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

1001

1002

1003

1004

1005

1006

1007

1008

1009

1010

1011

1012

1013

1014

1015

1016

1017

1018

1019

1020

1021

1022

1023

1024

1025

1026

1027

1028

1029

1030

1031

1032

1033

1034

1035

1036

1037

1038

1039

1040

1041

1042

1043

1044

1045

1046

1047

1048

1049

1050

1051

1052

1053

1054

1055

1056

1057

1058

1059

1060

1061

1062

1063

1064

1065

1066

1067

1068

1069

1070

1071

1072

1073

1074

1075

1076

1077

1078

1079

1080

1081

1082

1083

1084

1085

1086

1087

1088

1089

1090

1091

1092

1093

1094

1095

1096

1097

1098

1099

1100

1101

1102

1103

1104

1105

1106

1107

1108

1109

1110

1111

1112

1113

1114

1115

1116

1117

1118

1119

1120

1121

1122

1123

1124

1125

1126

1127

1128

1129

1130

1131

1132

1133

1134

1135

1136

1137

1138

1139

1140

1141

1142

1143

1144

1145

1146

1147

1148

1149

1150

1151

1152

1153

1154

1155

1156

1157

1158

1159

1160

1161

1162

1163

1164

1165

1166

1167

1168

1169

1170

1171

1172

1173

1174

1175

1176

1177

1178

1179

1180

1181

1182

1183

1184

1185

1186

1187

1188

1189

1190

1191

1192

1193

1194

1195

1196

1197

1198

1199

1200

1201

1202

1203

1204

1205

1206

1207

1208

1209

1210

1211

1212

1213

1214

1215

1216

1217

1218

1219

1220

1221

1222

1223

1224

1225

1226

1227

1228

1229

1230

1231

1232

1233

1234

1235

1236

1237

1238

1239

1240

1241

1242

1243

1244

1245

1246

1247

1248

1249

1250

1251

1252

1253

1254

1255

1256

1257

1258

1259

1260

1261

1262

1263

1264

1265

1266

1267

1268

1269

1270

1271

1272

1273

1274

1275

1276

1277

1278

1279

1280

1281

1282

1283

1284

1285

1286

1287

1288

1289

1290

1291

1292

1293

1294

1295

1296

1297

1298

1299

1300

1301

1302

1303

1304

1305

1306

1307

1308

1309

1310

1311

1312

1313

1314

1315

1316

1317

1318

1319

1320

1321

1322

1323

1324

1325

1326

1327

1328

1329

1330

1331

1332

1333

1334

1335

1336

1337

1338

1339

1340

1341

1342

1343

1344

1345

1346

1347

1348

1349

1350

1351

1352

1353

1354

1355

1356

1357

1358

1359

1360

1361

1362

1363

1364

1365

1366

1367

1368

1369

1370

1371

1372

1373

1374

1375

1376

1377

1378

1379

1380

1381

1382

1383

1384

1385

1386

1387

1388

1389

1390

1391

1392

1393

1394

1395

1396

1397

1398

1399

1400

1401

1402

1403

1404

1405

1406

1407

1408

1409

1410

1411

1412

1413

1414

1415

1416

1417

1418

1419

1420

1421

1422

1423

1424

1425

1426

1427

1428

1429

1430

1431

1432

1433

1434

1435

1436

1437

1438

1439

1440

1441

1442

1443

1444

1445

1446

1447

1448

1449

1450

1451

1452

1453

1454

1455

1456

1457

1458

1459

1460

1461

1462

1463

1464

1465

1466

1467

1468

1469

1470

1471

1472

1473

1474

1475

1476

1477

1478

1479

1480

1481

1482

1483

1484

1485

1486

1487

1488

1489

1490

1491

1492

1493

1494

1495

1496

1497

1498

1499

1500

1501

1502

1503

1504

1505

1506

1507

1508

1509

1510

1511

1512

1513

1514

1515

1516

1517

1518

1519

1520

1521

1522

1523

1524

1525

1526

1527

1528

1529

1530

1531

1532

1533

1534

1535

1536

1537

1538

1539

1540

1541

1542

1543

1544

1545

1546

1547

1548

1549

1550

1551

1552

1553

1554

1555

1556

1557

1558

1559

1560

1561

1562

1563

1564

1565

1566

1567

1568

1569

1570

1571

1572

1573

1574

1575

1576

1577

1578

1579

1580

1581

1582

1583

1584

1585

1586

1587

1588

1589

1590

1591

1592

1593

1594

1595

1596

1597

1598

1599

1600

1601

1602

1603

1604

1605

1606

1607

1608

1609

1610

1611

1612

1613

1614

1615

1616

1617

1618

1619

1620

1621

1622

1623

1624

1625

1626

1627

1628

1629

1630

1631

1632

1633

1634

1635

1636

1637

1638

1639

1640

1641

1642

1643

1644

1645

1646

1647

1648

1649

1650

1651

1652

1653

1654

1655

1656

1657

1658

1659

1660

1661

1662

1663

1664

1665

1666

1667

1668

1669

1670

1671

1672

1673

1674

1675

1676

1677

1678

1679

1680

1681

1682

1683

1684

1685

1686

1687

1688

1689

1690

1691

1692

1693

1694

1695

1696

1697

1698

1699

1700

1701

1702

1703

1704

1705

1706

1707

1708

1709

1710

1711

1712

1713

1714

1715

1716

1717

1718

1719

1720

1721

1722

1723

1724

1725

1726

1727

1728

1729

1730

1731

1732

1733

1734

1735

1736

1737

1738

1739

1740

1741

1742

1743

1744

1745

1746

1747

1748

1749

1750

1751

1752

1753

1754

1755

1756

1757

1758

1759

1760

1761

1762

1763

1764

1765

1766

1767

1768

1769

1770

1771

1772

1773

1774

1775

1776

1777

1778

1779

1780

1781

1782

1783

1784

1785

1786

1787

1788

1789

1790

1791

1792

1793

1794

1795

1796

1797

1798

1799

1800

1801

1802

1803

1804

1805

1806

1807

1808

1809

1810

1811

1812

1813

1814

1815

1816

1817

1818

1819

1820

1821

1822

1823

1824

1825

1826

1827

1828

1829

1830

1831

1832

1833

1834

1835

1836

1837

1838

1839

1840

1841

1842

1843

1844

1845

1846

1847

1848

1849

1850

1851

1852

1853

1854

1855

1856

1857

1858

1859

1860

1861

1862

1863

1864

1865

1866

1867

1868

1869

1870

1871

1872

1873

1874

1875

1876

1877

1878

1879

1880

1881

1882

1883

1884

1885

1886

1887

1888

1889

1890

1891

1892

1893

1894

1895

1896

1897

1898

1899

1900

1901

1902

1903

1904

1905

1906

1907

1908

1909

1910

1911

1912

1913

1914

1915

1916

1917

1918

1919

1920

1921

1922

1923

1924

1925

1926

1927

1928

1929

1930

1931

1932

1933

1934

1935

1936

1937

1938

1939

1940

1941

1942

1943

1944

1945

1946

1947

1948

1949

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

2022

2023

2024

2025

2026

2027

2028

2029

2030

2031

2032

2033

2034

2035

2036

2037

2038

2039

2040

2041

2042

2043

2044

2045

2046

2047

2048

2049

2050

2051

2052

2053

2054

2055

2056

2057

2058

2059

2060

2061

2062

2063

2064

2065

2066

2067

2068

2069

2070

2071

2072

2073

2074

2075

2076

2077

2078

2079

2080

2081

2082

2083

2084

2085

2086

2087

2088

2089

2090

2091

2092

2093

2094

2095

2096

2097

2098

2099

2100

2101

2102

2103

2104

2105

2106

2107

2108

2109

2110

2111

2112

2113

2114

2115

2116

2117

2118

2119

2120

2121

2122

2123

2124

2125

2126

2127

2128

2129

2130

2131

2132

2133

2134

2135

2136

2137

2138

2139

2140

2141

2142

2143

2144

2145

2146

2147

2148

2149

2150

2151

2152

2153

2154

2155

2156

2157

2158

2159

2160

2161

2162

2163

2164

2165

2166

2167

2168

2169

2170

2171

2172

2173

2174

2175

2176

2177

2178

2179

2180

2181

2182

2183

2184

2185

2186

2187

2188

2189

2190

2191

2192

2193

2194

2195

2196

2197

2198

2199

2200

2201

2202

2203

2204

2205

2206

2207

2208

2209

2210

2211

2212

2213

2214

2215

2216

2217

2218

2219

2220

2221

2222

2223

2224

2225

2226

2227

2228

2229

2230

2231

2232

2233

2234

2235

2236

2237

2238

2239

2240

2241

2242

2243

2244

2245

2246

2247

2248

2249

2250

2251

2252

2253

2254

2255

2256

2257

2258

2259

2260

2261

2262

2263

2264

2265

2266

2267

2268

2269

2270

2271

2272

2273

2274

2275

2276

2277

2278

2279

2280

2281

2282

2283

2284

2285

2286

2287

2288

2289

2290

2291

2292

2293

2294

2295

2296

2297

2298

2299

2300

2301

2302

2303

2304

2305

2306

2307

2308

2309

2310

2311

2312

2313

2314

2315

2316

2317

2318

2319

2320

2321

2322

2323

2324

2325

2326

2327

2328

2329

2330

2331

2332

2333

2334

2335

2336

2337

2338

2339

2340

2341

2342

2343

2344

2345

2346

2347

2348

2349

2350

2351

2352

2353

2354

2355

2356

2357

2358

2359

2360

2361

2362

2363

2364

2365

2366

2367

2368

2369

2370

2371

2372

2373

2374

2375

2376

2377

2378

2379

2380

2381

2382

2383

2384

2385

2386

2387

2388

2389

2390

2391

2392

2393

2394

2395

2396

2397

2398

2399

2400

2401

2402

2403

2404

2405

2406

2407

2408

2409

2410

2411

2412

2413

2414

2415

2416

2417

2418

2419

2420

2421

2422

2423

2424

2425

2426

2427

2428

2429

2430

2431

2432

2433

2434

2435

2436

2437

2438

2439

2440

2441

2442

2443

2444

2445

2446

2447

2448

2449

2450

2451

2452

2453

2454

2455

2456

2457

2458

2459

2460

2461

2462

2463

2464

2465

2466

2467

2468

2469

2470

2471

2472

2473

2474

2475

2476

2477

2478

2479

2480

2481

2482

2483

2484

2485

2486

2487

2488

2489

2490

2491

2492

2493

2494

2495

2496

2497

2498

2499

2500

2501

2502

2503

2504

2505

2506

2507

2508

2509

2510

2511

2512

2513

2514

2515

2516

2517

2518

2519

2520

2521

2522

2523

2524

2525

2526

2527

2528

2529

2530

2531

2532

2533

2534

2535

2536

2537

2538

2539

2540

2541

2542

2543

2544

2545

2546

2547

2548

2549

2550

2551

2552

2553

2554

2555

2556

2557

2558

2559

2560

2561

2562

2563

2564

2565

2566

2567

2568

2569

2570

2571

2572

2573

2574

2575

2576

2577

2578

2579

2580

2581

2582

2583

2584

2585

2586

2587

2588

2589

2590

2591

2592

2593

2594

2595

2596

2597

2598

2599

2600

2601

2602

2603

2604

2605

2606

2607

2608

2609

2610

2611

2612

2613

2614

2615

2616

2617

2618

2619

2620

2621

2622

2623

2624

2625

2626

2627

2628

2629

2630

2631

2632

2633

2634

2635

2636

2637

2638

2639

2640

2641

2642

2643

2644

2645

2646

2647

2648

2649

2650

2651

2652

2653

2654

2655

2656

2657

2658

2659

2660

2661

2662

2663

2664

2665

2666

2667

2668

2669

2670

2671

2672

2673

2674

2675

2676

2677

2678

2679

2680

2681

2682

2683

2684

2685

2686

2687

2688

2689

2690

2691

2692

2693

2694

2695

2696

2697

2698

2699

2700

2701

2702

2703

2704

2705

2706

2707

2708

2709

2710

2711

2712

2713

2714

2715

2716

2717

2718

2719

2720

2721

2722

2723

2724

2725

2726

2727

2728

2729

2730

2731

2732

2733

2734

2735

2736

2737

2738

2739

2740

2741

2742

2743

2744

2745

2746

2747

2748

2749

2750

2751

2752

2753

2754

2755

2756

2757

2758

2759

2760

2761

2762

2763

2764

2765

2766

2767

2768

2769

2770

2771

2772

2773

2774

2775

2776

2777

2778

2779

2780

2781

2782

2783

2784

2785

2786

2787

2788

2789

2790

2791

2792

2793

2794

2795

2796

2797

2798

2799

2800

2801

2802

2803

2804

2805

2806

2807

2808

2809

2810

2811

2812

2813

2814

2815

2816

2817

2818

2819

2820

2821

2822

2823

2824

2825

2826

2827

2828

2829

2830

2831

2832

2833

2834

2835

2836

2837

2838

2839

2840

2841

2842

2843

2844

2845

2846

2847

2848

2849

2850

2851

2852

2853

2854

2855

2856

2857

2858

2859

2860

2861

2862

2863

2864

2865

2866

2867

2868

2869

2870

2871

2872

2873

2874

2875

2876

2877

2878

2879

2880

2881

2882

2883

2884

2885

2886

2887

2888

2889

2890

2891

2892

2893

2894

2895

2896

2897

2898

2899

2900

2901

2902

2903

2904

2905

2906

2907

2908

2909

2910

2911

2912

2913

2914

2915

2916

2917

2918

2919

2920

2921

2922

2923

2924

2925

2926

2927

2928

2929

2930

2931

2932

2933

2934

2935

2936

2937

2938

2939

2940

2941

2942

2943

2944

2945

2946

2947

2948

2949

2950

2951

2952

2953

2954

2955

2956

2957

2958

2959

2960

2961

2962

2963

2964

2965

2966

2967

2968

2969

2970

2971

2972

2973

2974

2975

2976

2977

2978

2979

2980

2981

2982

2983

2984

2985

2986

2987

2988

2989

2990

2991

2992

2993

2994

2995

2996

2997

2998

2999

3000

3001

3002

3003

3004

3005

3006

3007

3008

3009

3010

3011

3012

3013

3014

3015

3016

3017

3018

3019

3020

3021

3022

3023

3024

3025

3026

3027

3028

3029

3030

3031

3032

3033

3034

3035

3036

3037

3038

3039

3040

3041

3042

3043

3044

3045

3046

3047

3048

3049

3050

3051

3052

3053

3054

3055

3056

3057

3058

3059

3060

3061

3062

3063

3064

3065

3066

3067

3068

3069

3070

3071

3072

3073

3074

3075

3076

3077

3078

3079

3080

3081

3082

3083

3084

3085

3086

3087

3088

3089

3090

3091

3092

3093

3094

3095

3096

3097

3098

3099

3100

3101

3102

3103

3104

3105

3106

3107

3108

3109

3110

3111

3112

3113

3114

3115

3116

3117

3118

3119

3120

3121

3122

3123

3124

3125

3126

3127

3128

3129

3130

3131

3132

3133

3134

3135

3136

3137

3138

3139

3140

3141

3142

3143

3144

3145

3146

3147

3148

3149

3150

3151

3152

3153

3154

3155

3156

3157

3158

3159

3160

3161

3162

3163

3164

3165

3166

3167

3168

3169

3170

3171

3172

3173

3174

3175

3176

3177

3178

3179

3180

3181

3182

3183

3184

3185

3186

3187

3188

3189

3190

3191

3192

3193

3194

3195

3196

3197

3198

3199

3200

3201

3202

3203

3204

3205

3206

3207

3208

3209

3210

3211

3212

3213

3214

3215

3216

3217

3218

3219

3220

3221

3222

3223

3224

3225

3226

3227

3228

3229

3230

3231

3232

3233

3234

3235

3236

3237

3238

3239

3240

3241

3242

3243

3244

3245

3246

3247

3248

3249

3250

3251

3252

3253

3254

3255

3256

3257

3258

3259

3260

3261

3262

3263

3264

3265

3266

3267

3268

3269

3270

3271

3272

3273

3274

3275

3276

3277

3278

3279

3280

3281

3282

3283

3284

3285

3286

3287

3288

3289

3290

3291

3292

3293

3294

3295

3296

3297

3298

3299

3300

3301

3302

3303

3304

3305

3306

3307

3308

3309

3310

3311

3312

3313

3314

3315

3316

3317

3318

3319

3320

3321

3322

3323

3324

3325

3326

3327

3328

3329

3330

3331

3332

3333

3334

3335

3336

3337

3338

3339

3340

3341

3342

3343

3344

3345

3346

3347

3348

3349

3350

3351

3352

3353

3354

3355

3356

3357

3358

3359

3360

3361

3362

3363

3364

3365

3366

3367

3368

3369

3370

3371

3372

3373

3374

3375

3376

3377

3378

3379

3380

3381

3382

3383

3384

3385

3386

3387

3388

3389

3390

3391

3392

3393

3394

3395

3396

3397

3398

3399

3400

3401

3402

3403

3404

3405

3406

3407

3408

3409

3410

3411

3412

3413

3414

3415

3416

3417

3418

3419

3420

3421

3422

3423

3424

3425

3426

3427

3428

3429

3430

3431

3432

3433

3434

3435

3436

3437

3438

3439

3440

3441

3442

3443

3444

3445

3446

3447

3448

3449

3450

3451

3452

3453

3454

3455

3456

3457

3458

3459

3460

3461

3462

3463

3464

3465

3466

3467

3468

3469

3470

3471

3472

3473

3474

3475

3476

3477

3478

3479

3480

3481

3482

3483

3484

3485

3486

3487

3488

3489

3490

3491

3492

3493

3494

3495

3496

3497

3498

3499

3500

3501

3502

3503

3504

3505

3506

3507

3508

3509

3510

3511

3512

3513

3514

3515

3516

3517

3518

3519

3520

3521

3522

3523

3524

3525

3526

3527

3528

3529

3530

3531

3532

3533

3534

3535

3536

3537

3538

3539

3540

3541

3542

3543

3544

3545

3546

3547

3548

3549

3550

3551

3552

3553

3554

3555

3556

3557

3558

3559

3560

3561

3562

3563

3564

3565

3566

3567

3568

3569

3570

3571

3572

3573

3574

3575

3576

3577

3578

3579

3580

3581

3582

3583

3584

3585

3586

3587

3588

3589

3590

3591

3592

3593

3594

3595

3596

3597

3598

3599

3600

3601

3602

3603

3604

3605

3606

3607

3608

3609

3610

3611

3612

3613

3614

3615

3616

3617

3618

3619

3620

3621

3622

3623

3624

3625

3626

3627

3628

3629

3630

3631

3632

3633

3634

3635

3636

3637

3638

3639

3640

3641

3642

3643

3644

3645

3646

3647

3648

3649

3650

3651

3652

3653

3654

3655

3656

3657

3658

3659

3660

3661

3662

3663

3664

3665

3666

3667

3668

3669

3670

3671

3672

3673

# http://pyrocko.org - GPLv3 

# 

# The Pyrocko Developers, 21st Century 

# ---|P------/S----------~Lg---------- 

'''This module provides basic signal processing for seismic traces.''' 

from __future__ import division, absolute_import 

 

import time 

import math 

import copy 

import logging 

 

import numpy as num 

from scipy import signal 

 

from . import util, evalresp, orthodrome, pchain, model 

from .util import reuse, hpfloat, UnavailableDecimation 

from .guts import Object, Float, Int, String, Complex, Tuple, List, \ 

StringChoice 

from .guts_array import Array 

 

try: 

newstr = unicode 

except NameError: 

newstr = str 

 

 

UnavailableDecimation # noqa 

 

guts_prefix = 'pf' 

 

logger = logging.getLogger('pyrocko.trace') 

 

 

class Trace(object): 

 

''' 

Create new trace object. 

 

A ``Trace`` object represents a single continuous strip of evenly sampled 

time series data. It is built from a 1D NumPy array containing the data 

samples and some attributes describing its beginning and ending time, its 

sampling rate and four string identifiers (its network, station, location 

and channel code). 

 

:param network: network code 

:param station: station code 

:param location: location code 

:param channel: channel code 

:param tmin: system time of first sample in [s] 

:param tmax: system time of last sample in [s] (if set to ``None`` it is 

computed from length of ``ydata``) 

:param deltat: sampling interval in [s] 

:param ydata: 1D numpy array with data samples (can be ``None`` when 

``tmax`` is not ``None``) 

:param mtime: optional modification time 

:param meta: additional meta information (not used, but maintained by the 

library) 

 

The length of the network, station, location and channel codes is not 

resricted by this software, but data formats like SAC, Mini-SEED or GSE 

have different limits on the lengths of these codes. The codes set here are 

silently truncated when the trace is stored 

''' 

 

cached_frequencies = {} 

 

def __init__(self, network='', station='STA', location='', channel='', 

tmin=0., tmax=None, deltat=1., ydata=None, mtime=None, 

meta=None): 

 

self._growbuffer = None 

 

if deltat < 0.001: 

tmin = hpfloat(tmin) 

if tmax is not None: 

tmax = hpfloat(tmax) 

 

if mtime is None: 

mtime = time.time() 

 

self.network, self.station, self.location, self.channel = [ 

reuse(x) for x in (network, station, location, channel)] 

 

self.tmin = tmin 

self.deltat = deltat 

 

if tmax is None: 

if ydata is not None: 

self.tmax = self.tmin + (ydata.size-1)*self.deltat 

else: 

raise Exception( 

'fixme: trace must be created with tmax or ydata') 

else: 

n = int(round((tmax - self.tmin) / self.deltat)) + 1 

self.tmax = self.tmin + (n - 1) * self.deltat 

 

self.meta = meta 

self.ydata = ydata 

self.mtime = mtime 

self._update_ids() 

self.file = None 

self._pchain = None 

 

def __str__(self): 

fmt = min(9, max(0, -int(math.floor(math.log10(self.deltat))))) 

s = 'Trace (%s, %s, %s, %s)\n' % self.nslc_id 

s += ' timerange: %s - %s\n' % ( 

util.time_to_str(self.tmin, format=fmt), 

util.time_to_str(self.tmax, format=fmt)) 

 

s += ' delta t: %g\n' % self.deltat 

if self.meta: 

for k in sorted(self.meta.keys()): 

s += ' %s: %s\n' % (k, self.meta[k]) 

return s 

 

def __getstate__(self): 

return (self.network, self.station, self.location, self.channel, 

self.tmin, self.tmax, self.deltat, self.mtime, 

self.ydata, self.meta) 

 

def __setstate__(self, state): 

if len(state) == 10: 

self.network, self.station, self.location, self.channel, \ 

self.tmin, self.tmax, self.deltat, self.mtime, \ 

self.ydata, self.meta = state 

 

else: 

# backward compatibility with old behaviour 

self.network, self.station, self.location, self.channel, \ 

self.tmin, self.tmax, self.deltat, self.mtime = state 

self.ydata = None 

self.meta = None 

 

self._growbuffer = None 

self._update_ids() 

 

def name(self): 

''' 

Get a short string description. 

''' 

 

s = '%s.%s.%s.%s, %s, %s' % (self.nslc_id + ( 

util.time_to_str(self.tmin), 

util.time_to_str(self.tmax))) 

 

return s 

 

def __eq__(self, other): 

return ( 

self.network == other.network 

and self.station == other.station 

and self.location == other.location 

and self.channel == other.channel 

and (abs(self.deltat - other.deltat) 

< (self.deltat + other.deltat)*1e-6) 

and abs(self.tmin-other.tmin) < self.deltat*0.01 

and abs(self.tmax-other.tmax) < self.deltat*0.01 

and num.all(self.ydata == other.ydata)) 

 

def almost_equal(self, other, rtol=1e-5, atol=0.0): 

return ( 

self.network == other.network 

and self.station == other.station 

and self.location == other.location 

and self.channel == other.channel 

and (abs(self.deltat - other.deltat) 

< (self.deltat + other.deltat)*1e-6) 

and abs(self.tmin-other.tmin) < self.deltat*0.01 

and abs(self.tmax-other.tmax) < self.deltat*0.01 

and num.allclose(self.ydata, other.ydata, rtol=rtol, atol=atol)) 

 

def assert_almost_equal(self, other, rtol=1e-5, atol=0.0): 

 

assert self.network == other.network, \ 

'network codes differ: %s, %s' % (self.network, other.network) 

assert self.station == other.station, \ 

'station codes differ: %s, %s' % (self.station, other.station) 

assert self.location == other.location, \ 

'location codes differ: %s, %s' % (self.location, other.location) 

assert self.channel == other.channel, 'channel codes differ' 

assert (abs(self.deltat - other.deltat) 

< (self.deltat + other.deltat)*1e-6), \ 

'sampling intervals differ %g, %g' % (self.deltat, other.delta) 

assert abs(self.tmin-other.tmin) < self.deltat*0.01, \ 

'start times differ: %s, %s' % ( 

util.time_to_str(self.tmin), util.time_to_str(other.tmin)) 

assert abs(self.tmax-other.tmax) < self.deltat*0.01, \ 

'end times differ: %s, %s' % ( 

util.time_to_str(self.tmax), util.time_to_str(other.tmax)) 

 

assert num.allclose(self.ydata, other.ydata, rtol=rtol, atol=atol), \ 

'trace values differ' 

 

def __hash__(self): 

return id(self) 

 

def __call__(self, t, clip=False, snap=round): 

it = int(snap((t-self.tmin)/self.deltat)) 

if clip: 

it = max(0, min(it, self.ydata.size-1)) 

else: 

if it < 0 or self.ydata.size <= it: 

raise IndexError() 

 

return self.tmin+it*self.deltat, self.ydata[it] 

 

def interpolate(self, t, clip=False): 

''' 

Value of trace between supporting points through linear interpolation. 

 

:param t: time instant 

:param clip: whether to clip indices to trace ends 

''' 

 

t0, y0 = self(t, clip=clip, snap=math.floor) 

t1, y1 = self(t, clip=clip, snap=math.ceil) 

if t0 == t1: 

return y0 

else: 

return y0+(t-t0)/(t1-t0)*(y1-y0) 

 

def index_clip(self, i): 

''' 

Clip index to valid range. 

''' 

 

return min(max(0, i), self.ydata.size) 

 

def add(self, other, interpolate=True, left=0., right=0.): 

''' 

Add values of other trace (self += other). 

 

Add values of ``other`` trace to the values of ``self``, where it 

intersects with ``other``. This method does not change the extent of 

``self``. If ``interpolate`` is ``True`` (the default), the values of 

``other`` to be added are interpolated at sampling instants of 

``self``. Linear interpolation is performed. In this case the sampling 

rate of ``other`` must be equal to or lower than that of ``self``. If 

``interpolate`` is ``False``, the sampling rates of the two traces must 

match. 

''' 

 

if interpolate: 

assert self.deltat <= other.deltat \ 

or same_sampling_rate(self, other) 

 

other_xdata = other.get_xdata() 

xdata = self.get_xdata() 

self.ydata += num.interp( 

xdata, other_xdata, other.ydata, left=left, right=left) 

else: 

assert self.deltat == other.deltat 

ioff = int(round((other.tmin-self.tmin)/self.deltat)) 

ibeg = max(0, ioff) 

iend = min(self.data_len(), ioff+other.data_len()) 

self.ydata[ibeg:iend] += other.ydata[ibeg-ioff:iend-ioff] 

 

def mult(self, other, interpolate=True): 

''' 

Muliply with values of other trace ``(self *= other)``. 

 

Multiply values of ``other`` trace to the values of ``self``, where it 

intersects with ``other``. This method does not change the extent of 

``self``. If ``interpolate`` is ``True`` (the default), the values of 

``other`` to be multiplied are interpolated at sampling instants of 

``self``. Linear interpolation is performed. In this case the sampling 

rate of ``other`` must be equal to or lower than that of ``self``. If 

``interpolate`` is ``False``, the sampling rates of the two traces must 

match. 

''' 

 

if interpolate: 

assert self.deltat <= other.deltat or \ 

same_sampling_rate(self, other) 

 

other_xdata = other.get_xdata() 

xdata = self.get_xdata() 

self.ydata *= num.interp( 

xdata, other_xdata, other.ydata, left=0., right=0.) 

else: 

assert self.deltat == other.deltat 

ibeg1 = int(round((other.tmin-self.tmin)/self.deltat)) 

ibeg2 = int(round((self.tmin-other.tmin)/self.deltat)) 

iend1 = int(round((other.tmax-self.tmin)/self.deltat))+1 

iend2 = int(round((self.tmax-other.tmin)/self.deltat))+1 

 

ibeg1 = self.index_clip(ibeg1) 

iend1 = self.index_clip(iend1) 

ibeg2 = self.index_clip(ibeg2) 

iend2 = self.index_clip(iend2) 

 

self.ydata[ibeg1:iend1] *= other.ydata[ibeg2:iend2] 

 

def max(self): 

''' 

Get time and value of data maximum. 

''' 

 

i = num.argmax(self.ydata) 

return self.tmin + i*self.deltat, self.ydata[i] 

 

def min(self): 

''' 

Get time and value of data minimum. 

''' 

 

i = num.argmin(self.ydata) 

return self.tmin + i*self.deltat, self.ydata[i] 

 

def absmax(self): 

''' 

Get time and value of maximum of the absolute of data. 

''' 

 

tmi, mi = self.min() 

tma, ma = self.max() 

if abs(mi) > abs(ma): 

return tmi, abs(mi) 

else: 

return tma, abs(ma) 

 

def set_codes( 

self, network=None, station=None, location=None, channel=None): 

 

''' 

Set network, station, location, and channel codes. 

''' 

 

if network is not None: 

self.network = network 

if station is not None: 

self.station = station 

if location is not None: 

self.location = location 

if channel is not None: 

self.channel = channel 

 

self._update_ids() 

 

def set_network(self, network): 

self.network = network 

self._update_ids() 

 

def set_station(self, station): 

self.station = station 

self._update_ids() 

 

def set_location(self, location): 

self.location = location 

self._update_ids() 

 

def set_channel(self, channel): 

self.channel = channel 

self._update_ids() 

 

def overlaps(self, tmin, tmax): 

''' 

Check if trace has overlap with a given time span. 

''' 

 

return not (tmax < self.tmin or self.tmax < tmin) 

 

def is_relevant(self, tmin, tmax, selector=None): 

''' 

Check if trace has overlap with a given time span and matches a 

condition callback. (internal use) 

''' 

 

return not (tmax <= self.tmin or self.tmax < tmin) \ 

and (selector is None or selector(self)) 

 

def _update_ids(self): 

''' 

Update dependent ids. 

''' 

 

self.full_id = ( 

self.network, self.station, self.location, self.channel, self.tmin) 

self.nslc_id = reuse( 

(self.network, self.station, self.location, self.channel)) 

 

def prune_from_reuse_cache(self): 

util.deuse(self.nslc_id) 

util.deuse(self.network) 

util.deuse(self.station) 

util.deuse(self.location) 

util.deuse(self.channel) 

 

def set_mtime(self, mtime): 

''' 

Set modification time of the trace. 

''' 

 

self.mtime = mtime 

 

def get_xdata(self): 

''' 

Create array for time axis. 

''' 

 

if self.ydata is None: 

raise NoData() 

 

return self.tmin \ 

+ num.arange(len(self.ydata), dtype=num.float64) * self.deltat 

 

def get_ydata(self): 

''' 

Get data array. 

''' 

 

if self.ydata is None: 

raise NoData() 

 

return self.ydata 

 

def set_ydata(self, new_ydata): 

''' 

Replace data array. 

''' 

 

self.drop_growbuffer() 

self.ydata = new_ydata 

self.tmax = self.tmin+(len(self.ydata)-1)*self.deltat 

 

def data_len(self): 

if self.ydata is not None: 

return self.ydata.size 

else: 

return int(round((self.tmax-self.tmin)/self.deltat)) + 1 

 

def drop_data(self): 

''' 

Forget data, make dataless trace. 

''' 

 

self.drop_growbuffer() 

self.ydata = None 

 

def drop_growbuffer(self): 

''' 

Detach the traces grow buffer. 

''' 

 

self._growbuffer = None 

self._pchain = None 

 

def copy(self, data=True): 

''' 

Make a deep copy of the trace. 

''' 

 

tracecopy = copy.copy(self) 

tracecopy.drop_growbuffer() 

if data: 

tracecopy.ydata = self.ydata.copy() 

tracecopy.meta = copy.deepcopy(self.meta) 

return tracecopy 

 

def crop_zeros(self): 

''' 

Remove any zeros at beginning and end. 

''' 

 

indices = num.where(self.ydata != 0.0)[0] 

if indices.size == 0: 

raise NoData() 

 

ibeg = indices[0] 

iend = indices[-1]+1 

if ibeg == 0 and iend == self.ydata.size-1: 

return 

 

self.drop_growbuffer() 

self.ydata = self.ydata[ibeg:iend].copy() 

self.tmin = self.tmin+ibeg*self.deltat 

self.tmax = self.tmin+(len(self.ydata)-1)*self.deltat 

self._update_ids() 

 

def append(self, data): 

''' 

Append data to the end of the trace. 

 

To make this method efficient when successively very few or even single 

samples are appended, a larger grow buffer is allocated upon first 

invocation. The traces data is then changed to be a view into the 

currently filled portion of the grow buffer array. 

''' 

 

assert self.ydata.dtype == data.dtype 

newlen = data.size + self.ydata.size 

if self._growbuffer is None or self._growbuffer.size < newlen: 

self._growbuffer = num.empty(newlen*2, dtype=self.ydata.dtype) 

self._growbuffer[:self.ydata.size] = self.ydata 

self._growbuffer[self.ydata.size:newlen] = data 

self.ydata = self._growbuffer[:newlen] 

self.tmax = self.tmin + (newlen-1)*self.deltat 

 

def chop( 

self, tmin, tmax, inplace=True, include_last=False, 

snap=(round, round), want_incomplete=True): 

 

''' 

Cut the trace to given time span. 

 

If the ``inplace`` argument is True (the default) the trace is cut in 

place, otherwise a new trace with the cut part is returned. By 

default, the indices where to start and end the trace data array are 

determined by rounding of ``tmin`` and ``tmax`` to sampling instances 

using Python's :py:func:`round` function. This behaviour can be changed 

with the ``snap`` argument, which takes a tuple of two functions (one 

for the lower and one for the upper end) to be used instead of 

:py:func:`round`. The last sample is by default not included unless 

``include_last`` is set to True. If the given time span exceeds the 

available time span of the trace, the available part is returned, 

unless ``want_incomplete`` is set to False - in that case, a 

:py:exc:`NoData` exception is raised. This exception is always raised, 

when the requested time span does dot overlap with the trace's time 

span. 

''' 

 

if want_incomplete: 

if tmax <= self.tmin-self.deltat or self.tmax+self.deltat < tmin: 

raise NoData() 

else: 

if tmin < self.tmin or self.tmax < tmax: 

raise NoData() 

 

ibeg = max(0, t2ind(tmin-self.tmin, self.deltat, snap[0])) 

iplus = 0 

if include_last: 

iplus = 1 

 

iend = min( 

self.data_len(), 

t2ind(tmax-self.tmin, self.deltat, snap[1])+iplus) 

 

if ibeg >= iend: 

raise NoData() 

 

obj = self 

if not inplace: 

obj = self.copy(data=False) 

 

self.drop_growbuffer() 

if self.ydata is not None: 

obj.ydata = self.ydata[ibeg:iend].copy() 

else: 

obj.ydata = None 

 

obj.tmin = obj.tmin+ibeg*obj.deltat 

obj.tmax = obj.tmin+((iend-ibeg)-1)*obj.deltat 

 

obj._update_ids() 

 

return obj 

 

def downsample(self, ndecimate, snap=False, initials=None, demean=False): 

''' 

Downsample trace by a given integer factor. 

 

:param ndecimate: decimation factor, avoid values larger than 8 

:param snap: whether to put the new sampling instances closest to 

multiples of the sampling rate. 

:param initials: ``None``, ``True``, or initial conditions for the 

anti-aliasing filter, obtained from a previous run. In the latter 

two cases the final state of the filter is returned instead of 

``None``. 

:param demean: whether to demean the signal before filtering. 

''' 

 

newdeltat = self.deltat*ndecimate 

if snap: 

ilag = int(round( 

(math.ceil(self.tmin / newdeltat) * newdeltat - self.tmin) 

/ self.deltat)) 

else: 

ilag = 0 

 

if snap and ilag > 0 and ilag < self.ydata.size: 

data = self.ydata.astype(num.float64) 

self.tmin += ilag*self.deltat 

else: 

data = self.ydata.astype(num.float64) 

 

if demean: 

data -= num.mean(data) 

 

if data.size != 0: 

result = util.decimate( 

data, ndecimate, ftype='fir', zi=initials, ioff=ilag) 

else: 

result = data 

 

if initials is None: 

self.ydata, finals = result, None 

else: 

self.ydata, finals = result 

 

self.deltat = reuse(self.deltat*ndecimate) 

self.tmax = self.tmin+(len(self.ydata)-1)*self.deltat 

self._update_ids() 

 

return finals 

 

def downsample_to(self, deltat, snap=False, allow_upsample_max=1, 

initials=None, demean=False): 

 

''' 

Downsample to given sampling rate. 

 

Tries to downsample the trace to a target sampling interval of 

``deltat``. This runs the :py:meth:`Trace.downsample` one or several 

times. If allow_upsample_max is set to a value larger than 1, 

intermediate upsampling steps are allowed, in order to increase the 

number of possible downsampling ratios. 

 

If the requested ratio is not supported, an exception of type 

:py:exc:`pyrocko.util.UnavailableDecimation` is raised. 

''' 

 

ratio = deltat/self.deltat 

rratio = round(ratio) 

 

ok = False 

for upsratio in range(1, allow_upsample_max+1): 

dratio = (upsratio/self.deltat) / (1./deltat) 

if abs(dratio - round(dratio)) / dratio < 0.0001 and \ 

util.decitab(int(round(dratio))): 

 

ok = True 

break 

 

if not ok: 

raise util.UnavailableDecimation('ratio = %g' % ratio) 

 

if upsratio > 1: 

self.drop_growbuffer() 

ydata = self.ydata 

self.ydata = num.zeros( 

ydata.size*upsratio-(upsratio-1), ydata.dtype) 

self.ydata[::upsratio] = ydata 

for i in range(1, upsratio): 

self.ydata[i::upsratio] = \ 

float(i)/upsratio * ydata[:-1] \ 

+ float(upsratio-i)/upsratio * ydata[1:] 

self.deltat = self.deltat/upsratio 

 

ratio = deltat/self.deltat 

rratio = round(ratio) 

 

deci_seq = util.decitab(int(rratio)) 

finals = [] 

for i, ndecimate in enumerate(deci_seq): 

if ndecimate != 1: 

xinitials = None 

if initials is not None: 

xinitials = initials[i] 

finals.append(self.downsample( 

ndecimate, snap=snap, initials=xinitials, demean=demean)) 

 

if initials is not None: 

return finals 

 

def resample(self, deltat): 

''' 

Resample to given sampling rate ``deltat``. 

 

Resampling is performed in the frequency domain. 

''' 

 

ndata = self.ydata.size 

ntrans = nextpow2(ndata) 

fntrans2 = ntrans * self.deltat/deltat 

ntrans2 = int(round(fntrans2)) 

deltat2 = self.deltat * float(ntrans)/float(ntrans2) 

ndata2 = int(round(ndata*self.deltat/deltat2)) 

if abs(fntrans2 - ntrans2) > 1e-7: 

logger.warning( 

'resample: requested deltat %g could not be matched exactly: ' 

'%g' % (deltat, deltat2)) 

 

data = self.ydata 

data_pad = num.zeros(ntrans, dtype=num.float) 

data_pad[:ndata] = data 

fdata = num.fft.rfft(data_pad) 

fdata2 = num.zeros((ntrans2+1)//2, dtype=fdata.dtype) 

n = min(fdata.size, fdata2.size) 

fdata2[:n] = fdata[:n] 

data2 = num.fft.irfft(fdata2) 

data2 = data2[:ndata2] 

data2 *= float(ntrans2) / float(ntrans) 

self.deltat = deltat2 

self.set_ydata(data2) 

 

def resample_simple(self, deltat): 

tyear = 3600*24*365. 

 

if deltat == self.deltat: 

return 

 

if abs(self.deltat - deltat) * tyear/deltat < deltat: 

logger.warning( 

'resample_simple: less than one sample would have to be ' 

'inserted/deleted per year. Doing nothing.') 

return 

 

ninterval = int(round(deltat / (self.deltat - deltat))) 

if abs(ninterval) < 20: 

logger.error( 

'resample_simple: sample insertion/deletion interval less ' 

'than 20. results would be erroneous.') 

raise ResamplingFailed() 

 

delete = False 

if ninterval < 0: 

ninterval = - ninterval 

delete = True 

 

tyearbegin = util.year_start(self.tmin) 

 

nmin = int(round((self.tmin - tyearbegin)/deltat)) 

 

ibegin = (((nmin-1)//ninterval)+1) * ninterval - nmin 

nindices = (len(self.ydata) - ibegin - 1) / ninterval + 1 

if nindices > 0: 

indices = ibegin + num.arange(nindices) * ninterval 

data_split = num.split(self.ydata, indices) 

data = [] 

for ln, h in zip(data_split[:-1], data_split[1:]): 

if delete: 

ln = ln[:-1] 

 

data.append(ln) 

if not delete: 

if ln.size == 0: 

v = h[0] 

else: 

v = 0.5*(ln[-1] + h[0]) 

data.append(num.array([v], dtype=ln.dtype)) 

 

data.append(data_split[-1]) 

 

ydata_new = num.concatenate(data) 

 

self.tmin = tyearbegin + nmin * deltat 

self.deltat = deltat 

self.set_ydata(ydata_new) 

 

def stretch(self, tmin_new, tmax_new): 

''' 

Stretch signal while preserving sample rate using sinc interpolation. 

 

:param tmin_new: new time of first sample 

:param tmax_new: new time of last sample 

 

This method can be used to correct for a small linear time drift or to 

introduce sub-sample time shifts. The amount of stretching is limited 

to 10% by the implementation and is expected to be much smaller than 

that by the approximations used. 

''' 

 

from pyrocko import signal_ext 

 

i_control = num.array([0, self.ydata.size-1], dtype=num.int64) 

t_control = num.array([tmin_new, tmax_new], dtype=num.float) 

 

r = (tmax_new - tmin_new) / self.deltat + 1.0 

n_new = int(round(r)) 

if abs(n_new - r) > 0.001: 

n_new = int(math.floor(r)) 

 

assert n_new >= 2 

 

tmax_new = tmin_new + (n_new-1) * self.deltat 

 

ydata_new = num.empty(n_new, dtype=num.float) 

signal_ext.antidrift(i_control, t_control, 

self.ydata.astype(num.float), 

tmin_new, self.deltat, ydata_new) 

 

self.tmin = tmin_new 

self.set_ydata(ydata_new) 

self._update_ids() 

 

def nyquist_check(self, frequency, intro='Corner frequency', warn=True, 

raise_exception=False): 

 

''' 

Check if a given frequency is above the Nyquist frequency of the trace. 

 

:param intro: string used to introduce the warning/error message 

:param warn: whether to emit a warning 

:param raise_exception: whether to raise an :py:exc:`AboveNyquist` 

exception. 

''' 

 

if frequency >= 0.5/self.deltat: 

message = '%s (%g Hz) is equal to or higher than nyquist ' \ 

'frequency (%g Hz). (Trace %s)' \ 

% (intro, frequency, 0.5/self.deltat, self.name()) 

if warn: 

logger.warning(message) 

if raise_exception: 

raise AboveNyquist(message) 

 

def lowpass(self, order, corner, nyquist_warn=True, 

nyquist_exception=False, demean=True): 

 

''' 

Apply Butterworth lowpass to the trace. 

 

:param order: order of the filter 

:param corner: corner frequency of the filter 

 

Mean is removed before filtering. 

''' 

 

self.nyquist_check( 

corner, 'Corner frequency of lowpass', nyquist_warn, 

nyquist_exception) 

 

(b, a) = _get_cached_filter_coefs( 

order, [corner*2.0*self.deltat], btype='low') 

 

if len(a) != order+1 or len(b) != order+1: 

logger.warning( 

'Erroneous filter coefficients returned by ' 

'scipy.signal.butter(). You may need to downsample the ' 

'signal before filtering.') 

 

data = self.ydata.astype(num.float64) 

if demean: 

data -= num.mean(data) 

self.drop_growbuffer() 

self.ydata = signal.lfilter(b, a, data) 

 

def highpass(self, order, corner, nyquist_warn=True, 

nyquist_exception=False, demean=True): 

 

''' 

Apply butterworth highpass to the trace. 

 

:param order: order of the filter 

:param corner: corner frequency of the filter 

 

Mean is removed before filtering. 

''' 

 

self.nyquist_check( 

corner, 'Corner frequency of highpass', nyquist_warn, 

nyquist_exception) 

 

(b, a) = _get_cached_filter_coefs( 

order, [corner*2.0*self.deltat], btype='high') 

 

data = self.ydata.astype(num.float64) 

if len(a) != order+1 or len(b) != order+1: 

logger.warning( 

'Erroneous filter coefficients returned by ' 

'scipy.signal.butter(). You may need to downsample the ' 

'signal before filtering.') 

if demean: 

data -= num.mean(data) 

self.drop_growbuffer() 

self.ydata = signal.lfilter(b, a, data) 

 

def bandpass(self, order, corner_hp, corner_lp, demean=True): 

''' 

Apply butterworth bandpass to the trace. 

 

:param order: order of the filter 

:param corner_hp: lower corner frequency of the filter 

:param corner_lp: upper corner frequency of the filter 

 

Mean is removed before filtering. 

''' 

 

self.nyquist_check(corner_hp, 'Lower corner frequency of bandpass') 

self.nyquist_check(corner_lp, 'Higher corner frequency of bandpass') 

(b, a) = _get_cached_filter_coefs( 

order, 

[corner*2.0*self.deltat for corner in (corner_hp, corner_lp)], 

btype='band') 

data = self.ydata.astype(num.float64) 

if demean: 

data -= num.mean(data) 

self.drop_growbuffer() 

self.ydata = signal.lfilter(b, a, data) 

 

def abshilbert(self): 

self.drop_growbuffer() 

self.ydata = num.abs(hilbert(self.ydata)) 

 

def envelope(self, inplace=True): 

''' 

Calculate the envelope of the trace. 

 

:param inplace: calculate envelope in place 

 

The calculation follows: 

 

.. math:: 

 

Y' = \\sqrt{Y^2+H(Y)^2} 

 

where H is the Hilbert-Transform of the signal Y. 

''' 

 

if inplace: 

self.drop_growbuffer() 

self.ydata = num.sqrt(self.ydata**2 + hilbert(self.ydata)**2) 

else: 

tr = self.copy(data=False) 

tr.ydata = num.sqrt(self.ydata**2 + hilbert(self.ydata)**2) 

return tr 

 

def taper(self, taperer, inplace=True, chop=False): 

''' 

Apply a :py:class:`Taper` to the trace. 

 

:param taperer: instance of :py:class:`Taper` subclass 

:param inplace: apply taper inplace 

:param chop: if ``True``: exclude tapered parts from the resulting 

trace 

''' 

 

if not inplace: 

tr = self.copy() 

else: 

tr = self 

 

if chop: 

i, n = taperer.span(tr.ydata, tr.tmin, tr.deltat) 

tr.shift(i*tr.deltat) 

tr.set_ydata(tr.ydata[i:i+n]) 

 

taperer(tr.ydata, tr.tmin, tr.deltat) 

 

if not inplace: 

return tr 

 

def whiten(self, order=6): 

''' 

Whiten signal in time domain using autoregression and recursive filter. 

 

:param order: order of the autoregression process 

''' 

 

b, a = self.whitening_coefficients(order) 

self.drop_growbuffer() 

self.ydata = signal.lfilter(b, a, self.ydata) 

 

def whitening_coefficients(self, order=6): 

ar = yulewalker(self.ydata, order) 

b, a = [1.] + ar.tolist(), [1.] 

return b, a 

 

def ampspec_whiten( 

self, 

width, 

td_taper='auto', 

fd_taper='auto', 

pad_to_pow2=True, 

demean=True): 

 

''' 

Whiten signal via frequency domain using moving average on amplitude 

spectra. 

 

:param width: width of smoothing kernel [Hz] 

:param td_taper: time domain taper, object of type :py:class:`Taper` or 

``None`` or ``'auto'``. 

:param fd_taper: frequency domain taper, object of type 

:py:class:`Taper` or ``None`` or ``'auto'``. 

:param pad_to_pow2: whether to pad the signal with zeros up to a length 

of 2^n 

:param demean: whether to demean the signal before tapering 

 

The signal is first demeaned and then tapered using ``td_taper``. Then, 

the spectrum is calculated and inversely weighted with a smoothed 

version of its amplitude spectrum. A moving average is used for the 

smoothing. The smoothed spectrum is then tapered using ``fd_taper``. 

Finally, the smoothed and tapered spectrum is back-transformed into the 

time domain. 

 

If ``td_taper`` is set to ``'auto'``, ``CosFader(1.0/width)`` is used. 

If ``fd_taper`` is set to ``'auto'``, ``CosFader(width)`` is used. 

''' 

 

ndata = self.data_len() 

 

if pad_to_pow2: 

ntrans = nextpow2(ndata) 

else: 

ntrans = ndata 

 

df = 1./(ntrans*self.deltat) 

nw = int(round(width/df)) 

if ndata//2+1 <= nw: 

raise TraceTooShort( 

'Samples in trace: %s, samples needed: %s' % (ndata, nw)) 

 

if td_taper == 'auto': 

td_taper = CosFader(1./width) 

 

if fd_taper == 'auto': 

fd_taper = CosFader(width) 

 

if td_taper: 

self.taper(td_taper) 

 

ydata = self.get_ydata().astype(num.float) 

if demean: 

ydata -= ydata.mean() 

 

spec = num.fft.rfft(ydata, ntrans) 

 

amp = num.abs(spec) 

nspec = amp.size 

csamp = num.cumsum(amp) 

amp_smoothed = num.empty(nspec, dtype=csamp.dtype) 

n1, n2 = nw//2, nw//2 + nspec - nw 

amp_smoothed[n1:n2] = (csamp[nw:] - csamp[:-nw]) / nw 

amp_smoothed[:n1] = amp_smoothed[n1] 

amp_smoothed[n2:] = amp_smoothed[n2-1] 

 

denom = amp_smoothed * amp 

numer = amp 

eps = num.mean(denom) * 1e-9 

if eps == 0.0: 

eps = 1e-9 

 

numer += eps 

denom += eps 

spec *= numer/denom 

 

if fd_taper: 

fd_taper(spec, 0., df) 

 

ydata = num.fft.irfft(spec) 

self.set_ydata(ydata[:ndata]) 

 

def _get_cached_freqs(self, nf, deltaf): 

ck = (nf, deltaf) 

if ck not in Trace.cached_frequencies: 

Trace.cached_frequencies[ck] = deltaf * num.arange( 

nf, dtype=num.float) 

 

return Trace.cached_frequencies[ck] 

 

def bandpass_fft(self, corner_hp, corner_lp): 

''' 

Apply boxcar bandbpass to trace (in spectral domain). 

''' 

 

n = len(self.ydata) 

n2 = nextpow2(n) 

data = num.zeros(n2, dtype=num.float64) 

data[:n] = self.ydata 

fdata = num.fft.rfft(data) 

freqs = self._get_cached_freqs(len(fdata), 1./(self.deltat*n2)) 

fdata[0] = 0.0 

fdata *= num.logical_and(corner_hp < freqs, freqs < corner_lp) 

data = num.fft.irfft(fdata) 

self.drop_growbuffer() 

self.ydata = data[:n] 

 

def shift(self, tshift): 

''' 

Time shift the trace. 

''' 

 

self.tmin += tshift 

self.tmax += tshift 

self._update_ids() 

 

def snap(self, inplace=True, interpolate=False): 

''' 

Shift trace samples to nearest even multiples of the sampling rate. 

 

:param inplace: (boolean) snap traces inplace 

 

If ``inplace`` is ``False`` and the difference of tmin and tmax of 

both, the snapped and the original trace is smaller than 0.01 x deltat, 

:py:func:`snap` returns the unsnapped instance of the original trace. 

''' 

 

tmin = round(self.tmin/self.deltat)*self.deltat 

tmax = tmin + (self.ydata.size-1)*self.deltat 

 

if inplace: 

xself = self 

else: 

if abs(self.tmin - tmin) < 1e-2*self.deltat and \ 

abs(self.tmax - tmax) < 1e-2*self.deltat: 

return self 

 

xself = self.copy() 

 

if interpolate: 

from pyrocko import signal_ext 

n = xself.data_len() 

ydata_new = num.empty(n, dtype=num.float) 

i_control = num.array([0, n-1], dtype=num.int64) 

t_control = num.array([xself.tmin, xself.tmax]) 

signal_ext.antidrift(i_control, t_control, 

xself.ydata.astype(num.float), 

tmin, xself.deltat, ydata_new) 

 

xself.ydata = ydata_new 

 

xself.tmin = tmin 

xself.tmax = tmax 

xself._update_ids() 

 

return xself 

 

def fix_deltat_rounding_errors(self): 

''' 

Try to undo sampling rate rounding errors. 

 

See :py:func:`fix_deltat_rounding_errors`. 

''' 

 

self.deltat = fix_deltat_rounding_errors(self.deltat) 

self.tmax = self.tmin + (self.data_len() - 1) * self.deltat 

 

def sta_lta_centered(self, tshort, tlong, quad=True, scalingmethod=1): 

''' 

Run special STA/LTA filter where the short time window is centered on 

the long time window. 

 

:param tshort: length of short time window in [s] 

:param tlong: length of long time window in [s] 

:param quad: whether to square the data prior to applying the STA/LTA 

filter 

:param scalingmethod: integer key to select how output values are 

scaled / normalized (``1``, ``2``, or ``3``) 

 

============= ====================================== =========== 

Scalingmethod Implementation Range 

============= ====================================== =========== 

``1`` As/Al* Tl/Ts [0,1] 

``2`` (As/Al - 1) / (Tl/Ts - 1) [-Ts/Tl,1] 

``3`` Like ``2`` but clipping range at zero [0,1] 

============= ====================================== =========== 

 

''' 

 

nshort = int(round(tshort/self.deltat)) 

nlong = int(round(tlong/self.deltat)) 

 

assert nshort < nlong 

if nlong > len(self.ydata): 

raise TraceTooShort( 

'Samples in trace: %s, samples needed: %s' 

% (len(self.ydata), nlong)) 

 

if quad: 

sqrdata = self.ydata**2 

else: 

sqrdata = self.ydata 

 

mavg_short = moving_avg(sqrdata, nshort) 

mavg_long = moving_avg(sqrdata, nlong) 

 

self.drop_growbuffer() 

 

if scalingmethod not in (1, 2, 3): 

raise Exception('Invalid argument to scalingrange argument.') 

 

if scalingmethod == 1: 

self.ydata = mavg_short/mavg_long * float(nshort)/float(nlong) 

elif scalingmethod in (2, 3): 

self.ydata = (mavg_short/mavg_long - 1.) \ 

/ ((float(nlong)/float(nshort)) - 1) 

 

if scalingmethod == 3: 

self.ydata = num.maximum(self.ydata, 0.) 

 

def sta_lta_right(self, tshort, tlong, quad=True, scalingmethod=1): 

''' 

Run special STA/LTA filter where the short time window is overlapping 

with the last part of the long time window. 

 

:param tshort: length of short time window in [s] 

:param tlong: length of long time window in [s] 

:param quad: whether to square the data prior to applying the STA/LTA 

filter 

:param scalingmethod: integer key to select how output values are 

scaled / normalized (``1``, ``2``, or ``3``) 

 

============= ====================================== =========== 

Scalingmethod Implementation Range 

============= ====================================== =========== 

``1`` As/Al* Tl/Ts [0,1] 

``2`` (As/Al - 1) / (Tl/Ts - 1) [-Ts/Tl,1] 

``3`` Like ``2`` but clipping range at zero [0,1] 

============= ====================================== =========== 

 

With ``scalingmethod=1``, the values produced by this variant of the 

STA/LTA are equivalent to 

 

.. math:: 

s_i = \\frac{s}{l} \\frac{\\frac{1}{s}\\sum_{j=i}{i+s-1} f_j} 

{\\frac{1}{l}\\sum_{j=i+s-l}^{i+s-1} f_j} 

 

where :math:`f_j` are the input samples, :math:`s` are the number of 

samples in the short time window and :math:`l` are the number of 

samples in the long time window. 

''' 

 

n = self.data_len() 

tmin = self.tmin 

 

nshort = max(1, int(round(tshort/self.deltat))) 

nlong = max(1, int(round(tlong/self.deltat))) 

 

assert nshort < nlong 

 

if nlong > len(self.ydata): 

raise TraceTooShort( 

'Samples in trace: %s, samples needed: %s' 

% (len(self.ydata), nlong)) 

 

if quad: 

sqrdata = self.ydata**2 

else: 

sqrdata = self.ydata 

 

nshift = int(math.floor(0.5 * (nlong - nshort))) 

if nlong % 2 != 0 and nshort % 2 == 0: 

nshift += 1 

 

mavg_short = moving_avg(sqrdata, nshort)[nshift:] 

mavg_long = moving_avg(sqrdata, nlong)[:sqrdata.size-nshift] 

 

self.drop_growbuffer() 

 

if scalingmethod not in (1, 2, 3): 

raise Exception('Invalid argument to scalingrange argument.') 

 

if scalingmethod == 1: 

ydata = mavg_short/mavg_long * nshort/nlong 

elif scalingmethod in (2, 3): 

ydata = (mavg_short/mavg_long - 1.) \ 

/ ((float(nlong)/float(nshort)) - 1) 

 

if scalingmethod == 3: 

ydata = num.maximum(self.ydata, 0.) 

 

self.set_ydata(ydata) 

 

self.shift((math.ceil(0.5*nlong) - nshort + 1) * self.deltat) 

 

self.chop( 

tmin + (nlong - nshort) * self.deltat, 

tmin + (n - nshort) * self.deltat) 

 

def peaks(self, threshold, tsearch, 

deadtime=False, 

nblock_duration_detection=100): 

 

''' 

Detect peaks above a given threshold (method 1). 

 

From every instant, where the signal rises above ``threshold``, a time 

length of ``tsearch`` seconds is searched for a maximum. A list with 

tuples (time, value) for each detected peak is returned. The 

``deadtime`` argument turns on a special deadtime duration detection 

algorithm useful in combination with recursive STA/LTA filters. 

''' 

 

y = self.ydata 

above = num.where(y > threshold, 1, 0) 

deriv = num.zeros(y.size, dtype=num.int8) 

deriv[1:] = above[1:]-above[:-1] 

itrig_positions = num.nonzero(deriv > 0)[0] 

tpeaks = [] 

apeaks = [] 

tzeros = [] 

tzero = self.tmin 

 

for itrig_pos in itrig_positions: 

ibeg = itrig_pos 

iend = min( 

len(self.ydata), 

itrig_pos + int(math.ceil(tsearch/self.deltat))) 

ipeak = num.argmax(y[ibeg:iend]) 

tpeak = self.tmin + (ipeak+ibeg)*self.deltat 

apeak = y[ibeg+ipeak] 

 

if tpeak < tzero: 

continue 

 

if deadtime: 

ibeg = itrig_pos 

iblock = 0 

nblock = nblock_duration_detection 

totalsum = 0. 

while True: 

if ibeg+iblock*nblock >= len(y): 

tzero = self.tmin + (len(y)-1) * self.deltat 

break 

 

logy = num.log( 

y[ibeg+iblock*nblock:ibeg+(iblock+1)*nblock]) 

logy[0] += totalsum 

ysum = num.cumsum(logy) 

totalsum = ysum[-1] 

below = num.where(ysum <= 0., 1, 0) 

deriv = num.zeros(ysum.size, dtype=num.int8) 

deriv[1:] = below[1:]-below[:-1] 

izero_positions = num.nonzero(deriv > 0)[0] + iblock*nblock 

if len(izero_positions) > 0: 

tzero = self.tmin + self.deltat * ( 

ibeg + izero_positions[0]) 

break 

iblock += 1 

else: 

tzero = ibeg*self.deltat + self.tmin + tsearch 

 

tpeaks.append(tpeak) 

apeaks.append(apeak) 

tzeros.append(tzero) 

 

if deadtime: 

return tpeaks, apeaks, tzeros 

else: 

return tpeaks, apeaks 

 

def peaks2(self, threshold, tsearch): 

 

''' 

Detect peaks above a given threshold (method 2). 

 

This variant of peak detection is a bit more robust (and slower) than 

the one implemented in :py:meth:`Trace.peaks`. First all samples with 

``a[i-1] < a[i] > a[i+1]`` are masked as potential peaks. From these, 

iteratively the one with the maximum amplitude ``a[j]`` and time 

``t[j]`` is choosen and potential peaks within 

``t[j] - tsearch, t[j] + tsearch`` 

are discarded. The algorithm stops, when ``a[j] < threshold`` or when 

no more potential peaks are left. 

''' 

 

a = num.copy(self.ydata) 

 

amin = num.min(a) 

 

a[0] = amin 

a[1: -1][num.diff(a, 2) <= 0.] = amin 

a[-1] = amin 

 

data = [] 

while True: 

imax = num.argmax(a) 

amax = a[imax] 

 

if amax < threshold or amax == amin: 

break 

 

data.append((self.tmin + imax * self.deltat, amax)) 

 

ntsearch = int(round(tsearch / self.deltat)) 

a[max(imax-ntsearch//2, 0):min(imax+ntsearch//2, a.size)] = amin 

 

if data: 

data.sort() 

tpeaks, apeaks = list(zip(*data)) 

else: 

tpeaks, apeaks = [], [] 

 

return tpeaks, apeaks 

 

def extend(self, tmin=None, tmax=None, fillmethod='zeros'): 

''' 

Extend trace to given span. 

 

:param tmin: begin time of new span 

:param tmax: end time of new span 

:param fillmethod: ``'zeros'``, ``'repeat'``, ``'mean'``, or 

``'median'`` 

''' 

 

nold = self.ydata.size 

 

if tmin is not None: 

nl = min(0, int(round((tmin-self.tmin)/self.deltat))) 

else: 

nl = 0 

 

if tmax is not None: 

nh = max(nold - 1, int(round((tmax-self.tmin)/self.deltat))) 

else: 

nh = nold - 1 

 

n = nh - nl + 1 

data = num.zeros(n, dtype=self.ydata.dtype) 

data[-nl:-nl + nold] = self.ydata 

if self.ydata.size >= 1: 

if fillmethod == 'repeat': 

data[:-nl] = self.ydata[0] 

data[-nl + nold:] = self.ydata[-1] 

elif fillmethod == 'median': 

v = num.median(self.ydata) 

data[:-nl] = v 

data[-nl + nold:] = v 

elif fillmethod == 'mean': 

v = num.mean(self.ydata) 

data[:-nl] = v 

data[-nl + nold:] = v 

 

self.drop_growbuffer() 

self.ydata = data 

 

self.tmin += nl * self.deltat 

self.tmax = self.tmin + (self.ydata.size - 1) * self.deltat 

 

self._update_ids() 

 

def transfer(self, 

tfade=0., 

freqlimits=None, 

transfer_function=None, 

cut_off_fading=True, 

invert=False): 

 

''' 

Return new trace with transfer function applied (convolution). 

 

:param tfade: rise/fall time in seconds of taper applied in timedomain 

at both ends of trace. 

:param freqlimits: 4-tuple with corner frequencies in Hz. 

:param transfer_function: FrequencyResponse object; must provide a 

method 'evaluate(freqs)', which returns the transfer function 

coefficients at the frequencies 'freqs'. 

:param cut_off_fading: whether to cut off rise/fall interval in output 

trace. 

:param invert: set to True to do a deconvolution 

''' 

 

if transfer_function is None: 

transfer_function = FrequencyResponse() 

 

if self.tmax - self.tmin <= tfade*2.: 

raise TraceTooShort( 

'Trace %s.%s.%s.%s too short for fading length setting. ' 

'trace length = %g, fading length = %g' 

% (self.nslc_id + (self.tmax-self.tmin, tfade))) 

 

if freqlimits is None and ( 

transfer_function is None or transfer_function.is_scalar()): 

 

# special case for flat responses 

 

output = self.copy() 

data = self.ydata 

ndata = data.size 

 

if transfer_function is not None: 

c = num.abs(transfer_function.evaluate(num.ones(1))[0]) 

 

if invert: 

c = 1.0/c 

 

data *= c 

 

if tfade != 0.0: 

data *= costaper( 

0., tfade, self.deltat*(ndata-1)-tfade, self.deltat*ndata, 

ndata, self.deltat) 

 

output.ydata = data 

 

else: 

ndata = self.ydata.size 

ntrans = nextpow2(ndata*1.2) 

coefs = self._get_tapered_coefs( 

ntrans, freqlimits, transfer_function, invert=invert) 

 

data = self.ydata 

 

data_pad = num.zeros(ntrans, dtype=num.float) 

data_pad[:ndata] = data - data.mean() 

if tfade != 0.0: 

data_pad[:ndata] *= costaper( 

0., tfade, self.deltat*(ndata-1)-tfade, self.deltat*ndata, 

ndata, self.deltat) 

 

fdata = num.fft.rfft(data_pad) 

fdata *= coefs 

ddata = num.fft.irfft(fdata) 

output = self.copy() 

output.ydata = ddata[:ndata] 

 

if cut_off_fading and tfade != 0.0: 

try: 

output.chop(output.tmin+tfade, output.tmax-tfade, inplace=True) 

except NoData: 

raise TraceTooShort( 

'Trace %s.%s.%s.%s too short for fading length setting. ' 

'trace length = %g, fading length = %g' 

% (self.nslc_id + (self.tmax-self.tmin, tfade))) 

else: 

output.ydata = output.ydata.copy() 

 

return output 

 

def drop_chain_cache(self): 

if self._pchain: 

self._pchain.clear() 

 

def init_chain(self): 

self._pchain = pchain.Chain( 

do_downsample, 

do_extend, 

do_pre_taper, 

do_fft, 

do_filter, 

do_ifft) 

 

def run_chain(self, tmin, tmax, deltat, setup, nocache): 

if setup.domain == 'frequency_domain': 

_, _, data = self._pchain( 

(self, deltat), 

(tmin, tmax), 

(setup.taper,), 

(setup.filter,), 

(setup.filter,), 

nocache=nocache) 

 

return num.abs(data), num.abs(data) 

 

else: 

processed = self._pchain( 

(self, deltat), 

(tmin, tmax), 

(setup.taper,), 

(setup.filter,), 

(setup.filter,), 

(), 

nocache=nocache) 

 

if setup.domain == 'time_domain': 

data = processed.get_ydata() 

 

elif setup.domain == 'envelope': 

processed = processed.envelope(inplace=False) 

 

elif setup.domain == 'absolute': 

processed.set_ydata(num.abs(processed.get_ydata())) 

 

return processed.get_ydata(), processed 

 

def misfit(self, candidate, setup, nocache=False, debug=False): 

""" 

Calculate misfit and normalization factor against candidate trace. 

 

:param candidate: :py:class:`Trace` object 

:param setup: :py:class:`MisfitSetup` object 

:returns: tuple ``(m, n)``, where m is the misfit value and n is the 

normalization divisor 

 

If the sampling rates of ``self`` and ``candidate`` differ, the trace 

with the higher sampling rate will be downsampled. 

""" 

 

a = self 

b = candidate 

 

for tr in (a, b): 

if not tr._pchain: 

tr.init_chain() 

 

deltat = max(a.deltat, b.deltat) 

tmin = min(a.tmin, b.tmin) - deltat 

tmax = max(a.tmax, b.tmax) + deltat 

 

adata, aproc = a.run_chain(tmin, tmax, deltat, setup, nocache) 

bdata, bproc = b.run_chain(tmin, tmax, deltat, setup, nocache) 

 

if setup.domain != 'cc_max_norm': 

m, n = Lx_norm(bdata, adata, norm=setup.norm) 

else: 

ctr = correlate(aproc, bproc, mode='full', normalization='normal') 

ccmax = ctr.max()[1] 

m = 0.5 - 0.5 * ccmax 

n = 0.5 

 

if debug: 

return m, n, aproc, bproc 

else: 

return m, n 

 

def spectrum(self, pad_to_pow2=False, tfade=None): 

''' 

Get FFT spectrum of trace. 

 

:param pad_to_pow2: whether to zero-pad the data to next larger 

power-of-two length 

:param tfade: ``None`` or a time length in seconds, to apply cosine 

shaped tapers to both 

 

:returns: a tuple with (frequencies, values) 

''' 

 

ndata = self.ydata.size 

 

if pad_to_pow2: 

ntrans = nextpow2(ndata) 

else: 

ntrans = ndata 

 

if tfade is None: 

ydata = self.ydata 

else: 

ydata = self.ydata * costaper( 

0., tfade, self.deltat*(ndata-1)-tfade, self.deltat*ndata, 

ndata, self.deltat) 

 

fydata = num.fft.rfft(ydata, ntrans) 

df = 1./(ntrans*self.deltat) 

fxdata = num.arange(len(fydata))*df 

return fxdata, fydata 

 

def multi_filter(self, filter_freqs, bandwidth): 

 

class Gauss(FrequencyResponse): 

def __init__(self, f0, a=1.0): 

self._omega0 = 2.*math.pi*f0 

self._a = a 

 

def evaluate(self, freqs): 

omega = 2.*math.pi*freqs 

return num.exp(-((omega-self._omega0) 

/ (self._a*self._omega0))**2) 

 

freqs, coefs = self.spectrum() 

n = self.data_len() 

nfilt = len(filter_freqs) 

signal_tf = num.zeros((nfilt, n)) 

centroid_freqs = num.zeros(nfilt) 

for ifilt, f0 in enumerate(filter_freqs): 

taper = Gauss(f0, a=bandwidth) 

weights = taper.evaluate(freqs) 

nhalf = freqs.size 

analytic_spec = num.zeros(n, dtype=num.complex) 

analytic_spec[:nhalf] = coefs*weights 

 

enorm = num.abs(analytic_spec[:nhalf])**2 

enorm /= num.sum(enorm) 

 

if n % 2 == 0: 

analytic_spec[1:nhalf-1] *= 2. 

else: 

analytic_spec[1:nhalf] *= 2. 

 

analytic = num.fft.ifft(analytic_spec) 

signal_tf[ifilt, :] = num.abs(analytic) 

 

enorm = num.abs(analytic_spec[:nhalf])**2 

enorm /= num.sum(enorm) 

centroid_freqs[ifilt] = num.sum(freqs*enorm) 

 

return centroid_freqs, signal_tf 

 

def _get_tapered_coefs( 

self, ntrans, freqlimits, transfer_function, invert=False): 

 

deltaf = 1./(self.deltat*ntrans) 

nfreqs = ntrans//2 + 1 

transfer = num.ones(nfreqs, dtype=num.complex) 

hi = snapper(nfreqs, deltaf) 

if freqlimits is not None: 

a, b, c, d = freqlimits 

freqs = num.arange(hi(d)-hi(a), dtype=num.float)*deltaf \ 

+ hi(a)*deltaf 

 

if invert: 

coeffs = transfer_function.evaluate(freqs) 

if num.any(coeffs == 0.0): 

raise InfiniteResponse('%s.%s.%s.%s' % self.nslc_id) 

 

transfer[hi(a):hi(d)] = 1.0 / transfer_function.evaluate(freqs) 

else: 

transfer[hi(a):hi(d)] = transfer_function.evaluate(freqs) 

 

tapered_transfer = costaper(a, b, c, d, nfreqs, deltaf)*transfer 

else: 

if invert: 

raise Exception( 

'transfer: `freqlimits` must be given when `invert` is ' 

'set to `True`') 

 

freqs = num.arange(nfreqs) * deltaf 

tapered_transfer = transfer_function.evaluate(freqs) 

 

tapered_transfer[0] = 0.0 # don't introduce static offsets 

return tapered_transfer 

 

def fill_template(self, template, **additional): 

''' 

Fill string template with trace metadata. 

 

Uses normal python '%(placeholder)s' string templates. The following 

placeholders are considered: ``network``, ``station``, ``location``, 

``channel``, ``tmin`` (time of first sample), ``tmax`` (time of last 

sample), ``tmin_ms``, ``tmax_ms``, ``tmin_us``, ``tmax_us``, 

``tmin_year``, ``tmax_year``, ``julianday``. The variants ending with 

``'_ms'`` include milliseconds, those with ``'_us'`` include 

microseconds, those with ``'_year'`` contain only the year. 

''' 

 

template = template.replace('%n', '%(network)s')\ 

.replace('%s', '%(station)s')\ 

.replace('%l', '%(location)s')\ 

.replace('%c', '%(channel)s')\ 

.replace('%b', '%(tmin)s')\ 

.replace('%e', '%(tmax)s')\ 

.replace('%j', '%(julianday)s') 

 

params = dict( 

zip(('network', 'station', 'location', 'channel'), self.nslc_id)) 

params['tmin'] = util.time_to_str( 

self.tmin, format='%Y-%m-%d_%H-%M-%S') 

params['tmax'] = util.time_to_str( 

self.tmax, format='%Y-%m-%d_%H-%M-%S') 

params['tmin_ms'] = util.time_to_str( 

self.tmin, format='%Y-%m-%d_%H-%M-%S.3FRAC') 

params['tmax_ms'] = util.time_to_str( 

self.tmax, format='%Y-%m-%d_%H-%M-%S.3FRAC') 

params['tmin_us'] = util.time_to_str( 

self.tmin, format='%Y-%m-%d_%H-%M-%S.6FRAC') 

params['tmax_us'] = util.time_to_str( 

self.tmax, format='%Y-%m-%d_%H-%M-%S.6FRAC') 

params['tmin_year'] = util.time_to_str( 

self.tmin, format='%Y') 

params['tmax_year'] = util.time_to_str( 

self.tmax, format='%Y') 

params['julianday'] = util.julian_day_of_year(self.tmin) 

params.update(additional) 

return template % params 

 

def plot(self): 

''' 

Show trace with matplotlib. 

 

See also: :py:meth:`Trace.snuffle`. 

''' 

 

import pylab 

pylab.plot(self.get_xdata(), self.get_ydata()) 

name = '%s %s %s - %s' % ( 

self.channel, 

self.station, 

time.strftime("%d-%m-%y %H:%M:%S", time.gmtime(self.tmin)), 

time.strftime("%d-%m-%y %H:%M:%S", time.gmtime(self.tmax))) 

 

pylab.title(name) 

pylab.show() 

 

def snuffle(self, **kwargs): 

''' 

Show trace in a snuffler window. 

 

:param stations: list of `pyrocko.model.Station` objects or ``None`` 

:param events: list of `pyrocko.model.Event` objects or ``None`` 

:param markers: list of `pyrocko.gui.util.Marker` objects or ``None`` 

:param ntracks: float, number of tracks to be shown initially (default: 

12) 

:param follow: time interval (in seconds) for real time follow mode or 

``None`` 

:param controls: bool, whether to show the main controls (default: 

``True``) 

:param opengl: bool, whether to use opengl (default: ``False``) 

''' 

 

return snuffle([self], **kwargs) 

 

 

def snuffle(traces, **kwargs): 

''' 

Show traces in a snuffler window. 

 

:param stations: list of `pyrocko.model.Station` objects or ``None`` 

:param events: list of `pyrocko.model.Event` objects or ``None`` 

:param markers: list of `pyrocko.gui.util.Marker` objects or ``None`` 

:param ntracks: float, number of tracks to be shown initially (default: 12) 

:param follow: time interval (in seconds) for real time follow mode or 

``None`` 

:param controls: bool, whether to show the main controls (default: 

``True``) 

:param opengl: bool, whether to use opengl (default: ``False``) 

''' 

 

from pyrocko import pile 

from pyrocko.gui import snuffler 

p = pile.Pile() 

if traces: 

trf = pile.MemTracesFile(None, traces) 

p.add_file(trf) 

return snuffler.snuffle(p, **kwargs) 

 

 

class InfiniteResponse(Exception): 

''' 

This exception is raised by :py:class:`Trace` operations when deconvolution 

of a frequency response (instrument response transfer function) would 

result in a division by zero. 

''' 

 

 

class MisalignedTraces(Exception): 

''' 

This exception is raised by some :py:class:`Trace` operations when tmin, 

tmax or number of samples do not match. 

''' 

 

pass 

 

 

class NoData(Exception): 

''' 

This exception is raised by some :py:class:`Trace` operations when no or 

not enough data is available. 

''' 

 

pass 

 

 

class AboveNyquist(Exception): 

''' 

This exception is raised by some :py:class:`Trace` operations when given 

frequencies are above the Nyquist frequency. 

''' 

 

pass 

 

 

class TraceTooShort(Exception): 

''' 

This exception is raised by some :py:class:`Trace` operations when the 

trace is too short. 

''' 

 

pass 

 

 

class ResamplingFailed(Exception): 

pass 

 

 

def minmax(traces, key=None, mode='minmax'): 

 

''' 

Get data range given traces grouped by selected pattern. 

 

:param key: a callable which takes as single argument a trace and returns a 

key for the grouping of the results. If this is ``None``, the default, 

``lambda tr: (tr.network, tr.station, tr.location, tr.channel)`` is 

used. 

:param mode: 'minmax' or floating point number. If this is 'minmax', 

minimum and maximum of the traces are used, if it is a number, mean +- 

standard deviation times ``mode`` is used. 

 

:returns: a dict with the combined data ranges. 

 

Examples:: 

 

ranges = minmax(traces, lambda tr: tr.channel) 

print ranges['N'] # print min & max of all traces with channel == 'N' 

print ranges['E'] # print min & max of all traces with channel == 'E' 

 

ranges = minmax(traces, lambda tr: (tr.network, tr.station)) 

print ranges['GR', 'HAM3'] # print min & max of all traces with 

# network == 'GR' and station == 'HAM3' 

 

ranges = minmax(traces, lambda tr: None) 

print ranges[None] # prints min & max of all traces 

''' 

 

if key is None: 

key = _default_key 

 

ranges = {} 

for trace in traces: 

if isinstance(mode, str) and mode == 'minmax': 

mi, ma = trace.ydata.min(), trace.ydata.max() 

else: 

mean = trace.ydata.mean() 

std = trace.ydata.std() 

mi, ma = mean-std*mode, mean+std*mode 

 

k = key(trace) 

if k not in ranges: 

ranges[k] = mi, ma 

else: 

tmi, tma = ranges[k] 

ranges[k] = min(tmi, mi), max(tma, ma) 

 

return ranges 

 

 

def minmaxtime(traces, key=None): 

 

''' 

Get time range given traces grouped by selected pattern. 

 

:param key: a callable which takes as single argument a trace and returns a 

key for the grouping of the results. If this is ``None``, the default, 

``lambda tr: (tr.network, tr.station, tr.location, tr.channel)`` is 

used. 

 

:returns: a dict with the combined data ranges. 

''' 

 

if key is None: 

key = _default_key 

 

ranges = {} 

for trace in traces: 

mi, ma = trace.tmin, trace.tmax 

k = key(trace) 

if k not in ranges: 

ranges[k] = mi, ma 

else: 

tmi, tma = ranges[k] 

ranges[k] = min(tmi, mi), max(tma, ma) 

 

return ranges 

 

 

def degapper( 

traces, 

maxgap=5, 

fillmethod='interpolate', 

deoverlap='use_second', 

maxlap=None): 

 

''' 

Try to connect traces and remove gaps. 

 

This method will combine adjacent traces, which match in their network, 

station, location and channel attributes. Overlapping parts are handled 

according to the ``deoverlap`` argument. 

 

:param traces: input traces, must be sorted by their full_id attribute. 

:param maxgap: maximum number of samples to interpolate. 

:param fillmethod: what to put into the gaps: 'interpolate' or 'zeros'. 

:param deoverlap: how to handle overlaps: 'use_second' to use data from 

second trace (default), 'use_first' to use data from first trace, 

'crossfade_cos' to crossfade with cosine taper, 'add' to add amplitude 

values. 

:param maxlap: maximum number of samples of overlap which are removed 

 

:returns: list of traces 

''' 

 

in_traces = traces 

out_traces = [] 

if not in_traces: 

return out_traces 

out_traces.append(in_traces.pop(0)) 

while in_traces: 

 

a = out_traces[-1] 

b = in_traces.pop(0) 

 

avirt, bvirt = a.ydata is None, b.ydata is None 

assert avirt == bvirt, \ 

'traces given to degapper() must either all have data or have ' \ 

'no data.' 

 

virtual = avirt and bvirt 

 

if (a.nslc_id == b.nslc_id and a.deltat == b.deltat 

and a.data_len() >= 1 and b.data_len() >= 1 

and (virtual or a.ydata.dtype == b.ydata.dtype)): 

 

dist = (b.tmin-(a.tmin+(a.data_len()-1)*a.deltat))/a.deltat 

idist = int(round(dist)) 

if abs(dist - idist) > 0.05 and idist <= maxgap: 

# logger.warning('Cannot degap traces with displaced sampling ' 

# '(%s, %s, %s, %s)' % a.nslc_id) 

pass 

else: 

if 1 < idist <= maxgap: 

if not virtual: 

if fillmethod == 'interpolate': 

filler = a.ydata[-1] + ( 

((1.0 + num.arange(idist-1, dtype=num.float)) 

/ idist) * (b.ydata[0]-a.ydata[-1]) 

).astype(a.ydata.dtype) 

elif fillmethod == 'zeros': 

filler = num.zeros(idist-1, dtype=a.ydist.dtype) 

a.ydata = num.concatenate((a.ydata, filler, b.ydata)) 

a.tmax = b.tmax 

if a.mtime and b.mtime: 

a.mtime = max(a.mtime, b.mtime) 

continue 

 

elif idist == 1: 

if not virtual: 

a.ydata = num.concatenate((a.ydata, b.ydata)) 

a.tmax = b.tmax 

if a.mtime and b.mtime: 

a.mtime = max(a.mtime, b.mtime) 

continue 

 

elif idist <= 0 and (maxlap is None or -maxlap < idist): 

if b.tmax > a.tmax: 

if not virtual: 

na = a.ydata.size 

n = -idist+1 

if deoverlap == 'use_second': 

a.ydata = num.concatenate( 

(a.ydata[:-n], b.ydata)) 

elif deoverlap in ('use_first', 'crossfade_cos'): 

a.ydata = num.concatenate( 

(a.ydata, b.ydata[n:])) 

elif deoverlap == 'add': 

a.ydata[-n:] += b.ydata[:n] 

a.ydata = num.concatenate( 

(a.ydata, b.ydata[n:])) 

else: 

assert False, 'unknown deoverlap method' 

 

if deoverlap == 'crossfade_cos': 

n = -idist+1 

taper = 0.5-0.5*num.cos( 

(1.+num.arange(n))/(1.+n)*num.pi) 

a.ydata[na-n:na] *= 1.-taper 

a.ydata[na-n:na] += b.ydata[:n] * taper 

 

a.tmax = b.tmax 

if a.mtime and b.mtime: 

a.mtime = max(a.mtime, b.mtime) 

continue 

else: 

# make short second trace vanish 

continue 

 

if b.data_len() >= 1: 

out_traces.append(b) 

 

for tr in out_traces: 

tr._update_ids() 

 

return out_traces 

 

 

def rotate(traces, azimuth, in_channels, out_channels): 

''' 

2D rotation of traces. 

 

:param traces: list of input traces 

:param azimuth: difference of the azimuths of the component directions 

(azimuth of out_channels[0]) - (azimuth of in_channels[0]) 

:param in_channels: names of the input channels (e.g. 'N', 'E') 

:param out_channels: names of the output channels (e.g. 'R', 'T') 

:returns: list of rotated traces 

''' 

 

phi = azimuth/180.*math.pi 

cphi = math.cos(phi) 

sphi = math.sin(phi) 

rotated = [] 

in_channels = tuple(_channels_to_names(in_channels)) 

out_channels = tuple(_channels_to_names(out_channels)) 

for a in traces: 

for b in traces: 

if ((a.channel, b.channel) == in_channels and 

a.nslc_id[:3] == b.nslc_id[:3] and 

abs(a.deltat-b.deltat) < a.deltat*0.001): 

tmin = max(a.tmin, b.tmin) 

tmax = min(a.tmax, b.tmax) 

 

if tmin < tmax: 

ac = a.chop(tmin, tmax, inplace=False, include_last=True) 

bc = b.chop(tmin, tmax, inplace=False, include_last=True) 

if abs(ac.tmin - bc.tmin) > ac.deltat*0.01: 

logger.warning( 

'Cannot rotate traces with displaced sampling ' 

'(%s, %s, %s, %s)' % a.nslc_id) 

continue 

 

acydata = ac.get_ydata()*cphi+bc.get_ydata()*sphi 

bcydata = -ac.get_ydata()*sphi+bc.get_ydata()*cphi 

ac.set_ydata(acydata) 

bc.set_ydata(bcydata) 

 

ac.set_codes(channel=out_channels[0]) 

bc.set_codes(channel=out_channels[1]) 

rotated.append(ac) 

rotated.append(bc) 

 

return rotated 

 

 

def rotate_to_rt(n, e, source, receiver, out_channels=('R', 'T')): 

azimuth = orthodrome.azimuth(receiver, source) + 180. 

in_channels = n.channel, e.channel 

out = rotate( 

[n, e], azimuth, 

in_channels=in_channels, 

out_channels=out_channels) 

 

assert len(out) == 2 

for tr in out: 

if tr.channel == 'R': 

r = tr 

elif tr.channel == 'T': 

t = tr 

 

return r, t 

 

 

def rotate_to_lqt(traces, backazimuth, incidence, in_channels, 

out_channels=('L', 'Q', 'T')): 

'''Rotate traces from ZNE to LQT system. 

 

:param traces: list of traces in arbitrary order 

:param backazimuth: backazimuth in degrees clockwise from north 

:param incidence: incidence angle in degrees from vertical 

:param in_channels: input channel names 

:param out_channels: output channel names (default: ('L', 'Q', 'T')) 

:returns: list of transformed traces 

''' 

i = incidence/180.*num.pi 

b = backazimuth/180.*num.pi 

 

ci = num.cos(i) 

cb = num.cos(b) 

si = num.sin(i) 

sb = num.sin(b) 

 

rotmat = num.array( 

[[ci, -cb*si, -sb*si], [si, cb*ci, sb*ci], [0., sb, -cb]]) 

return project(traces, rotmat, in_channels, out_channels) 

 

 

def _decompose(a): 

''' 

Decompose matrix into independent submatrices. 

''' 

 

def depends(iout, a): 

row = a[iout, :] 

return set(num.arange(row.size).compress(row != 0.0)) 

 

def provides(iin, a): 

col = a[:, iin] 

return set(num.arange(col.size).compress(col != 0.0)) 

 

a = num.asarray(a) 

outs = set(range(a.shape[0])) 

systems = [] 

while outs: 

iout = outs.pop() 

 

gout = set() 

for iin in depends(iout, a): 

gout.update(provides(iin, a)) 

 

if not gout: 

continue 

 

gin = set() 

for iout2 in gout: 

gin.update(depends(iout2, a)) 

 

if not gin: 

continue 

 

for iout2 in gout: 

if iout2 in outs: 

outs.remove(iout2) 

 

gin = list(gin) 

gin.sort() 

gout = list(gout) 

gout.sort() 

 

systems.append((gin, gout, a[gout, :][:, gin])) 

 

return systems 

 

 

def _channels_to_names(channels): 

names = [] 

for ch in channels: 

if isinstance(ch, model.Channel): 

names.append(ch.name) 

else: 

names.append(ch) 

return names 

 

 

def project(traces, matrix, in_channels, out_channels): 

''' 

Affine transform of three-component traces. 

 

Compute matrix-vector product of three-component traces, to e.g. rotate 

traces into a different basis. The traces are distinguished and ordered by 

their channel attribute. The tranform is applied to overlapping parts of 

any appropriate combinations of the input traces. This should allow this 

function to be robust with data gaps. It also tries to apply the 

tranformation to subsets of the channels, if this is possible, so that, if 

for example a vertical compontent is missing, horizontal components can 

still be rotated. 

 

:param traces: list of traces in arbitrary order 

:param matrix: tranformation matrix 

:param in_channels: input channel names 

:param out_channels: output channel names 

:returns: list of transformed traces 

''' 

 

in_channels = tuple(_channels_to_names(in_channels)) 

out_channels = tuple(_channels_to_names(out_channels)) 

systems = _decompose(matrix) 

 

# fallback to full matrix if some are not quadratic 

for iins, iouts, submatrix in systems: 

if submatrix.shape[0] != submatrix.shape[1]: 

return _project3(traces, matrix, in_channels, out_channels) 

 

projected = [] 

for iins, iouts, submatrix in systems: 

in_cha = tuple([in_channels[iin] for iin in iins]) 

out_cha = tuple([out_channels[iout] for iout in iouts]) 

if submatrix.shape[0] == 1: 

projected.extend(_project1(traces, submatrix, in_cha, out_cha)) 

elif submatrix.shape[1] == 2: 

projected.extend(_project2(traces, submatrix, in_cha, out_cha)) 

else: 

projected.extend(_project3(traces, submatrix, in_cha, out_cha)) 

 

return projected 

 

 

def project_dependencies(matrix, in_channels, out_channels): 

''' 

Figure out what dependencies project() would produce. 

''' 

 

in_channels = tuple(_channels_to_names(in_channels)) 

out_channels = tuple(_channels_to_names(out_channels)) 

systems = _decompose(matrix) 

 

subpro = [] 

for iins, iouts, submatrix in systems: 

if submatrix.shape[0] != submatrix.shape[1]: 

subpro.append((matrix, in_channels, out_channels)) 

 

if not subpro: 

for iins, iouts, submatrix in systems: 

in_cha = tuple([in_channels[iin] for iin in iins]) 

out_cha = tuple([out_channels[iout] for iout in iouts]) 

subpro.append((submatrix, in_cha, out_cha)) 

 

deps = {} 

for mat, in_cha, out_cha in subpro: 

for oc in out_cha: 

if oc not in deps: 

deps[oc] = [] 

 

for ic in in_cha: 

deps[oc].append(ic) 

 

return deps 

 

 

def _project1(traces, matrix, in_channels, out_channels): 

assert len(in_channels) == 1 

assert len(out_channels) == 1 

assert matrix.shape == (1, 1) 

 

projected = [] 

for a in traces: 

if not (a.channel,) == in_channels: 

continue 

 

ac = a.copy() 

ac.set_ydata(matrix[0, 0]*a.get_ydata()) 

ac.set_codes(channel=out_channels[0]) 

projected.append(ac) 

 

return projected 

 

 

def _project2(traces, matrix, in_channels, out_channels): 

assert len(in_channels) == 2 

assert len(out_channels) == 2 

assert matrix.shape == (2, 2) 

projected = [] 

for a in traces: 

for b in traces: 

if not ((a.channel, b.channel) == in_channels and 

a.nslc_id[:3] == b.nslc_id[:3] and 

abs(a.deltat-b.deltat) < a.deltat*0.001): 

continue 

 

tmin = max(a.tmin, b.tmin) 

tmax = min(a.tmax, b.tmax) 

 

if tmin > tmax: 

continue 

 

ac = a.chop(tmin, tmax, inplace=False, include_last=True) 

bc = b.chop(tmin, tmax, inplace=False, include_last=True) 

if abs(ac.tmin - bc.tmin) > ac.deltat*0.01: 

logger.warning( 

'Cannot project traces with displaced sampling ' 

'(%s, %s, %s, %s)' % a.nslc_id) 

continue 

 

acydata = num.dot(matrix[0], (ac.get_ydata(), bc.get_ydata())) 

bcydata = num.dot(matrix[1], (ac.get_ydata(), bc.get_ydata())) 

 

ac.set_ydata(acydata) 

bc.set_ydata(bcydata) 

 

ac.set_codes(channel=out_channels[0]) 

bc.set_codes(channel=out_channels[1]) 

 

projected.append(ac) 

projected.append(bc) 

 

return projected 

 

 

def _project3(traces, matrix, in_channels, out_channels): 

assert len(in_channels) == 3 

assert len(out_channels) == 3 

assert matrix.shape == (3, 3) 

projected = [] 

for a in traces: 

for b in traces: 

for c in traces: 

if not ((a.channel, b.channel, c.channel) == in_channels 

and a.nslc_id[:3] == b.nslc_id[:3] 

and b.nslc_id[:3] == c.nslc_id[:3] 

and abs(a.deltat-b.deltat) < a.deltat*0.001 

and abs(b.deltat-c.deltat) < b.deltat*0.001): 

 

continue 

 

tmin = max(a.tmin, b.tmin, c.tmin) 

tmax = min(a.tmax, b.tmax, c.tmax) 

 

if tmin >= tmax: 

continue 

 

ac = a.chop(tmin, tmax, inplace=False, include_last=True) 

bc = b.chop(tmin, tmax, inplace=False, include_last=True) 

cc = c.chop(tmin, tmax, inplace=False, include_last=True) 

if (abs(ac.tmin - bc.tmin) > ac.deltat*0.01 

or abs(bc.tmin - cc.tmin) > bc.deltat*0.01): 

 

logger.warning( 

'Cannot project traces with displaced sampling ' 

'(%s, %s, %s, %s)' % a.nslc_id) 

continue 

 

acydata = num.dot( 

matrix[0], 

(ac.get_ydata(), bc.get_ydata(), cc.get_ydata())) 

bcydata = num.dot( 

matrix[1], 

(ac.get_ydata(), bc.get_ydata(), cc.get_ydata())) 

ccydata = num.dot( 

matrix[2], 

(ac.get_ydata(), bc.get_ydata(), cc.get_ydata())) 

 

ac.set_ydata(acydata) 

bc.set_ydata(bcydata) 

cc.set_ydata(ccydata) 

 

ac.set_codes(channel=out_channels[0]) 

bc.set_codes(channel=out_channels[1]) 

cc.set_codes(channel=out_channels[2]) 

 

projected.append(ac) 

projected.append(bc) 

projected.append(cc) 

 

return projected 

 

 

def correlate(a, b, mode='valid', normalization=None, use_fft=False): 

''' 

Cross correlation of two traces. 

 

:param a,b: input traces 

:param mode: ``'valid'``, ``'full'``, or ``'same'`` 

:param normalization: ``'normal'``, ``'gliding'``, or ``None`` 

:param use_fft: bool, whether to do cross correlation in spectral domain 

 

:returns: trace containing cross correlation coefficients 

 

This function computes the cross correlation between two traces. It 

evaluates the discrete equivalent of 

 

.. math:: 

 

c(t) = \\int_{-\\infty}^{\\infty} a^{\\ast}(\\tau) b(t+\\tau) d\\tau 

 

where the star denotes complex conjugate. Note, that the arguments here are 

swapped when compared with the :py:func:`numpy.correlate` function, 

which is internally called. This function should be safe even with older 

versions of NumPy, where the correlate function has some problems. 

 

A trace containing the cross correlation coefficients is returned. The time 

information of the output trace is set so that the returned cross 

correlation can be viewed directly as a function of time lag. 

 

Example:: 

 

# align two traces a and b containing a time shifted similar signal: 

c = pyrocko.trace.correlate(a,b) 

t, coef = c.max() # get time and value of maximum 

b.shift(-t) # align b with a 

 

''' 

 

assert_same_sampling_rate(a, b) 

 

ya, yb = a.ydata, b.ydata 

 

# need reversed order here: 

yc = numpy_correlate_fixed(yb, ya, mode=mode, use_fft=use_fft) 

kmin, kmax = numpy_correlate_lag_range(yb, ya, mode=mode, use_fft=use_fft) 

 

if normalization == 'normal': 

normfac = num.sqrt(num.sum(ya**2))*num.sqrt(num.sum(yb**2)) 

yc = yc/normfac 

 

elif normalization == 'gliding': 

if mode != 'valid': 

assert False, 'gliding normalization currently only available ' \ 

'with "valid" mode.' 

 

if ya.size < yb.size: 

yshort, ylong = ya, yb 

else: 

yshort, ylong = yb, ya 

 

epsilon = 0.00001 

normfac_short = num.sqrt(num.sum(yshort**2)) 

normfac = normfac_short * num.sqrt( 

moving_sum(ylong**2, yshort.size, mode='valid')) \ 

+ normfac_short*epsilon 

 

if yb.size <= ya.size: 

normfac = normfac[::-1] 

 

yc /= normfac 

 

c = a.copy() 

c.set_ydata(yc) 

c.set_codes(*merge_codes(a, b, '~')) 

c.shift(-c.tmin + b.tmin-a.tmin + kmin * c.deltat) 

 

return c 

 

 

def deconvolve( 

a, b, waterlevel, 

tshift=0., 

pad=0.5, 

fd_taper=None, 

pad_to_pow2=True): 

 

same_sampling_rate(a, b) 

assert abs(a.tmin - b.tmin) < a.deltat * 0.001 

deltat = a.deltat 

npad = int(round(a.data_len()*pad + tshift / deltat)) 

 

ndata = max(a.data_len(), b.data_len()) 

ndata_pad = ndata + npad 

 

if pad_to_pow2: 

ntrans = nextpow2(ndata_pad) 

else: 

ntrans = ndata 

 

aspec = num.fft.rfft(a.ydata, ntrans) 

bspec = num.fft.rfft(b.ydata, ntrans) 

 

out = aspec * num.conj(bspec) 

 

bautocorr = bspec*num.conj(bspec) 

denom = num.maximum(bautocorr, waterlevel * bautocorr.max()) 

 

out /= denom 

df = 1/(ntrans*deltat) 

 

if fd_taper is not None: 

fd_taper(out, 0.0, df) 

 

ydata = num.roll(num.fft.irfft(out), int(round(tshift/deltat))) 

c = a.copy(data=False) 

c.set_ydata(ydata[:ndata]) 

c.set_codes(*merge_codes(a, b, '/')) 

return c 

 

 

def assert_same_sampling_rate(a, b, eps=1.0e-6): 

assert same_sampling_rate(a, b, eps), \ 

'Sampling rates differ: %g != %g' % (a.deltat, b.deltat) 

 

 

def same_sampling_rate(a, b, eps=1.0e-6): 

''' 

Check if two traces have the same sampling rate. 

 

:param a,b: input traces 

:param eps: relative tolerance 

''' 

 

return abs(a.deltat - b.deltat) < (a.deltat + b.deltat)*eps 

 

 

def fix_deltat_rounding_errors(deltat): 

''' 

Try to undo sampling rate rounding errors. 

 

Fix rounding errors of sampling intervals when these are read from single 

precision floating point values. 

 

Assumes that the true sampling rate or sampling interval was an integer 

value. No correction will be applied if this would change the sampling 

rate by more than 0.001%. 

''' 

 

if deltat <= 1.0: 

deltat_new = 1.0 / round(1.0 / deltat) 

else: 

deltat_new = round(deltat) 

 

if abs(deltat_new - deltat) / deltat > 1e-5: 

deltat_new = deltat 

 

return deltat_new 

 

 

def merge_codes(a, b, sep='-'): 

''' 

Merge network-station-location-channel codes of a pair of traces. 

''' 

 

o = [] 

for xa, xb in zip(a.nslc_id, b.nslc_id): 

if xa == xb: 

o.append(xa) 

else: 

o.append(sep.join((xa, xb))) 

return o 

 

 

class Taper(Object): 

''' 

Base class for tapers. 

 

Does nothing by default. 

''' 

 

def __call__(self, y, x0, dx): 

pass 

 

 

class CosTaper(Taper): 

''' 

Cosine Taper. 

 

:param a: start of fading in 

:param b: end of fading in 

:param c: start of fading out 

:param d: end of fading out 

''' 

 

a = Float.T() 

b = Float.T() 

c = Float.T() 

d = Float.T() 

 

def __init__(self, a, b, c, d): 

Taper.__init__(self, a=a, b=b, c=c, d=d) 

 

def __call__(self, y, x0, dx): 

apply_costaper(self.a, self.b, self.c, self.d, y, x0, dx) 

 

def span(self, y, x0, dx): 

return span_costaper(self.a, self.b, self.c, self.d, y, x0, dx) 

 

def time_span(self): 

return self.a, self.d 

 

 

class CosFader(Taper): 

''' 

Cosine Fader. 

 

:param xfade: fade in and fade out time in seconds (optional) 

:param xfrac: fade in and fade out as fraction between 0. and 1. (optional) 

 

Only one argument can be set. The other should to be ``None``. 

''' 

 

xfade = Float.T(optional=True) 

xfrac = Float.T(optional=True) 

 

def __init__(self, xfade=None, xfrac=None): 

Taper.__init__(self, xfade=xfade, xfrac=xfrac) 

assert (xfade is None) != (xfrac is None) 

self._xfade = xfade 

self._xfrac = xfrac 

 

def __call__(self, y, x0, dx): 

 

xfade = self._xfade 

 

xlen = (y.size - 1)*dx 

if xfade is None: 

xfade = xlen * self._xfrac 

 

a = x0 

b = x0 + xfade 

c = x0 + xlen - xfade 

d = x0 + xlen 

 

apply_costaper(a, b, c, d, y, x0, dx) 

 

def span(self, y, x0, dx): 

return 0, y.size 

 

def time_span(self): 

return None, None 

 

 

def none_min(l): 

if None in l: 

return None 

else: 

return min(x for x in l if x is not None) 

 

 

def none_max(l): 

if None in l: 

return None 

else: 

return max(x for x in l if x is not None) 

 

 

class MultiplyTaper(Taper): 

''' 

Multiplication of several tapers. 

''' 

 

tapers = List.T(Taper.T()) 

 

def __init__(self, tapers=None): 

if tapers is None: 

tapers = [] 

 

Taper.__init__(self, tapers=tapers) 

 

def __call__(self, y, x0, dx): 

for taper in self.tapers: 

taper(y, x0, dx) 

 

def span(self, y, x0, dx): 

spans = [] 

for taper in self.tapers: 

spans.append(taper.span(y, x0, dx)) 

 

mins, maxs = list(zip(*spans)) 

return min(mins), max(maxs) 

 

def time_span(self): 

spans = [] 

for taper in self.tapers: 

spans.append(taper.time_span()) 

 

mins, maxs = list(zip(*spans)) 

return none_min(mins), none_max(maxs) 

 

 

class GaussTaper(Taper): 

''' 

Frequency domain Gaussian filter. 

''' 

 

alpha = Float.T() 

 

def __init__(self, alpha): 

Taper.__init__(self, alpha=alpha) 

self._alpha = alpha 

 

def __call__(self, y, x0, dx): 

f = x0 + num.arange(y.size)*dx 

y *= num.exp(-num.pi**2 / (self._alpha**2) * f**2) 

 

 

class FrequencyResponse(Object): 

''' 

Evaluates frequency response at given frequencies. 

''' 

 

def evaluate(self, freqs): 

coefs = num.ones(freqs.size, dtype=num.complex) 

return coefs 

 

def is_scalar(self): 

''' 

Check if this is a flat response. 

''' 

 

if type(self) == FrequencyResponse: 

return True 

else: 

return False # default for derived classes 

 

 

class Evalresp(FrequencyResponse): 

''' 

Calls evalresp and generates values of the instrument response transfer 

function. 

 

:param respfile: response file in evalresp format 

:param trace: trace for which the response is to be extracted from the file 

:param target: ``'dis'`` for displacement or ``'vel'`` for velocity 

''' 

 

respfile = String.T() 

nslc_id = Tuple.T(4, String.T()) 

target = String.T(default='dis') 

instant = Float.T() 

 

def __init__( 

self, respfile, trace=None, target='dis', nslc_id=None, time=None): 

 

if trace is not None: 

nslc_id = trace.nslc_id 

time = (trace.tmin + trace.tmax) / 2. 

 

FrequencyResponse.__init__( 

self, 

respfile=respfile, 

nslc_id=nslc_id, 

instant=time, 

target=target) 

 

def evaluate(self, freqs): 

network, station, location, channel = self.nslc_id 

x = evalresp.evalresp( 

sta_list=station, 

cha_list=channel, 

net_code=network, 

locid=location, 

instant=self.instant, 

freqs=freqs, 

units=self.target.upper(), 

file=self.respfile, 

rtype='CS') 

 

transfer = x[0][4] 

return transfer 

 

 

class InverseEvalresp(FrequencyResponse): 

''' 

Calls evalresp and generates values of the inverse instrument response for 

deconvolution of instrument response. 

 

:param respfile: response file in evalresp format 

:param trace: trace for which the response is to be extracted from the file 

:param target: ``'dis'`` for displacement or ``'vel'`` for velocity 

''' 

 

respfile = String.T() 

nslc_id = Tuple.T(4, String.T()) 

target = String.T(default='dis') 

instant = Float.T() 

 

def __init__(self, respfile, trace, target='dis'): 

FrequencyResponse.__init__( 

self, 

respfile=respfile, 

nslc_id=trace.nslc_id, 

instant=(trace.tmin + trace.tmax)/2., 

target=target) 

 

def evaluate(self, freqs): 

network, station, location, channel = self.nslc_id 

x = evalresp.evalresp(sta_list=station, 

cha_list=channel, 

net_code=network, 

locid=location, 

instant=self.instant, 

freqs=freqs, 

units=self.target.upper(), 

file=self.respfile, 

rtype='CS') 

 

transfer = x[0][4] 

return 1./transfer 

 

 

class PoleZeroResponse(FrequencyResponse): 

''' 

Evaluates frequency response from pole-zero representation. 

 

:param zeros: :py:class:`numpy.array` containing complex positions of zeros 

:param poles: :py:class:`numpy.array` containing complex positions of poles 

:param constant: gain as floating point number 

 

:: 

 

(j*2*pi*f - zeros[0]) * (j*2*pi*f - zeros[1]) * ... 

T(f) = constant * ---------------------------------------------------- 

(j*2*pi*f - poles[0]) * (j*2*pi*f - poles[1]) * ... 

 

 

The poles and zeros should be given as angular frequencies, not in Hz. 

''' 

 

zeros = List.T(Complex.T()) 

poles = List.T(Complex.T()) 

constant = Complex.T(default=1.0+0j) 

 

def __init__(self, zeros=None, poles=None, constant=1.0+0j): 

if zeros is None: 

zeros = [] 

if poles is None: 

poles = [] 

FrequencyResponse.__init__( 

self, zeros=zeros, poles=poles, constant=constant) 

 

def evaluate(self, freqs): 

jomeg = 1.0j * 2.*num.pi*freqs 

 

a = num.ones(freqs.size, dtype=num.complex)*self.constant 

for z in self.zeros: 

a *= jomeg-z 

for p in self.poles: 

a /= jomeg-p 

 

return a 

 

def is_scalar(self): 

return len(self.zeros) == 0 and len(self.poles) == 0 

 

 

class ButterworthResponse(FrequencyResponse): 

''' 

Butterworth frequency response. 

 

:param corner: corner frequency of the response 

:param order: order of the response 

:param type: either ``high`` or ``low`` 

''' 

 

corner = Float.T(default=1.0) 

order = Int.T(default=4) 

type = StringChoice.T(choices=['low', 'high'], default='low') 

 

def evaluate(self, freqs): 

b, a = signal.butter( 

int(self.order), float(self.corner), self.type, analog=True) 

w, h = signal.freqs(b, a, freqs) 

return h 

 

 

class SampledResponse(FrequencyResponse): 

''' 

Interpolates frequency response given at a set of sampled frequencies. 

 

:param frequencies,values: frequencies and values of the sampled response 

function. 

:param left,right: values to return when input is out of range. If set to 

``None`` (the default) the endpoints are returned. 

''' 

 

frequencies = Array.T(shape=(None,), dtype=num.float, serialize_as='list') 

values = Array.T(shape=(None,), dtype=num.complex, serialize_as='list') 

left = Complex.T(optional=True) 

right = Complex.T(optional=True) 

 

def __init__(self, frequencies, values, left=None, right=None): 

FrequencyResponse.__init__( 

self, 

frequencies=asarray_1d(frequencies, num.float), 

values=asarray_1d(values, num.complex)) 

 

def evaluate(self, freqs): 

ereal = num.interp( 

freqs, self.frequencies, num.real(self.values), 

left=self.left, right=self.right) 

eimag = num.interp( 

freqs, self.frequencies, num.imag(self.values), 

left=self.left, right=self.right) 

transfer = ereal + 1.0j*eimag 

return transfer 

 

def inverse(self): 

''' 

Get inverse as a new :py:class:`SampledResponse` object. 

''' 

 

def inv_or_none(x): 

if x is not None: 

return 1./x 

 

return SampledResponse( 

self.frequencies, 1./self.values, 

left=inv_or_none(self.left), 

right=inv_or_none(self.right)) 

 

 

class IntegrationResponse(FrequencyResponse): 

''' 

The integration response, optionally multiplied by a constant gain. 

 

:param n: exponent (integer) 

:param gain: gain factor (float) 

 

:: 

 

gain 

T(f) = -------------- 

(j*2*pi * f)^n 

''' 

 

n = Int.T(optional=True, default=1) 

gain = Float.T(optional=True, default=1.0) 

 

def __init__(self, n=1, gain=1.0): 

FrequencyResponse.__init__(self, n=n, gain=gain) 

 

def evaluate(self, freqs): 

nonzero = freqs != 0.0 

resp = num.empty(freqs.size, dtype=num.complex) 

resp[nonzero] = self.gain / (1.0j * 2. * num.pi*freqs[nonzero])**self.n 

resp[num.logical_not(nonzero)] = 0.0 

return resp 

 

 

class DifferentiationResponse(FrequencyResponse): 

''' 

The differentiation response, optionally multiplied by a constant gain. 

 

:param n: exponent (integer) 

:param gain: gain factor (float) 

 

:: 

 

T(f) = gain * (j*2*pi * f)^n 

''' 

 

n = Int.T(optional=True, default=1) 

gain = Float.T(optional=True, default=1.0) 

 

def __init__(self, n=1, gain=1.0): 

FrequencyResponse.__init__(self, n=n, gain=gain) 

 

def evaluate(self, freqs): 

return self.gain * (1.0j * 2. * num.pi * freqs)**self.n 

 

 

class AnalogFilterResponse(FrequencyResponse): 

''' 

Frequency response of an analog filter. 

 

(see :py:func:`scipy.signal.freqs`). 

''' 

 

b = List.T(Float.T()) 

a = List.T(Float.T()) 

 

def __init__(self, b, a): 

FrequencyResponse.__init__(self, b=b, a=a) 

 

def evaluate(self, freqs): 

return signal.freqs(self.b, self.a, freqs/(2.*num.pi))[1] 

 

 

class MultiplyResponse(FrequencyResponse): 

''' 

Multiplication of several :py:class:`FrequencyResponse` objects. 

''' 

 

responses = List.T(FrequencyResponse.T()) 

 

def __init__(self, responses=None): 

if responses is None: 

responses = [] 

FrequencyResponse.__init__(self, responses=responses) 

 

def evaluate(self, freqs): 

a = num.ones(freqs.size, dtype=num.complex) 

for resp in self.responses: 

a *= resp.evaluate(freqs) 

 

return a 

 

def is_scalar(self): 

return all(resp.is_scalar() for resp in self.responses) 

 

 

def asarray_1d(x, dtype): 

if isinstance(x, (list, tuple)) and x and isinstance(x[0], (str, newstr)): 

return num.asarray(list(map(dtype, x)), dtype=dtype) 

else: 

a = num.asarray(x, dtype=dtype) 

if not a.ndim == 1: 

raise ValueError('could not convert to 1D array') 

return a 

 

 

cached_coefficients = {} 

 

 

def _get_cached_filter_coefs(order, corners, btype): 

ck = (order, tuple(corners), btype) 

if ck not in cached_coefficients: 

if len(corners) == 0: 

cached_coefficients[ck] = signal.butter( 

order, corners[0], btype=btype) 

else: 

cached_coefficients[ck] = signal.butter( 

order, corners, btype=btype) 

 

return cached_coefficients[ck] 

 

 

class _globals(object): 

_numpy_has_correlate_flip_bug = None 

 

 

def _default_key(tr): 

return (tr.network, tr.station, tr.location, tr.channel) 

 

 

def numpy_has_correlate_flip_bug(): 

''' 

Check if NumPy's correlate function reveals old behaviour 

''' 

 

if _globals._numpy_has_correlate_flip_bug is None: 

a = num.array([0, 0, 1, 0, 0, 0, 0]) 

b = num.array([0, 0, 0, 0, 1, 0, 0, 0]) 

ab = num.correlate(a, b, mode='same') 

ba = num.correlate(b, a, mode='same') 

_globals._numpy_has_correlate_flip_bug = num.all(ab == ba) 

 

return _globals._numpy_has_correlate_flip_bug 

 

 

def numpy_correlate_fixed(a, b, mode='valid', use_fft=False): 

''' 

Call :py:func:`numpy.correlate` with fixes. 

 

c[k] = sum_i a[i+k] * conj(b[i]) 

 

Note that the result produced by newer numpy.correlate is always flipped 

with respect to the formula given in its documentation (if ascending k 

assumed for the output). 

''' 

 

if use_fft: 

if a.size < b.size: 

c = signal.fftconvolve(b[::-1], a, mode=mode) 

else: 

c = signal.fftconvolve(a, b[::-1], mode=mode) 

return c 

 

else: 

buggy = numpy_has_correlate_flip_bug() 

 

a = num.asarray(a) 

b = num.asarray(b) 

 

if buggy: 

b = num.conj(b) 

 

c = num.correlate(a, b, mode=mode) 

 

if buggy and a.size < b.size: 

return c[::-1] 

else: 

return c 

 

 

def numpy_correlate_emulate(a, b, mode='valid'): 

''' 

Slow version of :py:func:`numpy.correlate` for comparison. 

''' 

 

a = num.asarray(a) 

b = num.asarray(b) 

kmin = -(b.size-1) 

klen = a.size-kmin 

kmin, kmax = numpy_correlate_lag_range(a, b, mode=mode) 

kmin = int(kmin) 

kmax = int(kmax) 

klen = kmax - kmin + 1 

c = num.zeros(klen, dtype=num.find_common_type((b.dtype, a.dtype), ())) 

for k in range(kmin, kmin+klen): 

imin = max(0, -k) 

ilen = min(b.size, a.size-k) - imin 

c[k-kmin] = num.sum( 

a[imin+k:imin+ilen+k] * num.conj(b[imin:imin+ilen])) 

 

return c 

 

 

def numpy_correlate_lag_range(a, b, mode='valid', use_fft=False): 

''' 

Get range of lags for which :py:func:`numpy.correlate` produces values. 

''' 

 

a = num.asarray(a) 

b = num.asarray(b) 

 

kmin = -(b.size-1) 

if mode == 'full': 

klen = a.size-kmin 

elif mode == 'same': 

klen = max(a.size, b.size) 

kmin += (a.size+b.size-1 - max(a.size, b.size)) // 2 + \ 

int(not use_fft and a.size % 2 == 0 and b.size > a.size) 

elif mode == 'valid': 

klen = abs(a.size - b.size) + 1 

kmin += min(a.size, b.size) - 1 

 

return kmin, kmin + klen - 1 

 

 

def autocorr(x, nshifts): 

''' 

Compute biased estimate of the first autocorrelation coefficients. 

 

:param x: input array 

:param nshifts: number of coefficients to calculate 

''' 

 

mean = num.mean(x) 

std = num.std(x) 

n = x.size 

xdm = x - mean 

r = num.zeros(nshifts) 

for k in range(nshifts): 

r[k] = 1./((n-num.abs(k))*std) * num.sum(xdm[:n-k] * xdm[k:]) 

 

return r 

 

 

def yulewalker(x, order): 

''' 

Compute autoregression coefficients using Yule-Walker method. 

 

:param x: input array 

:param order: number of coefficients to produce 

 

A biased estimate of the autocorrelation is used. The Yule-Walker equations 

are solved by :py:func:`numpy.linalg.inv` instead of Levinson-Durbin 

recursion which is normally used. 

''' 

 

gamma = autocorr(x, order+1) 

d = gamma[1:1+order] 

a = num.zeros((order, order)) 

gamma2 = num.concatenate((gamma[::-1], gamma[1:order])) 

for i in range(order): 

ioff = order-i 

a[i, :] = gamma2[ioff:ioff+order] 

 

return num.dot(num.linalg.inv(a), -d) 

 

 

def moving_avg(x, n): 

n = int(n) 

cx = x.cumsum() 

nn = len(x) 

y = num.zeros(nn, dtype=cx.dtype) 

y[n//2:n//2+(nn-n)] = (cx[n:]-cx[:-n])/n 

y[:n//2] = y[n//2] 

y[n//2+(nn-n):] = y[n//2+(nn-n)-1] 

return y 

 

 

def moving_sum(x, n, mode='valid'): 

n = int(n) 

cx = x.cumsum() 

nn = len(x) 

 

if mode == 'valid': 

if nn-n+1 <= 0: 

return num.zeros(0, dtype=cx.dtype) 

y = num.zeros(nn-n+1, dtype=cx.dtype) 

y[0] = cx[n-1] 

y[1:nn-n+1] = cx[n:nn]-cx[0:nn-n] 

 

if mode == 'full': 

y = num.zeros(nn+n-1, dtype=cx.dtype) 

if n <= nn: 

y[0:n] = cx[0:n] 

y[n:nn] = cx[n:nn]-cx[0:nn-n] 

y[nn:nn+n-1] = cx[-1]-cx[nn-n:nn-1] 

else: 

y[0:nn] = cx[0:nn] 

y[nn:n] = cx[nn-1] 

y[n:nn+n-1] = cx[nn-1] - cx[0:nn-1] 

 

if mode == 'same': 

n1 = (n-1)//2 

y = num.zeros(nn, dtype=cx.dtype) 

if n <= nn: 

y[0:n-n1] = cx[n1:n] 

y[n-n1:nn-n1] = cx[n:nn]-cx[0:nn-n] 

y[nn-n1:nn] = cx[nn-1] - cx[nn-n:nn-n+n1] 

else: 

y[0:max(0, nn-n1)] = cx[min(n1, nn):nn] 

y[max(nn-n1, 0):min(n-n1, nn)] = cx[nn-1] 

y[min(n-n1, nn):nn] = cx[nn-1] - cx[0:max(0, nn-(n-n1))] 

 

return y 

 

 

def nextpow2(i): 

return 2**int(math.ceil(math.log(i)/math.log(2.))) 

 

 

def snapper_w_offset(nmax, offset, delta, snapfun=math.ceil): 

def snap(x): 

return max(0, min(int(snapfun((x-offset)/delta)), nmax)) 

return snap 

 

 

def snapper(nmax, delta, snapfun=math.ceil): 

def snap(x): 

return max(0, min(int(snapfun(x/delta)), nmax)) 

return snap 

 

 

def apply_costaper(a, b, c, d, y, x0, dx): 

hi = snapper_w_offset(y.size, x0, dx) 

y[:hi(a)] = 0. 

y[hi(a):hi(b)] *= 0.5 \ 

- 0.5*num.cos((dx*num.arange(hi(a), hi(b))-(a-x0))/(b-a)*num.pi) 

y[hi(c):hi(d)] *= 0.5 \ 

+ 0.5*num.cos((dx*num.arange(hi(c), hi(d))-(c-x0))/(d-c)*num.pi) 

y[hi(d):] = 0. 

 

 

def span_costaper(a, b, c, d, y, x0, dx): 

hi = snapper_w_offset(y.size, x0, dx) 

return hi(a), hi(d) - hi(a) 

 

 

def costaper(a, b, c, d, nfreqs, deltaf): 

hi = snapper(nfreqs, deltaf) 

tap = num.zeros(nfreqs) 

tap[hi(a):hi(b)] = 0.5 \ 

- 0.5*num.cos((deltaf*num.arange(hi(a), hi(b))-a)/(b-a)*num.pi) 

tap[hi(b):hi(c)] = 1. 

tap[hi(c):hi(d)] = 0.5 \ 

+ 0.5*num.cos((deltaf*num.arange(hi(c), hi(d))-c)/(d-c)*num.pi) 

 

return tap 

 

 

def t2ind(t, tdelta, snap=round): 

return int(snap(t/tdelta)) 

 

 

def hilbert(x, N=None): 

''' 

Return the hilbert transform of x of length N. 

 

(from scipy.signal, but changed to use fft and ifft from numpy.fft) 

''' 

 

x = num.asarray(x) 

if N is None: 

N = len(x) 

if N <= 0: 

raise ValueError("N must be positive.") 

if num.iscomplexobj(x): 

logger.warning('imaginary part of x ignored.') 

x = num.real(x) 

Xf = num.fft.fft(x, N, axis=0) 

h = num.zeros(N) 

if N % 2 == 0: 

h[0] = h[N//2] = 1 

h[1:N//2] = 2 

else: 

h[0] = 1 

h[1:(N+1)//2] = 2 

 

if len(x.shape) > 1: 

h = h[:, num.newaxis] 

x = num.fft.ifft(Xf*h) 

return x 

 

 

def near(a, b, eps): 

return abs(a-b) < eps 

 

 

def coroutine(func): 

def wrapper(*args, **kwargs): 

gen = func(*args, **kwargs) 

next(gen) 

return gen 

 

wrapper.__name__ = func.__name__ 

wrapper.__dict__ = func.__dict__ 

wrapper.__doc__ = func.__doc__ 

return wrapper 

 

 

class States(object): 

''' 

Utility to store channel-specific state in coroutines. 

''' 

 

def __init__(self): 

self._states = {} 

 

def get(self, tr): 

k = tr.nslc_id 

if k in self._states: 

tmin, deltat, dtype, value = self._states[k] 

if (near(tmin, tr.tmin, deltat/100.) 

and near(deltat, tr.deltat, deltat/10000.) 

and dtype == tr.ydata.dtype): 

 

return value 

 

return None 

 

def set(self, tr, value): 

k = tr.nslc_id 

if k in self._states and self._states[k][-1] is not value: 

self.free(self._states[k][-1]) 

 

self._states[k] = (tr.tmax+tr.deltat, tr.deltat, tr.ydata.dtype, value) 

 

def free(self, value): 

pass 

 

 

@coroutine 

def co_list_append(list): 

while True: 

list.append((yield)) 

 

 

class ScipyBug(Exception): 

pass 

 

 

@coroutine 

def co_lfilter(target, b, a): 

''' 

Successively filter broken continuous trace data (coroutine). 

 

Create coroutine which takes :py:class:`Trace` objects, filters their data 

through :py:func:`scipy.signal.lfilter` and sends new :py:class:`Trace` 

objects containing the filtered data to target. This is useful, if one 

wants to filter a long continuous time series, which is split into many 

successive traces without producing filter artifacts at trace boundaries. 

 

Filter states are kept *per channel*, specifically, for each (network, 

station, location, channel) combination occuring in the input traces, a 

separate state is created and maintained. This makes it possible to filter 

multichannel or multistation data with only one :py:func:`co_lfilter` 

instance. 

 

Filter state is reset, when gaps occur. 

 

Use it like this:: 

 

from pyrocko.trace import co_lfilter, co_list_append 

 

filtered_traces = [] 

pipe = co_lfilter(co_list_append(filtered_traces), a, b) 

for trace in traces: 

pipe.send(trace) 

 

pipe.close() 

 

''' 

 

try: 

states = States() 

output = None 

while True: 

input = (yield) 

 

zi = states.get(input) 

if zi is None: 

zi = num.zeros(max(len(a), len(b))-1, dtype=num.float) 

 

output = input.copy(data=False) 

try: 

ydata, zf = signal.lfilter(b, a, input.get_ydata(), zi=zi) 

except ValueError: 

raise ScipyBug( 

'signal.lfilter failed: could be related to a bug ' 

'in some older scipy versions, e.g. on opensuse42.1') 

 

output.set_ydata(ydata) 

states.set(input, zf) 

target.send(output) 

 

except GeneratorExit: 

target.close() 

 

 

def co_antialias(target, q, n=None, ftype='fir'): 

b, a, n = util.decimate_coeffs(q, n, ftype) 

anti = co_lfilter(target, b, a) 

return anti 

 

 

@coroutine 

def co_dropsamples(target, q, nfir): 

try: 

states = States() 

while True: 

tr = (yield) 

newdeltat = q * tr.deltat 

ioffset = states.get(tr) 

if ioffset is None: 

# for fir filter, the first nfir samples are pulluted by 

# boundary effects; cut it off. 

# for iir this may be (much) more, we do not correct for that. 

# put sample instances to a time which is a multiple of the 

# new sampling interval. 

newtmin_want = math.ceil( 

(tr.tmin+(nfir+1)*tr.deltat) / newdeltat) * newdeltat \ 

- (nfir/2*tr.deltat) 

ioffset = int(round((newtmin_want - tr.tmin)/tr.deltat)) 

if ioffset < 0: 

ioffset = ioffset % q 

 

newtmin_have = tr.tmin + ioffset * tr.deltat 

newtr = tr.copy(data=False) 

newtr.deltat = newdeltat 

# because the fir kernel shifts data by nfir/2 samples: 

newtr.tmin = newtmin_have - (nfir/2*tr.deltat) 

newtr.set_ydata(tr.get_ydata()[ioffset::q].copy()) 

states.set(tr, (ioffset % q - tr.data_len() % q) % q) 

target.send(newtr) 

 

except GeneratorExit: 

target.close() 

 

 

def co_downsample(target, q, n=None, ftype='fir'): 

''' 

Successively downsample broken continuous trace data (coroutine). 

 

Create coroutine which takes :py:class:`Trace` objects, downsamples their 

data and sends new :py:class:`Trace` objects containing the downsampled 

data to target. This is useful, if one wants to downsample a long 

continuous time series, which is split into many successive traces without 

producing filter artifacts and gaps at trace boundaries. 

 

Filter states are kept *per channel*, specifically, for each (network, 

station, location, channel) combination occuring in the input traces, a 

separate state is created and maintained. This makes it possible to filter 

multichannel or multistation data with only one :py:func:`co_lfilter` 

instance. 

 

Filter state is reset, when gaps occur. The sampling instances are choosen 

so that they occur at (or as close as possible) to even multiples of the 

sampling interval of the downsampled trace (based on system time). 

''' 

 

b, a, n = util.decimate_coeffs(q, n, ftype) 

return co_antialias(co_dropsamples(target, q, n), q, n, ftype) 

 

 

@coroutine 

def co_downsample_to(target, deltat): 

 

decimators = {} 

try: 

while True: 

tr = (yield) 

ratio = deltat / tr.deltat 

rratio = round(ratio) 

if abs(rratio - ratio)/ratio > 0.0001: 

raise util.UnavailableDecimation('ratio = %g' % ratio) 

 

deci_seq = tuple(x for x in util.decitab(int(rratio)) if x != 1) 

if deci_seq not in decimators: 

pipe = target 

for q in deci_seq[::-1]: 

pipe = co_downsample(pipe, q) 

 

decimators[deci_seq] = pipe 

 

decimators[deci_seq].send(tr) 

 

except GeneratorExit: 

for g in decimators.values(): 

g.close() 

 

 

class DomainChoice(StringChoice): 

choices = [ 

'time_domain', 

'frequency_domain', 

'envelope', 

'absolute', 

'cc_max_norm'] 

 

 

class MisfitSetup(Object): 

''' 

Contains misfit setup to be used in :py:func:`trace.misfit` 

 

:param description: Description of the setup 

:param norm: L-norm classifier 

:param taper: Object of :py:class:`Taper` 

:param filter: Object of :py:class:`FrequencyResponse` 

:param domain: ['time_domain', 'frequency_domain', 'envelope', 'absolute', 

'cc_max_norm'] 

 

Can be dumped to a yaml file. 

''' 

 

xmltagname = 'misfitsetup' 

description = String.T(optional=True) 

norm = Int.T(optional=False) 

taper = Taper.T(optional=False) 

filter = FrequencyResponse.T(optional=True) 

domain = DomainChoice.T(default='time_domain') 

 

 

def equalize_sampling_rates(trace_1, trace_2): 

''' 

Equalize sampling rates of two traces (reduce higher sampling rate to 

lower). 

 

:param trace_1: :py:class:`Trace` object 

:param trace_2: :py:class:`Trace` object 

 

Returns a copy of the resampled trace if resampling is needed. 

''' 

 

if same_sampling_rate(trace_1, trace_2): 

return trace_1, trace_2 

 

if trace_1.deltat < trace_2.deltat: 

t1_out = trace_1.copy() 

t1_out.downsample_to(deltat=trace_2.deltat, snap=True) 

logger.debug('Trace downsampled (return copy of trace): %s' 

% '.'.join(t1_out.nslc_id)) 

return t1_out, trace_2 

 

elif trace_1.deltat > trace_2.deltat: 

t2_out = trace_2.copy() 

t2_out.downsample_to(deltat=trace_1.deltat, snap=True) 

logger.debug('Trace downsampled (return copy of trace): %s' 

% '.'.join(t2_out.nslc_id)) 

return trace_1, t2_out 

 

 

def Lx_norm(u, v, norm=2): 

''' 

Calculate the misfit denominator *m* and the normalization devisor *n* 

according to norm. 

 

The normalization divisor *n* is calculated from ``v``. 

 

:param u: :py:class:`numpy.array` 

:param v: :py:class:`numpy.array` 

:param norm: (default = 2) 

 

``u`` and ``v`` must be of same size. 

''' 

 

if norm == 1: 

return ( 

num.sum(num.abs(v-u)), 

num.sum(num.abs(v))) 

 

elif norm == 2: 

return ( 

num.sqrt(num.sum((v-u)**2)), 

num.sqrt(num.sum(v**2))) 

 

else: 

return ( 

num.power(num.sum(num.abs(num.power(v - u, norm))), 1./norm), 

num.power(num.sum(num.abs(num.power(v, norm))), 1./norm)) 

 

 

def do_downsample(tr, deltat): 

if abs(tr.deltat - deltat) / tr.deltat > 1e-6: 

tr = tr.copy() 

tr.downsample_to(deltat, snap=True, demean=False) 

else: 

if tr.tmin/tr.deltat > 1e-6 or tr.tmax/tr.deltat > 1e-6: 

tr = tr.copy() 

tr.snap() 

return tr 

 

 

def do_extend(tr, tmin, tmax): 

if tmin < tr.tmin or tmax > tr.tmax: 

tr = tr.copy() 

tr.extend(tmin=tmin, tmax=tmax, fillmethod='repeat') 

 

return tr 

 

 

def do_pre_taper(tr, taper): 

return tr.taper(taper, inplace=False, chop=True) 

 

 

def do_fft(tr, filter): 

if filter is None: 

return tr 

else: 

ndata = tr.ydata.size 

nfft = nextpow2(ndata) 

padded = num.zeros(nfft, dtype=num.float) 

padded[:ndata] = tr.ydata 

spectrum = num.fft.rfft(padded) 

df = 1.0 / (tr.deltat * nfft) 

frequencies = num.arange(spectrum.size)*df 

return [tr, frequencies, spectrum] 

 

 

def do_filter(inp, filter): 

if filter is None: 

return inp 

else: 

tr, frequencies, spectrum = inp 

spectrum *= filter.evaluate(frequencies) 

return [tr, frequencies, spectrum] 

 

 

def do_ifft(inp): 

if isinstance(inp, Trace): 

return inp 

else: 

tr, _, spectrum = inp 

ndata = tr.ydata.size 

tr = tr.copy(data=False) 

tr.set_ydata(num.fft.irfft(spectrum)[:ndata]) 

return tr 

 

 

def check_alignment(t1, t2): 

if abs(t1.tmin-t2.tmin) > t1.deltat * 1e-4 or \ 

abs(t1.tmax - t2.tmax) > t1.deltat * 1e-4 or \ 

t1.ydata.shape != t2.ydata.shape: 

raise MisalignedTraces( 

'Cannot calculate misfit of %s and %s due to misaligned ' 

'traces.' % ('.'.join(t1.nslc_id), '.'.join(t2.nslc_id)))