for name in outDF.columns:
if 'AreaSqKm' in name:
areaName = name
if 'PctFull' in name:
pct = name
outDF.loc[(outDF[areaName] == 0), outDF.columns[2:]] = np.nan # identifies that there is no area in catchment mask, then NA values across the table
outDF.loc[(outDF[pct] == 0), outDF.columns[2:-1]] = np.nan
return outDF
##############################################################################
def children(token, tree, chkset=None):
'''
__author__ = "Ryan Hill <[email protected]>"
"Marc Weber <[email protected]>"
returns a list of every child
Arguments
---------
token : a single COMID
tree : Full dictionary of list of upstream COMIDs for each COMID in the zone
chkset : set of all the NHD catchment COMIDs used to remove flowlines with no associated catchment
'''
visited = set()
to_crawl = deque([token])
while to_crawl:
current = to_crawl.popleft()
if current in visited:
continue
visited.add(current)
node_children = set(tree[current])
to_crawl.extendleft(node_children - visited)
#visited.remove(token)
if chkset != None:
visited = visited.intersection(chkset)
return list(visited)
##############################################################################
def bastards(token, tree, chkset=None):
'''
__author__ = "Ryan Hill <[email protected]>"
"Marc Weber <[email protected]>"
returns a list of every child w/ out father (key) included
Arguments
---------
token : a single COMID
tree : Full dictionary of list of upstream COMIDs for each COMID in the zone
chkset : set of all the NHD catchment COMIDs, used to remove flowlines with no associated catchment
'''
visited = set()
to_crawl = deque([token])
while to_crawl:
current = to_crawl.popleft()
if current in visited:
continue
visited.add(current)
node_children = set(tree[current])
to_crawl.extendleft(node_children - visited)
visited.remove(token)
if chkset != None:
visited = visited.intersection(chkset)
return list(visited)
##############################################################################
def swapper(coms, upStream):
'''
__author__ = "Marc Weber <[email protected]>"
"Ryan Hill <[email protected]>"
Creates array of indexes for all upstream COMIDs that will be summarized for each local catchment.
Arguments
---------
coms : numpy array of all COMIDs in the zone
upstream : numpy array of all upstream COMIDs for each local catchment
'''
bsort = np.argsort(coms)
apos = np.searchsorted(coms[bsort], upStream)
indices = bsort[apos]
return indices
##############################################################################
def findUpstreamNpy(numpy_dir, com):
""" Unpacks Numpy files describing the array of upstream COMID's for
each catchment in NHD. Similar to the Arc add-in tool that Marc made to
identify upstream flowline/catchments. NOT USED IN LAKECAT PROCESS! for QA.
Arguments
---------
numpy_dir : Loccation of numpy files
com : ID of feature
"""
comids = np.load(numpy_dir + '/comids.npy')
lengths= np.load(numpy_dir + '/lengths.npy')
upStream = np.load(numpy_dir + '/upStream.npy')
itemindex = int(np.where(comids == com)[0])
n = lengths[:itemindex].sum()
arrlen = lengths[itemindex]
return upStream[n:n+arrlen]
##############################################################################
def PointInPoly(points, inZoneData, pct_full, fld='GRIDCODE', summaryfield=None):
'''
__author__ = "Marc Weber <[email protected]>"
"Rick Debbout <[email protected]>"
Returns either the count of spatial points feature in every polygon in a spatial polygons feature or the summary of
an attribute field for all the points in every polygon of a spatial polygons feature
Arguments
---------
points : input points geographic features as a GeoPandas GeoDataFrame
InZoneData : input polygon shapefile as a string, i.e. 'C:/Temp/outshape.shp'
pct_full : table that links COMIDs to pct_full, determined from catchments that are not within the US Census border
fld : the field in the InZoneData file that uniquely identifies each polygon
summaryfield : a list of the field/s in points feature to use for getting summary stats in polygons
'''
polys = gpd.GeoDataFrame.from_file(inZoneData)
if not points.crs == polys.crs:
points = points.to_crs(polys.crs)
# Get list of lat/long fields in the table
latlon = [s for s in points.columns if any(xs in s.upper() for xs in ['LONGIT','LATIT'])]
# Remove duplicate points for 'Count'
points2 = points.ix[~points.duplicated(latlon)]
try:
point_poly_join = sjoin(points2, polys, how="left", op="within")
except:
polys['link'] = np.nan
point_poly_join = polys #gpd.GeoDataFrame( pd.concat( [points2, polys], ignore_index=True) )
fld = 'link'
# Create group of all points in catchment
grouped = point_poly_join.groupby(fld)
point_poly_count = grouped[fld].count() # point_poly_count.head() next((x for x in points2.columns if x != 'geometry'),None)
point_poly_count.name = 'COUNT'
# Join Count column on to NHDCatchments table and keep only 'COMID','CatAreaSqKm','CatCount'
final = polys.join(point_poly_count, on=fld, how='left')
final = final[[fld, 'AreaSqKM', 'COUNT']].fillna(0)
cols = [fld, 'CatAreaSqKm', 'CatCount']
if not summaryfield == None: # Summarize fields in list with gpd table including duplicates
point_poly_dups = sjoin(points, polys, how="left", op="within")
grouped2 = point_poly_dups.groupby(fld)
for x in summaryfield: # Sum the field in summary field list for each catchment
point_poly_stats = grouped2[x].sum()
final = final.join(point_poly_stats, on=fld, how='left').fillna(0)
cols.append('Cat' + x)
final.columns = cols
# Merge final table with Pct_Full table based on COMID and fill NA's with 0
final = pd.merge(final, pct_full, on=fld, how='left')
final['CatPctFull'] = final['CatPctFull'].fillna(100) # final.head() final.ix[final.CatCount == 0]
#print "elapsed time " + str(dt.now()-startTime)
for name in final.columns:
if 'AreaSqKm' in name:
area = name
# replace CatAreaSqKm with NANs where value is zero
final.loc[(final[area] == 0), final.columns[2:]] = np.nan
return final
##############################################################################
def chkColumnLength(table, LandscapeLayer):
'''
__author__ = "Marc Weber <[email protected]>"
"Ryan Hill <[email protected]>"
Checks the number of columns returned from zonal stats and adds any of the
categorical values that that didn't exist within the zone and fills the
column with zeros so that all categories will be represented in the table.
Arguments
---------
table : Results table of catchment summarizations
LandscapeLayer : string to file holding the table of inter VPU COMIDs
'''
# Get ALL categorical values from the dbf associated with the raster to retain all values
# in the raster in every table, even when a given value doesn't exist in a given hydroregion
AllCols = dbf2DF(LandscapeLayer + '.vat.dbf').VALUE.tolist()
col_list = table.columns.tolist()
col_list.sort()
col_list.sort(key=len) # table.columns
table = table[col_list]
if len(AllCols) != len(col_list[1:]):
AllCols = ['VALUE_'+str(x) for x in AllCols]
diff = list(set(AllCols) - set(col_list[1:]))
diff.sort()
diff.sort(key=len)
for spot in diff:
here = AllCols.index(spot) + 1
table.insert(here, spot ,0)
return table
##############################################################################
def getOnNetLakes2(metric, StreamCat, LakeComs, npy_files, nhd):
'''
__author__ = "Rick Debbout <[email protected]>"
Grabs records from StreamCat for on-network lakes. Adjusts
cat results to be an accumulated result of all associated catchments
to the waterbody comid.
Arguments
---------
metric : Metric name
StreamCat : Location of intermediate StreamCat csv files
LakeComs : Location of csv's that join waterbody COMID to catchment COMID
npy_files : Location of files that associate all catchments with WBAREACOMID
'''
final = pd.DataFrame()
for zone in inputs:
tbl = pd.read_csv('%s/join_%s.csv' % (LakeComs,zone))[['catCOMID','wbCOMID']] # remove
strmCat = pd.read_csv('%s/%s_%s.csv' % (StreamCat, metric, zone))
if metric == 'RdCrs':
strmCat = strmCat.drop([x for x in strmCat.columns if 'SlpWtd' in x], axis=1)
# if metric == 'Elev':
# strmCat = strmCat.drop([x for x in strmCat.columns if 'MAX' in x or 'MIN' in x], axis=1)
cols = [col for col in strmCat.columns if col[:3] =='Cat']
iso = strmCat[['COMID'] + cols]
accum = np.load(npy_files)['vpus'].item()[zone]
accumCats = Accumulation(iso, accum['comids'],
accum['lengths'],
accum['upstream'],
'', 'COMID')
# # shouldn't be needed if keep tbl_type arg as empty string in Accumulation
# accumCats.columns = [col.replace('Ws','Cat') for col in accumCats.columns]
upWs = strmCat.ix[strmCat.COMID.isin(tbl.catCOMID)].drop(cols, axis=1)
newCats = pd.merge(accumCats, upWs, on="COMID")
tbl2 = pd.merge(tbl, newCats, left_on='catCOMID', right_on='COMID')
tbl2 = tbl2.drop(['COMID','catCOMID'], axis=1)
tbl2.rename(columns={'wbCOMID': 'COMID'}, inplace=True)
final = pd.concat([final, tbl2])
return final
##############################################################################
def getOnNetLakes(metric, StreamCat, LakeComs):
"""Using the LakeCat join tables linking waterbody COMID to catchment COMID,
Streamcat ouput is collected and manipulated to accumulate catchment stats
for lakes that have more than one catchment associated through the
WBAREACOMI attribute in NHD flowlines.
Arguments
---------
metric : name of the metric to collect data for in StreamCat
StreamCat : location of StreamCat outputs(AllocationandAccumulation)
LakeComs : location of join tables in LakeCat
"""
final = pd.DataFrame()
for zone in inputs:
tbl = pd.read_csv('%s/LakeCOMs%s.csv' % (LakeComs,zone))
strmCat = pd.read_csv('%s/%s_%s.csv' % (StreamCat, metric, zone))
tbl2 = pd.merge(tbl, strmCat.ix[strmCat.COMID.isin(tbl.catCOMID)], left_on='catCOMID', right_on='COMID')
tbl2 = tbl2.drop(['COMID','catCOMID'], axis=1)
if metric == 'RdCrs':
tbl2 = tbl2.drop([x for x in tbl2.columns if 'SlpWtd' in x], axis=1)
if metric == 'Elev':
tbl2 = tbl2.drop([x for x in tbl2.columns if 'MAX' in x or 'MIN' in x], axis=1)
tbl2.rename(columns={'wbCOMID': 'COMID'}, inplace=True)
final = pd.concat([final, tbl2])
return final
##############################################################################
def makeRat(fn):
'''
__author__ = "Matt Gregory <[email protected] >"
Adds a Raster Attribute Table to the .tif.aux.xml file, then passes those
values to rat_to_df function to return the RAT in a pandas DataFrame.
Arguments
---------
fn : raster filename
'''
ds = gdal.Open(fn)
rb = ds.GetRasterBand(1)
nd = rb.GetNoDataValue()
data = rb.ReadAsArray()
# Get unique values in the band and return counts for COUNT val
u = np.array(np.unique(data, return_counts=True))
# remove NoData value
u = np.delete(u, np.argwhere(u==nd), axis=1)
# Create and populate the RAT
rat = gdal.RasterAttributeTable()
rat.CreateColumn('Value', gdal.GFT_Integer, gdal.GFU_Generic)
rat.CreateColumn('Count', gdal.GFT_Integer, gdal.GFU_Generic)
for i in range(u[0].size):
rat.SetValueAsInt(i, 0, int(u[0][i]))
rat.SetValueAsInt(i, 1, int(u[1][i]))
# Associate with the band
rb.SetDefaultRAT(rat)
# Close the dataset and persist the RAT
ds = None
#return the rat to build DataFrame
df = rat_to_df(rat)
return df
##############################################################################
def rat_to_df(in_rat):
"""
__author__ = "Matt Gregory <[email protected] >"
Given a GDAL raster attribute table, convert to a pandas DataFrame
Parameters
----------
in_rat : gdal.RasterAttributeTable
The input raster attribute table
Returns
-------
df : pd.DataFrame
The output data frame
"""
# Read in each column from the RAT and convert it to a series infering
# data type automatically
s = [pd.Series(in_rat.ReadAsArray(i), name=in_rat.GetNameOfCol(i))
for i in xrange(in_rat.GetColumnCount())]
# Concatenate all series together into a dataframe and return
return pd.concat(s, axis=1)
##############################################################################
def DF2dbf(df, dbf_path, my_specs=None):
'''
Convert a pandas.DataFrame into a dbf.
__author__ = "Dani Arribas-Bel <[email protected]> "
...
Arguments
---------
df : DataFrame
Pandas dataframe object to be entirely written out to a dbf
dbf_path : str
Path to the output dbf. It is also returned by the function
my_specs : list
List with the field_specs to use for each column.
Defaults to None and applies the following scheme:
* int: ('N', 14, 0)
* float: ('N', 14, 14)
* str: ('C', 14, 0)
'''
if my_specs:
specs = my_specs
else:
type2spec = {int: ('N', 20, 0),
np.int64: ('N', 20, 0),
float: ('N', 36, 15),
np.float64: ('N', 36, 15),
str: ('C', 14, 0),
np.int32: ('N', 14, 0)
}
types = [type(df[i].iloc[0]) for i in df.columns]
specs = [type2spec[t] for t in types]
db = ps.open(dbf_path, 'w')
db.header = list(df.columns)
db.field_spec = specs
for i, row in df.T.iteritems():
db.write(row)
db.close()
return dbf_path
##############################################################################
def purge(directory, pattern):
'''
__author__ = "Rick Debbout <[email protected]>"
Clears directory of created rasters that will need to be re-written due to
holding on-network like properties, i.e basins created are larger than the
associated catchment.
Arguments
---------
directory : directory to be cleared
pattern : string value to find in filename for removal
'''
for f in os.listdir(directory):
if re.search(pattern, f):
os.remove(os.path.join(directory, f))
##############################################################################
def updateSinks(wbDF,flDF):
'''
__author__ = "Rick Debbout <[email protected]>"
Updates the WBARECOMI field in the NHDFlowline GeoDatFrame where NHDSinks
intersect with NHDWaterbodies. Not currently held in the NHDPlusV21, but
we can process the waterbodies with our on-network approach.
Arguments
---------
wbDF : Metric name
flDF : Location of intermediate StreamCat csv files
'''
flow = flDF.set_index('COMID')
sinks = wbDF.ix[wbDF.COMID_sink.notnull()].copy()
sinks.rename(columns={'COMID': 'WBAREACOMI'}, inplace=True)
flow.update(sinks)
flow.WBAREACOMI = flow.WBAREACOMI.astype(np.int64)
return flow.reset_index(level=0)
##############################################################################
def rollArray(a, d):
if len(d) == 4:
a = a[0,:]
new = np.roll(np.roll(a, d[0], axis=0), d[1], axis=1)
new[d[2],:] = a[d[2],:]
new[:, d[3]] = a[:, d[3]]
if len(d) == 3:
new = np.roll(a[0,:], d[0], axis=d[1])
if d[1] == 0:
new[d[2],:] = a[0,d[2],:]
if d[1] == 1:
new[:,d[2]] = a[0,:,d[2]]
return np.expand_dims(new, axis=0)
##############################################################################
def makeFlows(arr, shiftd, fdr, path, nd):
iso = np.not_equal(arr, shiftd) * np.not_equal(shiftd, nd) * np.not_equal(arr, nd) # cells change value after shift * cells not equal to NoData
pth = np.equal(fdr,path) # True when equal to path value
val = iso * pth * arr
shiftval = iso * pth * shiftd
idx = np.not_equal(val,shiftd)
fromcom = val[idx]
tocom = shiftval[idx]
fromcom = fromcom[fromcom > 0]
tocom = tocom[tocom > 0]
# don't load-in the entire array to the DF, just connection vals
df = pd.DataFrame({'TOCOMID' : tocom,
'FROMCOMID' : fromcom,
'move' : path})
return df.drop_duplicates(['FROMCOMID','TOCOMID'])
##############################################################################
def compAll(arr, fdr ,moves, from_to, nd):
for move in moves:
flow = makeFlows(arr, rollArray(np.copy(arr), moves[move][0]), fdr, moves[move][1], nd)
from_to = pd.concat([from_to,flow])
return from_to
##############################################################################
def expand(window, size=1):
r, c = window
return ((r[0] - size, r[1] + size), (c[0] - size, c[1] + size))
def check_window(window, w, h):
r, c = window
return ((max(0, r[0]), min(h, r[1])), (max(0, c[0]), min(w, c[1])))
##############################################################################
def chunk_windows(r, indexes=None, max_ram=250000000):
if indexes is None:
indexes = r.indexes
elif isinstance(indexes, int):
indexes = [indexes]
if not indexes:
raise ValueError('No indexes to read')
pixel_size = 0
for bidx in indexes:
if bidx not in r.indexes:
raise IndexError('band index out of range')
idx = r.indexes.index(bidx)
pixel_size += np.dtype(r.dtypes[idx]).itemsize
chunk_size, _ = divmod(max_ram, pixel_size)
r_h, r_w = r.height, r.width
if chunk_size >= r_h * r_w:
yield (0, 0), ((0, r_h), (0, r_w))
else:
b_h, b_w = r.block_shapes[0]
d, _ = divmod(chunk_size, r_w * b_h)
chunk_height = d * b_h
d, m = divmod(r_h, chunk_height)
n_chunks = d + int(m>0)
for i in range(n_chunks):
row = i * chunk_height
# height = min(chunk_height, r_h - row)
yield (i, 0), ((row, row+chunk_height), (0, r_w))
##############################################################################
def findFlows(zone_file, fdr_file):
moves = {'up':[(-1,0,-1),4],'left':[(-1,1,-1),1],'down' :[(1,0,0),64],
'right':[(1,1,0),16],'downRight':[(1,1,0,0),32],
'downLeft':[(1,-1,0,-1), 128],'upRight':[(-1,1,-1,0),8],
'upLeft':[(-1,-1,-1,-1),2]}
flows = pd.DataFrame()
with rs.open(zone_file) as z:
with rs.open(fdr_file) as f: # 0 is NoData for fdr
profile = z.profile.copy()
nd = profile['nodata']
assert z.shape == f.shape, "Rasters have different extents!"
for _, w in chunk_windows(z): # currently defaults to 250MB
new_w = check_window(expand(w,2), z.width, z.height)
data = z.read(window=new_w)
f_r = f.read(window=new_w)
flows = pd.concat([flows,compAll(data,f_r,moves,flows,nd)])
return flows.drop_duplicates(['FROMCOMID','TOCOMID'])
##############################################################################
def NHDtblMerge(nhd, bounds, out):
'''
__author__ = "Rick Debbout <[email protected]>"
Merges all of the NHD tables needed to find on-network lakes. Returns the
GeoDataFrames that will be used to find off-network lakes. Attribute fields
COMID, WBARECOMI, and FEATUREID are used to link waterbodies to catchments.
Arguments
---------
nhd : string value of prefix to NHD directory
bounds : GeoDataFrame of Vector Processing Unit boundaries
out : directory to write out to
'''
# build dict of hr/vpu labels to read through NHD
vpus = bounds.query("UnitType == 'VPU'").copy()
# initialize containers to append to through processing
onNet_connect = {}
Obounds = gpd.GeoDataFrame(crs={'init': u'epsg:4269'})
qa_cols=['Total Waterbodies','On-Network','Off-network','FTYPE_drop',
'Sink_add','Out_of_bounds']
qa_tbl = pd.DataFrame()
ons = []
for zone in inputs:
print zone
hr = inputs[zone]
pre = "%s/NHDPlus%s/NHDPlus%s" % (nhd, hr, zone)
wbShp = gpd.read_file("%s/NHDSnapshot/Hydrography/NHDWaterbody.shp"%(pre))
# hold length of total Waterbodies
ttl_WB = len(wbShp)
# format columns and select out FTYPE
wbShp.columns = wbShp.columns[:-1].str.upper().tolist() + ['geometry']
wbShp = wbShp[['AREASQKM','COMID','FTYPE','geometry']]
wbShp = wbShp.loc[wbShp['FTYPE'].isin(['LakePond','Reservoir'])]
# hold number of lakes removed from FTYPE
ttl_FTYPE = ttl_WB - len(wbShp)
fl = dbf2DF("%s/NHDSnapshot/Hydrography/NHDFlowline.dbf"%(pre))[['COMID',
'WBAREACOMI']]
cat = gpd.read_file('%s/NHDPlusCatchment/Catchment.shp'%(pre)).drop(
['GRIDCODE', 'SOURCEFC'], axis=1)
cat.columns = cat.columns[:-1].str.upper().tolist() + ['geometry']
vaa = dbf2DF('%s/NHDPlusAttributes/PlusFlowlineVAA.dbf'%(pre))[['COMID',
'HYDROSEQ']]
# merge all necessary NHD tables
final = pd.merge(cat.drop('geometry', axis=1),fl,left_on='FEATUREID',
right_on='COMID',how='inner')
final = pd.merge(wbShp.drop('geometry',axis=1),final,left_on='COMID',
right_on='WBAREACOMI',how='left',
suffixes=('_wb','_cat'))
final = pd.merge(final,vaa,left_on='COMID_cat',
right_on='COMID',how='left')
# initialize containers for on-net lakes
cols = ['catCOMID','wbCOMID','CatAreaSqKm']
onNetDF = pd.DataFrame(columns=cols)
catDict = {} # holds associated lake catchments to an on-network lake
# group by the waterbody COMID to find associated catchment
for name, group in final.groupby('COMID_wb'):
if not pd.isnull(group.FEATUREID).any():
base = group.ix[group.HYDROSEQ.idxmin()]
row = pd.Series([int(base.COMID_cat), int(base.COMID_wb),
base.AREASQKM_cat], index=cols)
onNetDF = onNetDF.append(row, ignore_index=True)
catDict[int(base.COMID_cat)] = group.FEATUREID.astype(int).tolist()
# hold length of on-net lakes
ttl_ON = len(onNetDF)
# add in related sinks -- this could be done in tables and found in
# groupby object above, but this allows for isolation of number added
sinks = gpd.read_file("%s/NHDPlusBurnComponents/Sink.shp"%(pre))
exp = '(SOURCEFC== "NHDWaterbody")&(PURPDESC== "NHDWaterbody closed lake")'
if len(sinks) > 0:
sinks = sinks.query(exp)
try:
assert len(sinks) > 0
catSink = sjoin(sinks, cat)
catSink = catSink[['FEATUREID_right',
'FEATUREID_left',
'AREASQKM']]
catSink.columns = cols
catSink = catSink.ix[catSink.wbCOMID.isin(wbShp.COMID)] #this will remove any NHDWaterbody COMIDs that have the wrong FTYPE
catSink = catSink.ix[~catSink.wbCOMID.isin(onNetDF.wbCOMID)] #remove any COMIDs that are already in the onNetDF
ttl_SINK = len(catSink)
for idx, line in catSink.iterrows():
catDict[line.catCOMID] = [line.wbCOMID]
onNetDF = pd.concat([onNetDF,catSink])
except AssertionError:
ttl_SINK = 0 # all sinks got queried out!
pass
else:
ttl_SINK = len(sinks) # get val if no sinks for QA
# create numpy arrays for connected catchments to waterbody
allLinkd = map(lambda x: catDict[x], catDict.keys())
onNet_connect[zone] = {'comids':np.array(catDict.keys()),
'lengths':np.array([len(v) for v in allLinkd]),
'upstream':np.int32(np.hstack(np.array(allLinkd)))}
# write-out table of catchment-lake COMID connections
onNetDF.to_csv("%s/joinTables/join_%s.csv" % (out, zone), index=False)
offLks = wbShp.ix[~wbShp.COMID.isin(onNetDF.wbCOMID)].copy()
ons = ons + onNetDF.wbCOMID.tolist()
# find off-netowrk lakes that are out-of-bounds
vpu = vpus.query("UnitID == '%s'" % zone)
offCen = offLks.copy()
offCen.geometry = offLks.geometry.centroid
# find centroids within the vpu
lkVPUjoin = sjoin(offCen, vpu, op='within')[['AREASQKM','COMID','FTYPE',
'UnitID','geometry']]
# hold lakes that aren't within the VPU boundary
out_of_bounds = offLks.ix[~offLks.COMID.isin(lkVPUjoin.COMID)].copy()
out_of_bounds['VPU_orig'] = zone # identify the zone it came from
# find the correct vpu for those out-of-bounds
outCen = offCen.ix[~offCen.COMID.isin(lkVPUjoin.COMID)]
unit = sjoin(outCen, vpus, op='within')[['COMID','UnitID']]
out_of_bounds = out_of_bounds.merge(unit, how='left', on='COMID')
# add out-of-bounds to GeoDF to hold all, and select only lakes within
# the vpu
Obounds = pd.concat([Obounds,out_of_bounds])
# Obounds = gpd.GeoDataFrame( pd.concat([Obounds,out_of_bounds],
# ignore_index=True) )
offLks = offLks.ix[offLks.COMID.isin(lkVPUjoin.COMID)].copy()
ttl_OOB = len(out_of_bounds)
ttl_OFF = len(offLks)
# add VPU info to offLks table
offLks['VPU_orig'] = zone
vpu_tbl = sjoin(offCen, vpus, op='within')[['COMID','UnitID']]
offLks = offLks.merge(vpu_tbl, on='COMID', how='left')
# write-out off-net lakes and add series of QA info to DF
offLks.to_file("%s/off_net_%s.shp" % (out, zone))
qa_tbl[zone] = [ttl_WB,ttl_ON,ttl_OFF,ttl_FTYPE,ttl_SINK,ttl_OOB]
# write-out all zone DF's and the numpy files created to
assert Obounds.crs == {'init': u'epsg:4269'}
# remove lakes out of bounds that would be off-network from that zone, but
# were established as on-net in the zone they exist
# 02/04 : coms 15516920, 15516922 NHD Problem.....again
Obounds = Obounds.ix[~Obounds.COMID.isin(ons)]
Obounds.to_file("%s/out_of_bounds.shp" % out)
np.savez_compressed('%s/LakeCat_npy/onNet_LakeCat.npz' % (out),
vpus=onNet_connect)
qa_tbl.index = qa_cols
qa_tbl.T.index.rename('VPU', inplace=True)
qa_tbl['TOTALS'] = qa_tbl.ix[:].sum(axis=1)
qa_tbl.T.to_csv("%s/Lake_QA.csv" % out)
##############################################################################
def makeBasins(nhd, bounds, out):
'''
__author__ = "Rick Debbout <[email protected]>"
Makes GeoDataFrame of all lakes within each raster processing unit, converts
each to raster, and then draws watershed basins for each lake identified as
off-network. Creates flow table for all adjacent lake basin boundaries to
identify zones that have hydrological connectivity.
Arguments
---------
nhd : string value of prefix to NHD directory
bounds : GeoDataFrame of Vector Processing Unit boundaries
out : directory to write out to
'''
problems = pd.DataFrame() # holding for overdrawn basin delineations,
# i.e. bigger watershed than respective catchment
allOff = gpd.GeoDataFrame(crs={'init': u'epsg:4269'}) # concat for all lakes
allBsns = gpd.GeoDataFrame() # concat for shpfile of all basins(PointInPoly)
runit = bounds.query("UnitType == 'RPU'").copy() # GeoDF of RPU bounds only
Obounds = gpd.read_file("%s/out_of_bounds.shp" % out) # lakes found out of
# their respective hydroregion
arcpy.env.workspace = "%s/rasters/lakes/scratchArc" % out
arcpy.env.outputCoordinateSystem = fiftyseventy # output crs in ESRI WKT
cols = ['VPU','out_of_raster']
addOut = pd.DataFrame(columns=cols) # DF to hold no. of lakes not in raster
# for QA table
#countTbl = pd.DataFrame() # concat for all RAT info into csv
flow_tbl = pd.DataFrame() # concat for all flow tables into one csv
uid = 1000
for zone in inputs:
print zone
hr = inputs[zone]
pre = "%s/NHDPlus%s/NHDPlus%s" % (nhd, hr, zone)
# get the lakes that were out-of-bounds into the correct vpu
addLks = Obounds.ix[Obounds.UnitID == zone].copy()
offLks = gpd.read_file("%s/off_net_%s.shp" % (out, zone))
# remove duplicated lakes across zones
addLks.drop(addLks.loc[addLks.COMID.isin(offLks.COMID)].index,inplace=True)
# add back-in lakes that are in other zones
offLks = pd.concat([offLks,addLks]).reset_index(drop=True)
assert offLks.crs == {'init': u'epsg:4269'}
offLks.rename(columns={'UnitID':'VPU_moved'}, inplace=True)
ttl_LOST = 0
cat = gpd.read_file('%s/NHDPlusCatchment/Catchment.shp'%(pre))
for rpu in rpus[zone]:
lakes = offLks.copy()
if len(rpus[zone]) > 1:
rpuShp = runit.query("UnitID == '%s'" % rpu).drop(['Hydroseq',
'UnitType'],
axis=1)
lakes = sjoin(lakes, rpuShp, op='within') #this line removes lakes that straddle RPU borders
lakes = lakes.drop('index_right',axis=1)
lakes.rename(columns={'UnitID': 'RPU'}, inplace=True)
if len(rpus[zone]) == 1:
lakes['RPU'] = rpu
lakes.drop_duplicates('COMID', inplace=True) # filter out duplicated
ln = len(lakes)
lakes['UID'] = range(uid,uid+ln)
uid += ln
lakes.to_file('%s/rasters/lakes/scratchArc/rasPrep_%s.shp' % (out, rpu))
fdr = arcpy.sa.Raster("%s/NHDPlusFdrFac%s/fdr" % (pre, rpu))
arcpy.env.extent = fdr.extent
arcpy.env.snapRaster = "%s/NHDPlusFdrFac%s/fdr" % (pre, rpu)
p2r('%s/rasters/lakes/scratchArc/rasPrep_%s.shp' % (out, rpu),'UID',
"%s/rasters/lakes/lakes_%s.tif" % (out, rpu),"CELL_CENTER",'',30)
purge('%s/rasters/lakes/scratchArc' % out, 'rasPrep_%s' % rpu)
outWshed = Watershed("%s/NHDPlusFdrNull%s/fdrnull" % (pre, rpu),