19
LAMPIRAN
20 Lampiran 1. Peta administrasi Riau dan plotting stasiun pengamatan wilayah Riau
Sumber : Badan Koordinasi Survei dan Pemetaan Nasional (Bakosurtanal)
21 Lampiran 2. Proses pengolahan data hujan satelit dari data CMORPH Data CMORPH memberikan gambaran curah hujan estimasi secara global. Informasi numerik tersebut diperoleh dengan mengkonversi informasi hujan format shapefile menjadi format text sehingga akan diperoleh nilai curah hujan yang diinginkan. Berikut ini langkah pengolahan data hujan satelit dari data CMORPH (Oktavariani 2008) : 1. Data CMORPH yang sudah diekstrak dari format zip, kemudian dibuka menggunakan Arcview 3.3 dengan mengaktifkan extention 3D-Analysi, Grid Analyst, dan Spatial Analyst. 2. Open basemap Indonesia, kemudian cropping wilayah kajian – save – ok. 3. Open data CMORPH. 4. Drid Analyst – extract grid theme using polygon – pilih hasil cropping pada langkah sebelumnya. 5. Aktifkan theme hasil extract – convert grid theme to XYZ text file. Maka akan diperoleh informasi hujan global sesuai dengan koordinat lintang dan bujur yang tersimpan dalam file .txt.
22 Lampiran 3. Uji dua regresi Stasiun Pekanbaru 2.1 Musim Hujan
Regression Analysis: Y versus X The regression equation is Y = 0.960 X Predictor Noconstant X
Coef
SE Coef
T
P
0.96003
0.05279
18.19
0.000
S = 60.0985 Analysis of Variance Source DF SS Regression 1 1194507 Residual Error 107 386466 Total 108 1580973
MS 1194507 3612
F 330.72
P 0.000
2.2 Musim Kemarau
Regression Analysis: Y versus X The regression equation is Y = 1.08 X Predictor Noconstant X
Coef
SE Coef
T
P
1.07893
0.07389
14.60
0.000
MS 591987 2776
F 213.24
S = 52.6893 Analysis of Variance Source DF SS Regression 1 591987 Residual Error 107 297050 Total 108 889037
P 0.000
Z = 0.8261 dengan α = 0.7967, dimana α > taraf nyata sehingga persamaan musim hujan dan musim kemarau tidak berbeda nyata.
23 Lampiran 4. Uji dua regresi Stasiun Japura Rengat 3.1 Musim Hujan
Regression Analysis: Y versus X The regression equation is Y = 0.495 X Predictor Noconstant X
Coef
SE Coef
T
P
0.49482
0.03271
15.13
0.000
MS 239077 1045
F 228.86
S = 32.3208 Analysis of Variance Source DF SS Regression 1 239077 Residual Error 107 111776 Total 108 350853
P 0.000
3.2 Musim Kemarau
Regression Analysis: Y versus X The regression equation is Y = 0.595 X Predictor Noconstant X
Coef
SE Coef
T
P
0.59536
0.05101
11.67
0.000
MS 122398 898
F 136.24
S = 29.9739 Analysis of Variance Source DF SS Regression 1 122398 Residual Error 107 96132 Total 108 218531
P 0.000
Z = 0.7392 dengan α = 0.7704, dimana α > taraf nyata sehingga persamaan musim hujan dan musim kemarau tidak berbeda nyata.
24 Lampiran 5. Uji dua regresi Stasiun Tanjung Pinang 4.1 Musim Hujan
Regression Analysis: Y versus X The regression equation is Y = 0.826 X Predictor Noconstant X
Coef
SE Coef
T
P
0.8259
0.1009
8.19
0.000
S = 53.8562 Analysis of Variance Source DF SS Regression 1 194322 Residual Error 107 310352 Total 108 504675
MS 194322 2900
F 67.00
P 0.000
4.2 Musim Kemarau
Regression Analysis: Y versus X The regression equation is Y = 1.36 X Predictor Noconstant X
Coef
SE Coef
T
P
1.35802
0.09341
14.54
0.000
MS 316800 1499
F 211.37
S = 38.7146 Analysis of Variance Source DF SS Regression 1 316800 Residual Error 107 160374 Total 108 477174
P 0.000
Z = 6.1702 dengan α = 0.9997, dimana α > taraf nyata sehingga persamaan musim hujan dan musim kemarau tidak berbeda nyata.
25 Lampiran 6. Uji dua regresi Stasiun Dabo Singkep 5.1 Musim Hujan
Regression Analysis: Y versus X The regression equation is Y = 0.799 X Predictor Noconstant X
Coef
SE Coef
T
P
0.79933
0.04956
16.13
0.000
MS 497924 1914
F 260.11
S = 43.7527 Analysis of Variance Source DF SS Regression 1 497924 Residual Error 107 204830 Total 108 702754
P 0.000
5.2 Musim Kemarau
Regression Analysis: Y versus X The regression equation is Y = 1.15 X Predictor Noconstant X
Coef
SE Coef
T
P
1.14893
0.06827
16.83
0.000
SS 496658 187631 684289
MS 496658 1754
F 283.23
S = 41.8755 Analysis of Variance Source Regression Residual Error Total
DF 1 107 108
P 0.000
Z = 2.5661 dengan α = 0.9949, dimana α > taraf nyata sehingga persamaan musim hujan dan musim kemarau tidak berbeda nyata.
26 Lampiran 7. Contoh hasil keluaran VIF (Variation Inflation Factor) Stasiun Pekanbaru
•
Domain 3x3
Regression Analysis: X1 versus X2, X3, X4, X5, X6, X7, X8, X9 The regression equation is X1 = 1.45 + 1.24 X2 - 0.450 X3 + 0.658 X4 - 0.680 X5 + 0.231 X6 0.0625 X7 - 0.0212 X8 + 0.0784 X9 Predictor Constant X2 X3 X4 X5 X6 X7 X8 X9
Coef 1.451 1.24220 -0.45022 0.65843 -0.6796 0.23121 -0.06250 -0.02120 0.07844
S = 12.9878
SE Coef 1.749 0.06916 0.06872 0.08441 0.1015 0.09228 0.07583 0.09154 0.07125
R-Sq = 94.7%
Analysis of Variance Source DF SS Regression 8 517066 Residual Error 171 28845 Total 179 545911
T 0.83 17.96 -6.55 7.80 -6.70 2.51 -0.82 -0.23 1.10
P 0.408 0.000 0.000 0.000 0.000 0.013 0.411 0.817 0.272
VIF 14.2 12.8 22.9 31.4 23.3 20.4 25.2 13.1
R-Sq(adj) = 94.5%
MS 64633 169
F 383.16
P 0.000
27 Lampiran 8. Contoh hasil keluaran VIF (Variation Inflation Factor) Stasiun Japura Rengat •
Domain 3x3
Regression Analysis: X1 versus X2, X3, X4, X5, X6, X7, X8, X9 The regression equation is X1 = 0.99 + 1.10 X2 - 0.316 X3 + 0.658 X4 - 0.457 X5 + 0.0749 X6 0.174 X7 + 0.261 X8 - 0.149 X9 Predictor Constant X2 X3 X4 X5 X6 X7 X8 X9
Coef 0.995 1.10488 -0.31624 0.65814 -0.4569 0.07490 -0.17449 0.26086 -0.14878
S = 12.7578
SE Coef 1.663 0.08217 0.07824 0.07614 0.1057 0.09448 0.07813 0.09875 0.07621
R-Sq = 94.1%
Analysis of Variance Source DF SS Regression 8 446312 Residual Error 171 27832 Total 179 474144
T 0.60 13.45 -4.04 8.64 -4.32 0.79 -2.23 2.64 -1.95
P 0.551 0.000 0.000 0.000 0.000 0.429 0.027 0.009 0.053
VIF 15.2 12.0 17.7 28.7 19.1 20.7 29.6 15.0
R-Sq(adj) = 93.9%
MS 55789 163
F 342.76
P 0.000
28 Lampiran 9. Contoh hasil keluaran VIF (Variation Inflation Factor) Stasiun Tanjung Pinang •
Domain 3x3
Regression Analysis: X1 versus X2, X3, X4, X5, X6, X7, X8, X9 The regression equation is X1 = 1.05 + 1.24 X2 - 0.317 X3 + 0.353 X4 - 0.501 X5 + 0.088 X6 + 0.218 X7 - 0.022 X8 - 0.0578 X9 Predictor Constant X2 X3 X4 X5 X6 X7 X8 X9
Coef 1.0506 1.23808 -0.31743 0.35316 -0.5011 0.0883 0.21799 -0.0220 -0.05778
S = 6.23610
SE Coef 0.6054 0.06971 0.07132 0.07571 0.1220 0.1030 0.07940 0.1154 0.08039
R-Sq = 96.5%
Analysis of Variance Source DF SS Regression 8 185294 Residual Error 171 6650 Total 179 191944
T 1.74 17.76 -4.45 4.66 -4.11 0.86 2.75 -0.19 -0.72
P 0.084 0.000 0.000 0.000 0.000 0.392 0.007 0.849 0.473
VIF 26.0 31.9 29.1 84.5 72.5 31.1 78.8 45.9
R-Sq(adj) = 96.4%
MS 23162 39
F 595.59
P 0.000
29 Lampiran 10. Contoh hasil keluaran VIF (Variation Inflation Factor) Stasiun Dabo Singkep •
Domain 3x3
Regression Analysis: X1 versus X2, X3, X4, X5, X6, X7, X8, X9 The regression equation is X1 = 0.455 + 1.03 X2 - 0.279 X3 + 0.594 X4 - 0.345 X5 - 0.0373 X6 0.0876 X7 - 0.128 X8 + 0.213 X9 Predictor Constant X2 X3 X4 X5 X6 X7 X8 X9
Coef 0.4545 1.02938 -0.27896 0.59405 -0.34537 -0.03726 -0.08762 -0.12783 0.21333
S = 8.22001
SE Coef 0.8934 0.07334 0.06156 0.05529 0.08517 0.05928 0.06275 0.09386 0.05913
R-Sq = 95.9%
Analysis of Variance Source DF SS Regression 8 270757 Residual Error 171 11554 Total 179 282311
T 0.51 14.04 -4.53 10.74 -4.06 -0.63 -1.40 -1.36 3.61
P 0.612 0.000 0.000 0.000 0.000 0.530 0.164 0.175 0.000
VIF 29.4 30.5 16.7 51.0 28.9 25.0 62.2 26.0
R-Sq(adj) = 95.7%
MS 33845 68
F 500.89
P 0.000
30 Lampiran 11. Partial Least Square (PLS) Stasiun Pekanbaru a. Domain 3x3
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, X9 Number of components specified: 5 Analysis of Variance Source DF Regression 5 Residual Error 174 Total 179
for Y SS 399225 442882 842107
MS 79845.0 2545.3
F 31.37
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.860830 502282 0.403541 2 0.896507 453751 0.461172 3 0.954814 450520 0.465008 4 0.973081 445720 0.470709 5 0.982214 442882 0.474079
b.
Domain 5x5
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 5 Analysis of Variance for Y Source DF SS Regression 5 3682367 Residual Error 174 579645 Total 179 4262012
MS 736473 3331
F 221.08
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.840239 942939 0.778757 2 0.945897 799293 0.812461 3 0.957915 636876 0.850569 4 0.967756 600215 0.859171 5 0.982104 579645 0.863997
c.
Domain 7x7
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 3 Analysis of Variance for Y Source DF SS Regression 3 41562899 Residual Error 176 538037 Total 179 42100936
MS 13854300 3057
F 4531.95
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.905286 1280752 0.969579 2 0.970463 571409 0.986428 3 0.975203 538037 0.987220
P 0.000
31 d.
Domain 9x9
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 11 Analysis of Variance Source DF Regression 11 Residual Error 168 Total 179
for Y SS 574214 267893 842107
MS 52201.2 1594.6
F 32.74
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.621955 547959 0.349300 2 0.702334 487081 0.421592 3 0.755631 438254 0.479575 4 0.792673 400086 0.524899 5 0.821672 370721 0.559769 6 0.830362 329533 0.608680 7 0.845836 308871 0.633216 8 0.863729 296217 0.648242 9 0.880338 286275 0.660048 10 0.890059 274865 0.673598 11 0.900857 267893 0.681878
P 0.000
32 Lampiran 12. Partial Least Square (PLS) Stasiun Japura Rengat a. Domain 3x3
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, X9 Number of components specified: 5 Analysis of Variance for Y Source DF SS Regression 5 66772 Residual Error 174 152053 Total 179 218824
MS 13354.3 873.9
F 15.28
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.839062 164460 0.248437 2 0.886706 160725 0.265506 3 0.902245 155724 0.288359 4 0.954468 154428 0.294281 5 0.978373 152053 0.305138
b.
Domain 5x5
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 4 Analysis of Variance for Y Source DF SS Regression 4 9043314 Residual Error 175 327257 Total 179 9370570
MS 2260828 1870
F 1208.97
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.828310 1685862 0.820090 2 0.944038 650152 0.930618 3 0.964630 352851 0.962345 4 0.980464 327257 0.965076 5 0.984394 293037 0.968728
c.
Domain 7x7
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 3 Analysis of Variance for Y Source DF SS Regression 3 27410368 Residual Error 176 349095 Total 179 27759464
MS 9136789 1983
F 4606.41
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.830031 1975675 0.928829 2 0.945332 462319 0.983346 3 0.951652 349095 0.987424
P 0.000
33 d.
Domain 9x9
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 12 Analysis of Variance Source DF Regression 12 Residual Error 167 Total 179
for Y SS 116608 102216 218824
MS 9717.36 612.07
F 15.88
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.609809 162236 0.258602 2 0.690525 154449 0.294184 3 0.728496 143092 0.346086 4 0.753411 136839 0.374664 5 0.767391 128869 0.411085 6 0.787903 124107 0.432845 7 0.805584 118179 0.459937 8 0.855198 116123 0.469330 9 0.871852 111811 0.489036 10 0.884067 107676 0.507934 11 0.895979 104446 0.522692 12 0.908082 102216 0.532886
P 0.000
34 Lampiran 13. Partial Least Square (PLS) Stasiun Tanjung Pinang a. Domain 3x3
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, X9 Number of components specified: 4 Analysis of Variance for Y Source DF SS Regression 4 83834 Residual Error 175 270834 Total 179 354668
MS 20958.4 1547.6
F 13.54
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.930431 286216 0.193004 2 0.939306 272444 0.231834 3 0.958772 271190 0.235371 4 0.984754 270834 0.236372
b.
Domain 5x5
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 5 Analysis of Variance for Y Source DF SS Regression 5 812309 Residual Error 174 280265 Total 179 1092574
MS 162462 1611
F 100.86
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.815661 552145 0.494638 2 0.946064 349036 0.680538 3 0.954771 309863 0.716392 4 0.967711 294457 0.730492 5 0.977193 280265 0.743482
c.
Domain 7x7
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 5 Analysis of Variance for Y Source DF SS Regression 5 2330496 Residual Error 174 309621 Total 179 2640117
MS 466099 1779
F 261.94
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.779742 828029 0.686367 2 0.939489 404162 0.846915 3 0.949731 344351 0.869570 4 0.964833 321995 0.878038 5 0.971654 309621 0.882725
P 0.000
35 d.
Domain 9x9
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 11 Analysis of Variance Source DF Regression 11 Residual Error 168 Total 179
for Y SS 176839 177814 354653
MS 16076.3 1058.4
F 15.19
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.753082 285120 0.196058 2 0.793677 262863 0.258816 3 0.837217 247479 0.302195 4 0.860591 233300 0.342172 5 0.903983 226447 0.361496 6 0.919291 210808 0.405594 7 0.931429 203533 0.426105 8 0.935733 192836 0.456268 9 0.941689 186380 0.474473 10 0.947443 181414 0.488473 11 0.952402 177814 0.498626
P 0.000
36 Lampiran 14. Partial Least Square (PLS) Stasiun Dabo Singkep a. Domain 3x3
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, X9 Number of components specified: 5 Analysis of Variance Source DF Regression 5 Residual Error 174 Total 179
for Y SS 324341 225785 550126
MS 64868.1 1297.6
F 49.99
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.904752 300257 0.454203 2 0.940044 243121 0.558063 3 0.970850 234345 0.574016 4 0.981111 228004 0.585541 5 0.988920 225785 0.589575
b.
Domain 5x5
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 5 Analysis of Variance for Y Source DF SS Regression 5 3973367 Residual Error 174 216545 Total 179 4189911
MS 794673 1245
F 638.54
P 0.000
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.794795 635450 0.848338 2 0.950092 292760 0.930128 3 0.969522 262479 0.937355 4 0.976460 230797 0.944916 5 0.981562 216545 0.948318
c.
Domain 7x7
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 3 Analysis of Variance for Y Source DF SS Regression 3 5921141 Residual Error 176 261041 Total 179 6182182
MS 1973714 1483
F 1330.72
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.857442 701592 0.886514 2 0.942271 309220 0.949982 3 0.958812 261041 0.957775
P 0.000
37 d.
Domain 9x9
PLS Regression: Y versus X1, X2, X3, X4, X5, X6, X7, X8, ... Number of components specified: 11 Analysis of Variance Source DF Regression 11 Residual Error 168 Total 179
for Y SS 419302 130824 550126
MS 38118.3 778.7
F 48.95
Model Selection and Validation for Y Components X Variance Error SS R-Sq 1 0.681429 317800 0.422315 2 0.732657 237731 0.567861 3 0.800585 225696 0.589738 4 0.845445 212952 0.612903 5 0.865842 198461 0.639244 6 0.877460 182578 0.668115 7 0.893917 170925 0.689299 8 0.904720 154741 0.718717 9 0.912864 144388 0.737536 10 0.923236 138315 0.748576 11 0.928684 130824 0.762192
P 0.000
38 Lampiran 15. Keragaman curah hujan yang dapat diterangkan oleh setiap komponen berdasarkan metode Partial Least Square (PLS)
Stasiun
Domain
Pekanbaru
3x3 5x5 7x7 9x9
Japura Rengat
3x3 5x5 7x7 9x9
Tanjung Pinang
3x3 5x5 7x7 9x9
Dabo Singkep
3x3 5x5 7x7 9x9
X variance PC1
PC2
PC3
PC4
PC5
PC6
PC7
PC8
PC9
PC10
PC11
0.8608 0.8402 0.9053 0.6220 0.8391 0.8283 0.8300 0.6098 0.9304 0.8157 0.7797 0.7531 0.9048 0.7948 0.8574 0.6814
0.8965 0.9459 0.9705 0.7023 0.8867 0.9440 0.9453 0.6905 0.9393 0.9461 0.9395 0.7937 0.9400 0.9501 0.9423 0.7327
0.9548 0.9579 0.9752 0.7556 0.9022 0.9646 0.9517 0.7285 0.9588 0.9548 0.9497 0.8372 0.9709 0.9695 0.9588 0.8006
0.9731 0.9678
0.9822 0.9821
0.7927 0.9545 0.9805
0.8217 0.9784
0.8304
0.8458
0.8637
0.8803
0.8901
0.9009
0.7534 0.9848 0.9677 0.9648 0.8606 0.9811 0.9765
0.7674
0.7879
0.8056
0.8552
0.8719
0.8841
0.8960
0.9772 0.9717 0.9040 0.9889 0.9816
0.9193
0.9314
0.9357
0.9417
0.9474
0.9524
0.8454
0.8658
0.8775
0.8939
0.9047
0.9129
0.9232
0.9287
PC12
0.9081
38
39 Lampiran 16. Koefisien determinasi berdasarkan metode Partial Least Square (PLS)
Stasiun
Domain
Pekanbaru
3x3 5x5 7x7 9x9
Japura Rengat
3x3 5x5 7x7 9x9
Tanjung Pinang
3x3 5x5 7x7 9x9
Dabo Singkep
3x3 5x5 7x7 9x9
R2 (%) PC1
PC2
PC3
PC4
PC5
PC6
PC7
PC8
PC9
PC10
PC11
40.4 77.9 97.0 34.9 24.8 82.0 92.9 25.9 19.3 49.5 68.6 19.6 45.4 84.8 88.7 42.2
46.1 81.2 98.6 42.2 26.6 93.1 98.3 29.4 23.2 68.1 84.7 25.9 55.8 93.0 95.0 56.8
46.5 85.1 98.7 48.0 28.8 96.2 98.7 34.6 23.5 71.6 87.0 30.2 57.4 93.7 95.8 59.0
47.1 85.9
47.4 86.4
52.5 29.4 96.5
56.0 30.5
60.9
63.3
64.8
66.0
67.4
68.2
37.5 23.6 73.0 87.8 34.2 58.6 94.5
41.1
43.3
46.0
46.9
48.9
50.8
52.3
74.3 88.3 36.1 59.0 94.8
40.6
42.6
45.6
47.4
48.8
49.9
61.3
63.9
66.8
68.9
71.9
73.8
74.9
76.2
PC12
53.3
39
40 Lampiran 17. Regresi sederhana Stasiun Pekanbaru
40
41 Lampiran 18. Regresi sederhana Stasiun Japura Rengat
41
42 Lampiran 19. Regresi sederhana Stasiun Tanjung Pinang
42
43 Lampiran 20. Regresi sederhana Stasiun Dabo Singkep
43
