Peak Oil
Donate Bitcoin

Donate Paypal


PeakOil is You

PeakOil is You

I'm confused. How do you calculate and model this data?

Discuss research and forecasts regarding hydrocarbon depletion.
Post a reply

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 01:54:19

$this->bbcode_second_pass_quote('pstarr', '1')5x reserve growth sounds abiotic? Are these guys kidding?


I believe it is caused by some very early estimates that were way off. Once they registered an estimate somewhat close within the first several years after discovery, then the error correction creep appears between 1x and 2x over long periods.

My current theory based on the quirks of human behavior is that underlings always give conservative estimates. That coupled with the fact that abandoned wells stop counting at some point right at the eventual estimate causes a monotonic error drift. This essentially falls into the class of estimates that suffer from truncation bias. Since error estimates are not centered around 0 with equal weights positive or negative, then it turns into a continual drift upwards. From the looks of the data, no one ever wants to make a guess that eventually overestimates, so the bias is always positive.

One quick analogy that I can think of is The Price is Right strategy. You always try to lowball the guess if someone just made one and you happen to be the last guesser. If you can only win if you don't overguess, then it makes sense to just add a single cent to the other guy's estimate. The worst that can happen is that you both overbid. The best case is that you win. The odds are very slim that the other guy can win if you make the margin small enough. I think if you look at how auction prices rise, it follows a similar pattern of a monotonically increase which starts to level off after a while. But, again, these are both human behaviors which deal with psychological battles more than anything else. And business is largely psychological -- how to trick the customer into parting with his money. You just have to know who the customer is, to know how it will play out.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby dub_scratch » Mon 23 Jan 2006, 03:34:20

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')
One quick analogy that I can think of is The Price is Right strategy. You always try to lowball the guess if someone just made one and you happen to be the last guesser. If you can only win if you don't overguess, then it makes sense to just add a single cent to the other guy's estimate. The worst that can happen is that you both overbid. The best case is that you win. The odds are very slim that the other guy can win if you make the margin small enough. I think if you look at how auction prices rise, it follows a similar pattern of a monotonically increase which starts to level off after a while. But, again, these are both human behaviors which deal with psychological battles more than anything else. And business is largely psychological -- how to trick the customer into parting with his money. You just have to know who the customer is, to know how it will play out.


WebHubbleTelescope, why is this reserve growth-- for whatever reason it happens-- is it not a factor in recently discovered oil. If last year's oil discoveries are about 9bbl and those discoveries end up producing three times that amount, then the idea that discoveries are insufficient in comparison to current consumption would be false. Why are these new fields not going to grow like the ones in the past?
dub_scratch
Tar Sands
Tar Sands
 
Posts: 700
Joined: Thu 16 Dec 2004, 04:00:00
Location: Santa Monica, CA
Top

Re: I'm confused. How do you calculate and model this data?

Postby Doly » Mon 23 Jan 2006, 09:58:46

$this->bbcode_second_pass_quote('dub_scratch', 'W')hy are these new fields not going to grow like the ones in the past?


Even assuming they will, the curve of discoveries is going well downward, without a doubt.
User avatar
Doly
Expert
Expert
 
Posts: 4370
Joined: Fri 03 Dec 2004, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby shakespear1 » Mon 23 Jan 2006, 10:18:20

$this->bbcode_second_pass_code('', 'Like I said before, this is basically a bean-counting exercise and I know that they know how much is under the ground. If they were to spill the beans, we would all be better off. ')

Woww, you are completly in the dark what we "oilers" do. I will need to include in the category "oilers" geologists.

The one thing you need to bare in mind is that we do not see the reservoir underground as you see your capacitor. Here is a way to visualize this.

Go to your back yard and take your drinking straw and make a hole of that diameter. Now based on what you see from the ground that you pulled out in the sample volume of that hole try to characterize the properties of the rest of your back yard under the grass. Now go "drill" a few more "wells" the same way and IMPROVE your estimate of what you think is going on spatially underground.

THIS IS WHAT we are doing in trying to evaluate the reserves in a field. Our bore hole diameters underground will be on the order of magnitude of 7 to 8 1/2 inches !!!

Hope that helps.

You might do a seach on my name of my oldest posts where I tried to explain in greater detail how we "petroleum" engineers do our work. :-D
Men argue, nature acts !
Voltaire

"...In the absence of the gold standard, there is no way to protect savings from confiscation through inflation."

Alan Greenspan
shakespear1
Heavy Crude
Heavy Crude
 
Posts: 1532
Joined: Fri 13 May 2005, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby 128shot » Mon 23 Jan 2006, 13:36:55

$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('128shot', '
')Tensions run high, particulary with doomsayers :)



Still, I'm confused on how to get some kind of data representation I can present in 3 weeks...this has to be as scientifical as possible..


For a minute there, I thought I recalled a Bushism. But then I looked it up and yes indeed scientifical is a word. It means scientific. Which I am for, BTW.



LOL.


Even though I'm not sure you ment that to be funny, I thought it was.


Never the less, my problem is still the same. Though I'm finding more on this forum than I thought I would.


I still can't find data I can represent because there is no calculation put up for it. Like a formula or something. garr..
User avatar
128shot
Lignite
Lignite
 
Posts: 303
Joined: Wed 18 Jan 2006, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 22:15:57

Here is the data from Attanasi & Root, converted from OCR software.


[sub]Discovery -- Year of Estimate
Year 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Unknown 1719 1735 1743 1826 1876 1829 1882 1857 1825 1830 1838 1868 1853 1870 1883
Pre-1900 5005 4886 5009 5097 5130 5196 5294 5424 5664 5747 5780 5835 5839 5856 5846
1900 658 669 674 680 684 705 706 721 741 725 732 740 738 747 745
1901 2640 2563 2666 2782 2823 2858 2915 3108 3333 3374 3423 3658 3829 3877 3871
1902 945 948 940 968 948 980 995 1039 1040 1056 1125 1136 1221 1188 1200
1903 110 110 111 112 113 115 116 117 118 118 120 120 122 123 120
1904 221 224 229 230 226 225 225 226 225 226 227 228 230 231 233
1905 1914 1867 1896 1874 1867 1887 1920 1911 1929 1957 1957 1969 2016 2023 2039
1906 801 811 823 839 845 841 845 865 883 908 916 935 940 962 953
1907 16 16 17 18 19 21 21 21 21 21 21 21 21 22 22
1908 56 57 59 59 59 59 60 61 61 61 62 63 63 64 66
1909 1546 1515 1529 1535 1531 1514 1511 1567 1576 1566 1591 1567 1558 1555 1559
1910 1742 1727 1749 1765 1801 1802 1813 1826 1862 1860 1882 1900 1929 1954 1981
1911 1271 1424 2112 2145 2173 2226 2291 2371 2047 2148 2255 2387 2435 2453 2501
1912 1170 1186 1198 1232 1261 1274 1283 1276 1322 1356 1365 1374 1396 1396 1399
1913 708 713 722 724 725 736 738 743 752 753 752 756 765 762 766
1914 403 409 418 434 446 446 467 468 473 465 470 471 474 474 471
1915 1314 1306 1336 1347 1368 1351 1370 1376 1426 1414 1453 1454 1481 1478 1474
1916 1968 1969 2009 2018 2032 2062 2098 2142 2178 2192 2231 2242 2302 2336 2347
1917 1263 1248 1234 1252 1266 1277 1284 1289 1307 1299 1302 1314 1319 1320 1326
1918 849 849 858 871 876 869 877 891 909 895 899 904 906 914 911
1919 2593 2435 2524 2583 2586 2603 2638 2658 2761 2781 2774 2782 2798 2809 2821
1920 2312 2322 2299 2313 2340 2352 2374 2400 2435 2541 2511 2571 2538 2546 2523
1921 1997 1974 1950 1945 1950 1952 1956 1964 1983 1971 1978 1980 1994 1990 1990
1922 1597 1604 1598 1616 1620 1632 1641 1656 1664 1652 1672 1681 1730 1742 1733
1923 1389 1376 1380 1400 1409 1424 1443 1456 1511 1506 1539 1520 1527 1539 1529
1924 1201 1210 1214 1228 1243 1250 1265 1279 1300 1294 1308 1321 1347 1493 1507
1925 1302 1290 1313 1328 1337 1351 1367 1381 1404 1405 1444 1441 1475 1490 1496
1926 3997 4046 4086 4097 4143 4076 4123 4145 4202 4226 4257 4282 4305 4321 4328
1927 1809 1801 1802 1813 1825 1831 1852 1894 1887 1893 1910 1903 1912 1918 1927
1928 3005 3008 3024 3061 3081 3121 3147 3164 3191 3190 3208 3206 3225 3231 3230
1929 3957 4009 3893 3893 3920 3937 4004 4092 4124 4130 4148 4209 4195 4215 4234
1930 8262 8229 8140 8210 8222 8180 8245 8273 8299 8295 8328 8348 8380 8398 8356
1931 2104 2088 2063 2062 2070 2071 2079 2090 2095 2105 2107 2121 2134 2145 2149
1932 2809 2816 3173 3197 3211 3249 3291 3376 3214 3195 3238 3255 3241 3230 3247
1933 1142 1153 1166 1158 1159 1160 1180 1189 1199 1212 1223 1233 1239 1244 1242
1934 2419 2357 2362 2364 2379 2385 2411 2458 2461 2464 2487 2476 2488 2494 2507
1935 2845 2832 2769 2774 2787 2798 2818 2833 2842 2845 2854 2877 2901 2906 2953
1936 2611 2612 2654 2687 2832 2837 2869 2884 2919 2933 2969 3010 2997 2998 3035
1937 5563 5692 5687 5993 6110 6249 6259 6414 6709 6636 6701 6853 6933 6994 7057
1938 4974 4923 4917 4953 4977 5029 5048 5074 5131 5138 5166 5159 5174 5197 5221
1939 1685 1684 1664 1661 1679 1662 1666 1682 1691 1699 1730 1745 1751 1772 1782
1940 4095 4074 3842 3869 3892 3910 3944 3990 4090 4131 4185 4215 4239 4263 4268
1941 1425 2099 1460 1454 1465 1468 1481 1505 1518 1520 1521 1530 1541 1548 1551
1942 1946 1951 2000 2033 2045 2057 2076 2105 2168 2218 2248 2268 2279 2372 2374
1943 1073 1095 1100 1113 1125 1137 1148 1166 1176 1193 1205 1218 1232 1239 1248
1944 2646 2670 2680 2736 2744 2770 2794 2846 2907 2898 2926 2953 2981 2991 3003

Table 1. Continued.
Discovery -- Year of Estimate
Year 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
1945 2752 2762 2849 2881 2929 2957 2997 3031 3135 3172 3245 3287 3316 3352 3338
1946 813 813 811 816 823 845 843 848 858 878 881 876 887 892 889
1947 1760 1738 1740 1834 1848 1841 1889 1902 1947 1947 1981 1956 1984 2010 2023
1948 3363 3394 3389 3417 3427 3499 3546 3572 3636 3617 3642 3680 3688 3694 3604
1949 3566 3584 3587 3625 3647 3704 3750 3817 3878 3940 3994 4017 4040 4106 4130
1950 3087 3169 3215 3235 3297 3327 3399 3475 3563 3589 3646 3724 3767 3851 3911
1951 1485 1488 1506 1522 1554 1576 1612 1630 1656 1679 1692 1714 1728 1747 1751
1952 1359 1352 1362 1378 1396 1410 1410 1432 1458 1459 1478 1484 1500 1510 1518
1953 2188 2212 2236 2282 2333 2363 2410 2453 2477 2489 2529 2548 2582 2609 2625
1954 1526 1532 1546 1542 1568 1574 1591 1587 1607 1612 1623 1635 1657 1666 1673
1955 1629 1667 1690 1747 1792 1820 1856 1908 1939 1977 2000 2038 2070 2106 2132
1956 2272 2297 2358 2398 2427 2446 2508 2562 2621 2641 2699 2735 2810 2850 2886
1957 1232 1254 1273 1290 1316 1368 1385 1429 1461 1476 1495 1512 1540 1550 1553
1958 1651 1664 1598 1614 1629 1617 1656 1687 1706 1710 1729 1759 1786 1797 1788
1959 728 729 739 759 762 788 796 807 830 836 848 858 857 868 878
1960 972 1006 1012 1129 1182 1248 1311 1343 1474 1409 1434 1462 1490 1509 1580
1961 633 645 634 642 659 686 701 718 735 733 747 751 763 769 782
1962 868 877 888 915 946 966 1003 1008 1026 1039 1064 1096 1098 1084 1089
1963 503 507 521 543 574 596 614 631 646 659 664 685 700 730 746
1964 690 668 677 682 708 727 740 756 790 794 807 807 823 848 880
1965 573 589 610 643 664 667 682 718 730 739 745 768 792 804 820
1966 634 624 631 650 679 665 679 713 736 742 743 748 768 788 777
1967 594 614 636 653 656 688 722 733 750 748 770 789 810 849 878
1968 640 664 660 671 677 683 724 744 761 767 774 783 794 811 822
1969 607 - 625 642 709 855 918 953 949 1016 1091 1112 1128 1136 1146 1166
1970 798 820 833 899 998 1069 1094 1128 1151 1228 1237 1256 1261 1273 1293
1971 561 583 567 589 609 647 699 702 720 735 754 770 777 803 835
1972 438 457 458 487 506 526 541 560 589 596 607 658 670 710 676
1973 392 424 460 508 623 615 639 698 743 768 787 807 835 851 852
1974 423 474 494 539 612 621 653 680 700 705 741 840 778 805 809
1975 311 339 451 504 506 519 576 616 672 686 708 732 786 831 836
1976 344 440 503 698 710 711 747 807 798 829 852 856 857 880 906
1977 110 171 21 264 376 385 393 428 452 458 465 489 557 588 584
1978 0 45 87 113 140 146 172 190 216 217 222 227 235 241 256
1979 0 0 77 217 394 432 479 475 470 510 517 559 589 608 603
1980 0 0 0 59 151 178 211 244 246 252 267 295 314 332 350
1981 0 0 0 0 91 282 353 390 467 474 502 557 587 667 706
1982 0 0 0 0 0 120 209 241 264 261 269 292 307 336 335
1983 0 0 0 0 0 0 97 219 406 307 332 329 345 374 397
1984 0 0 0 0 0 0 0 103 179 212 224 248 272 282 301
1985 0 0 0 0 0 0 0 0 111 192 224 258 264 284 303
1986 0 0 0 0 0 0 0 0 0 35 71 81 89 99 98
1987 0 0 0 0 0 0 0, 0 0 0 45 90 214 232 235
1988 0 0 0 0 0 0 0 0 0 0 0 44 90 100 107
1989 0 0 0 0 0 0 0 0 0 0 0 0 38 74 95
1990 0 0 0 0 0 0 0 0 0 0 0 0 0 37 79
1991 0 0 0 0 0 0 0 0 0 0 0 0 0 0 97

[/sub]

In the next post I will provide a correct stationary analysis of the data.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 22:54:50

If I plot the above data as a scatter plot which shows only fractional increases per year since the year of discovery it looks like the chart below. Note that this plot does not visualize the "multiplier" approach which A&R choose to do; IMO they do this rather unwisely, because it accentuates bad early estimates. The "fractional" approach is an accepted statistical way to look at this kind of data which avoids amplifying regions with poor statistics.

Image

I also put in a 20-point moving average filter to guide the eye (not weighted by size of field). Notice that in the sweet spot right in the middle of the chart, where we get the best statistics, the "reserve growth" fluctuates around 1% per year. There is some growth for fields for fields older than ~80 years, but those have have worse sampling statistics than the rest. The latest data also exhibits poor sampling statistics.

Basically this is how you extrapolate what is assumed stationary data correctly.

Now, who is going to make the claim that 1% reserve growth will effectively compensate the 5% to 20% depletion rate per year routinely estimated for many fields?
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 23:13:42

$this->bbcode_second_pass_quote('shakespear1', '
')The one thing you need to bare in mind is that we do not see the reservoir underground as you see your capacitor. Here is a way to visualize this.

Go to your back yard and take your drinking straw and make a hole of that diameter. Now based on what you see from the ground that you pulled out in the sample volume of that hole try to characterize the properties of the rest of your back yard under the grass. Now go "drill" a few more "wells" the same way and IMPROVE your estimate of what you think is going on spatially underground.

THIS IS WHAT we are doing in trying to evaluate the reserves in a field. Our bore hole diameters underground will be on the order of magnitude of 7 to 8 1/2 inches !!!

Hope that helps.

You might do a seach on my name of my oldest posts where I tried to explain in greater detail how we "petroleum" engineers do our work. :-D


You put in a proportional number of straws to how big the field is. That is my only assumption in extraction, a kind of greedy estimate. I would think that discovery estimates use a bunch of different techniques that improve as knowledge advances over time.

I never claim that it works the same way as an RC electrical circuit, just that the math happens to come out the same way (a spring/damper system uses the same math too!). An electrical circuit is stochastic in the sense of populations of electrons working in some average way toward a lower potential field. A large set of oil fields is stochastic in the sense that the aggregate works together as a set of many independent decisions, modeled as Markov rate probabilities. If my approach helps people to visualize stuff, nothing wrong with that.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Mon 23 Jan 2006, 23:23:19

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Now, who is going to make the claim that 1% reserve growth will effectively compensate the 5% to 20% depletion rate per year routinely estimated for many fields?


Nobody I know of claims that the data as presented by Attanasai and Root represents a 1% growth in reserves/ultimate. Not MMS, USGS, EIA, DOE or any one else who has published a paper on the subject using any of the proprietary reserve databases with which I am familiar thinks anything remotely resembling what you have just stated.

Could you explain your reasoning in a little more depth perhaps?
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Mon 23 Jan 2006, 23:33:03

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')
You put in a proportional number of straws to how big the field is. That is my only assumption in extraction, a kind of greedy estimate.


Saudi Arabia has perhaps what, 1500? 2000 production wells for a rate of 10 MBO/Day? The United States has perhaps hundreds of thousands of wells for a current rate of 5 MBO/Day?

Seems like one of several things should happen...the US shouldn't have any reserve growth because everyone knows everything, and this isn't true and can be easily demonstrated as recently as data differences between 95 and 2000.

Or perhaps the Saudi's have no idea what they are sitting on, and could be subject to orders of magnitude more reserve growth because they haven't explored their areas under your "drill everywhere to know more" theory.

Unfortunately, capital investment of the magnitude your method suggests doesn't happen.
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 23:44:51

$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Now, who is going to make the claim that 1% reserve growth will effectively compensate the 5% to 20% depletion rate per year routinely estimated for many fields?


Nobody I know of claims that the data as presented by Attanasai and Root represents a 1% growth in reserves/ultimate. Not MMS, USGS, EIA, DOE or any one else who has published a paper on the subject using any of the proprietary reserve databases with which I am familiar thinks anything remotely resembling what you have just stated.

Could you explain your reasoning in a little more depth perhaps?


I don't know, maybe by looking at the data? It's sitting right there. I would only offer that maybe we need somebody like Khebab or pup or some other statistical wizards on this board for a second opinion.

So take a look at the row corresponding to the discoveries of 1930:

$this->bbcode_second_pass_code('', '
Year 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
1930 8262 8229 8140 8210 8222 8180 8245 8273 8299 8295 8328 8348 8380 8398 8356
')
Notice that in this case the average yearly gowth is less than 1% from the previous year's estimate (actually way less). This is for "aged" discoveries ranging from 47 to 61 years. I essentially repeated this for the aggregate set of data. If the data is bad, hey, don't blame me, this stuff is from the government, according to A&R they got it as unpublished data from the EIA. And they saw fit to fill up 4 journal pages with the stuff, so it must have been used in their analysis, wouldn't you think?
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Mon 23 Jan 2006, 23:58:04

$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')
You put in a proportional number of straws to how big the field is. That is my only assumption in extraction, a kind of greedy estimate.


Saudi Arabia has perhaps what, 1500? 2000 production wells for a rate of 10 MBO/Day? The United States has perhaps hundreds of thousands of wells for a current rate of 5 MBO/Day?

Seems like one of several things should happen...the US shouldn't have any reserve growth because everyone knows everything, and this isn't true and can be easily demonstrated as recently as data differences between 95 and 2000.

Or perhaps the Saudi's have no idea what they are sitting on, and could be subject to orders of magnitude more reserve growth because they haven't explored their areas under your "drill everywhere to know more" theory.

Unfortunately, capital investment of the magnitude your method suggests doesn't happen.


Yes, the Saudis and USA contribute to an exponential distribution of proportional extraction rates. Saudis filling up the high output level with minimal fields and the USA doing it with lots of stripper wells. Such is the nature of stochastic processes and a maximum entropy estimator.

I realize that it might be hard for everyone to grok how this stuff works out. However, take some classes in statistical mechanics and you will realize this is kids stuff in comparison.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Tue 24 Jan 2006, 00:28:07

$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Now, who is going to make the claim that 1% reserve growth will effectively compensate the 5% to 20% depletion rate per year routinely estimated for many fields?


Nobody I know of claims that the data as presented by Attanasai and Root represents a 1% growth in reserves/ultimate. Not MMS, USGS, EIA, DOE or any one else who has published a paper on the subject using any of the proprietary reserve databases with which I am familiar thinks anything remotely resembling what you have just stated.

Could you explain your reasoning in a little more depth perhaps?


I don't know, maybe by looking at the data? It's sitting right there.
So take a look at the row corresponding to the discoveries of 1930:

Notice that in this case the average yearly gowth is less than 1% from the previous year's estimate (actually way less).


Okay, I see what you are doing. The data has to be parsed into separate lines on my screen, its a bit hard to read. But lets do better than selectively pull a subjective "Hubble" and do more than look at 1930 shall we?

1903. 9%
1904 5.4%
1906 19%

Maybe this is all too old of fields for your taste? Particularly since this is growth after the field is nearly what, 70 years old?

How about

1947? 15%
1950? 27%
1955? 30%

Maybe reserve growth was "cured" by modern technology?


1975? 160%
1977? 430%
1985? 173%

Well....now this brings up the question, who is always complaining about selective data displays?

I suppose it was unreasonable of you to actually recognize something which goes against what appears to be your natural subjectivity AGAINST reserve growth. Poor scientific method display I might venture.

Hundreds of percent of "bad guesses" early in the fields life it appears to me, and put together using the Root and Attanasai methodology, without much more than using a modification of the original Arrington method, it shows quite a bit of reserve growth. Quite a bit meaning NOT 1%, and mostly certainly in the 5X to 20X range I would venture.

Want to argue quantity of reserve growth now rather than trying to pretend that bad guesses should be fixed as petroleum engineers become less "conservative", by doubling the size of their initial estimates because someone told you thats some silly engineering "must" once?
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Tue 24 Jan 2006, 00:37:06

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Yes, the Saudis and USA contribute to an exponential distribution of proportional extraction rates. Saudis filling up the high output level with minimal fields and the USA doing it with lots of stripper wells. Such is the nature of stochastic processes and a maximum entropy estimator.

I realize that it might be hard for everyone to grok how this stuff works out. However, take some classes in statistical mechanics and you will realize this is kids stuff in comparison.


Holy crap, what the hell is "exponential distribution of proportional extraction rates" mean? Particularly when used as an answer to a simple question, which is why your "drill up eveything" theory doesn't jive with some basic production facts?

Maybe this is a way to conceal how far off base your original assumption is ..with no regard for capital expenditures of course, an idea I think you would find would be greated with much snickering and giggling if you ever expressed it to the people who write the checks for exploration and developement projects.

That makes about as much sense as whatever nitwit told you that using a 2X safety/conservative factor should cover any boo-boo's in reserves....when as best I can tell from the data YOU pulled from the Enigma paper reserve growth is more a 5X-15X function.
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Tue 24 Jan 2006, 00:49:12

$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Now, who is going to make the claim that 1% reserve growth will effectively compensate the 5% to 20% depletion rate per year routinely estimated for many fields?


Nobody I know of claims that the data as presented by Attanasai and Root represents a 1% growth in reserves/ultimate. Not MMS, USGS, EIA, DOE or any one else who has published a paper on the subject using any of the proprietary reserve databases with which I am familiar thinks anything remotely resembling what you have just stated.

Could you explain your reasoning in a little more depth perhaps?


I don't know, maybe by looking at the data? It's sitting right there.
So take a look at the row corresponding to the discoveries of 1930:

Notice that in this case the average yearly gowth is less than 1% from the previous year's estimate (actually way less).


Okay, I see what you are doing. The data has to be parsed into separate lines on my screen, its a bit hard to read. But lets do better than selectively pull a subjective "Hubble" and do more than look at 1930 shall we?

1903. 9%
1904 5.4%
1906 19%

Maybe this is all too old of fields for your taste? Particularly since this is growth after the field is nearly what, 70 years old?

How about

1947? 15%
1950? 27%
1955? 30%

Maybe reserve growth was "cured" by modern technology?


1975? 160%
1977? 430%
1985? 173%

Well....now this brings up the question, who is always complaining about selective data displays?

I suppose it was unreasonable of you to actually recognize something which goes against what appears to be your natural subjectivity AGAINST reserve growth. Poor scientific method display I might venture.

Hundreds of percent of "bad guesses" early in the fields life it appears to me, and put together using the Root and Attanasai methodology, without much more than using a modification of the original Arrington method, it shows quite a bit of reserve growth. Quite a bit meaning NOT 1%, and mostly certainly in the 5X to 20X range I would venture.

Want to argue quantity of reserve growth now rather than trying to pretend that bad guesses should be fixed as petroleum engineers become less "conservative", by doubling the size of their initial estimates because someone told you thats some silly engineering "must" once?


This is what the data says -- yes, there are fairly large gains in the set, and these gains drop back after about 5 years to around 5% on the average. From there the growth drops to 1%. If I was looking at this as a a single investment, I would consider the 430% you quote as nice and the 1% not so good. But we are not looking at these as single investments; they are more like mutual funds where the collective gains matter. And the collective gains average around 1% per year, like it or not.

Go ahead and invest in your Enron, while I stay with my staid mutual fund.

(The only reason I picked the 1930 was because it was one of the biggest discovery years -- not because I cherry picked the data as you claim. You wanted to understand and evidently you got it.)
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Tue 24 Jan 2006, 17:32:52

My response is this....

Read the paper...the claim by the authors and their method, which you haven't said a single thing about, using all of the data you quoted is what? 10X 15X?

Seems like compounding reserve growth through time has a cumulative effect over, say, 100 years, similar to compounding interest.

It doesn't look like much at the individual increment level, but give it time and its katie bar the door.

This lends itself to the idea that while guessing new field sizes from their first estimate is amusing, it has nothing to do with the fields ultimate and final size, which is apparently 10X and 15X higher.

Do your models accomodate this growth of ultimate through time, or do you use a static initial number which doesn't increase with time or field depletion?

Oh...other references for the "googling challenged"...

"A New Reserve Growth Model for United States Oil and Gas Fields". M Verma, Natural Resources Research Vol 14, No 2, June-2005

And if its MASSIVE amounts of interesting data which you prefer, might I recommend AAPG Memoir 86, "Global Resource Estimates from Total Petroleum Systems", published in 2005 by AAPG.

SPE Paper # 62616 is okay as a prelude to M. Verma's above listed work as well.

And to top it off, a nice summary could be found in Natural Resources Research, Vol 14, No 3, Sept 2005 by T. Klett. "USGS's Reserve Growth Models and Their Implementation"
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Fri 27 Jan 2006, 23:19:51

$this->bbcode_second_pass_quote('ReserveGrowthRulz', 'M')y response is this....

Read the paper...the claim by the authors and their method, which you haven't said a single thing about, using all of the data you quoted is what? 10X 15X?

Seems like compounding reserve growth through time has a cumulative effect over, say, 100 years, similar to compounding interest.


Not compound growth by any stretch of the imagination. Compound growth in the traditional sense has a fixed proportional rate. This would give an accelerating slope. However, reserve growth has a decreasing rate which leads to a decelerating slope. Think of it this way -- if the growth rate follows 1/x, then any increase in x gets counterbalanced by a smaller proportional amount or ~(1+1/x)[sup]x[/sup]. I plotted a 0.5/x curve (in green) on top of the moving-average fit below.

Image

Having been away in Silicon Valley on business the past few days, I haven't had a chance to respond more quickly, but oddly enough, there is a strong silicon analogy to what I see in the way this works. Take for example, the work of Andy Grove, one of the co-founders of Intel, who did his thesis work in diffusion-limited oxide growth. In a nutshell, silicon dioxide needs a source of silicon to form, but as the SiO2 layer gets thicker, it becomes harder and takes longer for the Si atoms to diffuse to the surface and react with oxygen. This leads to a law of the following form, where F(t) is thickness as a function of time:
$this->bbcode_second_pass_code('', 'dF/dt = k/F(t)
F = sqrt (2kt)')
Note that the fractional rate can be expressed as:
$this->bbcode_second_pass_code('', 'dF/dt / F = 0.5/t
')
Note that this follows the "reserve growth" curve fit fairly well, where the fractional growth is inversely proportional to time. This is called the parabolic growth law. For the Google-challenged, you can look it up in any standard microelectronics textbooks.

Now if we were stupid semiconductor neophytes living in the 1950's and thought that the oxide growth could only be "guessed" at, then we would never have been able to advance through the microelectronics revolution and process unpredictability would have killed us. We would still be working with crystal radio sets. None of the multi-million gate circuits would have ever gotten made!

But the fact was that material scientists and engineers like Andy Grove were able to characterize the phenomena within a few years time (mid to late 1960's) and get their process down to a gnat's eyelash speaks volumes about the difference between real engineers and the geologists who believe in magical, enigmatic reserve growth. (I don't know any fab engineer in a bunny suit who believes in "enigmatic" oxide growth)

I have a suggestion for the geologists and petroleum engineers. Figure out what the heck is going on in your measurements and estimates, and then perfect the formula to eliminate the magical guess work. The more I look at it, the more I seriously think that no one has figured out how to do estimates of volume correctly. Could they all be measuring the volume as an approximation to how much they have extracted, with the increase over time caused by diminishing returns? Much like a thick SiO2 layer prevents fast oxidation, that drilling "deeper" into a field starts to slows down further depletion and you need to work harder and wait longer times to get at it? It almost sounds as if no one wants to admit that a parabolic growth law has any kind of importance.

If done correctly, reserve growth would transform from magic to a measure of extractability over time. And this is all I have been saying on how crudely the estimates have been made in the past.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Sat 28 Jan 2006, 02:06:26

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')But the fact was that material scientists and engineers like Andy Grove were able to characterize the phenomena within a few years time (mid to late 1960's) and get their process down to a gnat's eyelash speaks volumes about the difference between real engineers and the geologists who believe in magical, enigmatic reserve growth. (I don't know any fab engineer in a bunny suit who believes in "enigmatic" oxide growth)


While I can't comment on common engineering practices in other industries, I would venture the following.....it appears you beleive that those of us in industry are quite incompetent and know nothing about our own industry, well, nothing that couldn't be fixed by bringing in some "real" engineers.

I might ask, are you familiar with Hubberts estimates of reserve growth? Him believing in things both "magical" and "enigmatic" apparently?

And please, before you rush off and classify him a consultant from Canada or some other BS which just brings out my natural tendencies to bash amateurs, realize I consider him at least an equal to whatever "really smart guy" from some other industry you would like to bring into the conversation in the hopes that this somehow MEANS something to the oil and gas industry.
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby WebHubbleTelescope » Sat 28 Jan 2006, 03:34:33

$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')But the fact was that material scientists and engineers like Andy Grove were able to characterize the phenomena within a few years time (mid to late 1960's) and get their process down to a gnat's eyelash speaks volumes about the difference between real engineers and the geologists who believe in magical, enigmatic reserve growth. (I don't know any fab engineer in a bunny suit who believes in "enigmatic" oxide growth)


While I can't comment on common engineering practices in other industries, I would venture the following.....it appears you beleive that those of us in industry are quite incompetent and know nothing about our own industry, well, nothing that couldn't be fixed by bringing in some "real" engineers.

I might ask, are you familiar with Hubberts estimates of reserve growth? Him believing in things both "magical" and "enigmatic" apparently?

And please, before you rush off and classify him a consultant from Canada or some other BS which just brings out my natural tendencies to bash amateurs, realize I consider him at least an equal to whatever "really smart guy" from some other industry you would like to bring into the conversation in the hopes that this somehow MEANS something to the oil and gas industry.


Hubbert is no hero of mine. He is just a guy who gave it a shot as far as I am concerned. The Canadian consultants provided some additional data that stressed that "heavy oil" gave much higher reserve growth. Go figure that oil sands fall into the heavy oil category.

[hr]

Here is a plot of the A&R data using averaged delta corrections. The horizontal lines are +/- standard deviation.

Image
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: I'm confused. How do you calculate and model this data?

Postby ReserveGrowthRulz » Sat 28 Jan 2006, 10:50:27

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Hubbert is no hero of mine. He is just a guy who gave it a shot as far as I am concerned. The Canadian consultants provided some additional data that stressed that "heavy oil" gave much higher reserve growth. Go figure that oil sands fall into the heavy oil category.

[hr]

Here is a plot of the A&R data using averaged delta corrections. The horizontal lines are +/- standard deviation.

Image


So....now Hubbert....on a Peak Oil website qualifies as "hardly a hero"? Wow....talk about the amateurs thinking highly of themselves.

Three points, you really should stop referring to them as "Canadian consultants", reserve growth isn't limited to heavy oil as any of the articles I referenced demonstrates, and you never answered an original question, which was, if you don't like Root and Attanasi's reserve growth measurements, or Hubberts, or Verma's, then what might yours be, and how do you account for it within any of your modelling?
User avatar
ReserveGrowthRulz
Permanently Banned
 
Posts: 813
Joined: Fri 30 Dec 2005, 04:00:00
Top
PreviousNext
Post a reply

Return to Peak oil studies, reports & models

Who is online

Users browsing this forum: No registered users and 0 guests



cron