[This narrative originally appeared on RangeNet.org, a website that is since defunct. Since I believe it is a useful document, I make it available here.]
Once Upon a SVIM
by: Larry Walker, 5/18/03
Well folks, I am going to talk about BLM’s former “Soil Vegetation Inventory Method” (SVIM) and other things that go bump in the night.
But first, a little background going back most of a century.
Following passage of the Taylor Grazing Act in 1934 at about the height of the “Great Dust Bowl”, the Mc Carren Holiday, theReorganization Act, and other sundry happenings; the Bureau of Land Management (BLM) (aka Bureau of Livestock and Mining) decided in about the 1950’s that it should figure out how much grazing it actually could allow on the public lands (then called the “public domain” since it was still assumed that those lands would eventually be homesteaded or otherwise disposed of).
Back then they knew that they couldn’t get everything right immediately, that inventories are expensive, and that things change over time. So – their strategy was to inventory once to establish a baseline, monitor the results, and make needed adjustments as indicated by their monitoring (does anybody detect the “new” concept of “adaptive management” in this?).
Now the Dust Bowl experts in the form of the Soil Conservation Service (SCS) had developed the “Range Site” approach for inventorying
rangelands thanks to the outstanding work of Dr. E. J. Dyksterhuis. The fledgling BLM looked at this approach, but found a few problems. (1) The methodology required quite a bit of knowledge about soils and their potentials, (2) it gave an objective estimate of condition, without directly quantifying cow chow very well, (3) it was expensive.
BLM decided against the Dyksterhuis approach as being too expensive. After all, they were going to monitor and adjust over time – so they needed something to cheaply give some numbers for initial allocation. Enter the Square Foot Density Method, the Ocular Reconnaissance Forage Survey, and the Weight Estimate Forage Survey.
BLM did LOTS of one-time inventories, one time after another time after another. They somehow never got around to a consistent follow-up monitoring program let alone making adjustments (does this ring a bell with West-side Forest Plan folks?), so they were always starting over with a new and different survey.
Over the next couple of decades a few more statutes were enacted (many under Republican administrations incidentally): Multiple Use and Sustained Yield Act, National Environmental Policy Act, Endangered Species Act, Clean Water Act, Federal Land Policy and Management Act, Public Rangeland Improvement Act, to name a few.
The National Environmental Policy Act (NEPA) threw the first big bump in BLM’s road, but they tried to make an end-run. Permitting domestic livestock on public lands was merely replacing the Bison, so no EIS was needed – they said!
That didn’t play very well, and even BLM’s top brass came to realize that it was dead on arrival. So, they went the “programmatic” EIS route and said that everything would be just fine if we did Allotment Management Plans (AMP’s) on everything, not withstanding that AMP’s had only been developed on a fraction of a percent of BLM administered lands (it is a small percentage even today).
Well, there were some folks who were not quite ready to shell out their cash for this swine in a sack – so they sued – and they won!
While the number changed over time due to various consolidations, BLM was faced with a Federal Court Order to do more than a hundred Grazing Environmental Impact Statements.
This is about where I came into the picture. I was working as a Range Con. (the “Con.” is an abbreviation for conservationist or con artist, take your pick) on BLM’s Prineville District in the mid 1970’s. We were looking up waiting for that grand piano to fall on our heads (deadlines for EIS’s) and getting about zip out of the front office in terms of guidance. The bottom line was that they did not know what in the hell to do and were waiting for EIS number 1 (Challis) to set the pattern for the rest of us to follow.
Being somewhat fond of our skins, and trusting the Washington Office about the square root of zero, we started scoping things out for ourselves in order to be prepared for when we would be expected to do the impossible yesterday.
Our bottom line was that, even if we were doing a good job of managing livestock to meet our statutory environmental goals (we thought), we had absolutely nothing to document or prove that fact!
Back to basics. What did we have and what was it capable of? How were we using it and what were the results? What could we have and how could we use it? What did we want?
Well, about three separate BLM offices came up with essentially the same approach pretty much simultaneously and independently: The use of soil surveys and SCS’s Range Site methodology (what BLM had discarded as a possibility a couple of decades earlier) would provide objective information on natural potential and present condition.
By supplementing that information with a few things like wildlife species life histories, susceptibility of soils/sites/condition classes to erosion, and the like – and doing a bit of modeling – we could say:
(1) We are HERE now
(2) Among the alternatives of where we could be, the best of possible worlds would be if we were THERE
(3) We are going to do XYZ to get THERE
(4) We will MONITOR and if we are not getting THERE, we will change XYZ so that we will get THERE
Sounds a bit like “adaptive management”, doesn’t it? BLM’s problem is not that adaptive management is a bad idea, it is just common sense after all. BLM’s problem is that it HAS NEVER DONE ALL OF THE ESSENTIAL STEPS, and there is no credibility that it ever will!
Anyway, I digress. Back to SVIM!
The Challis EIS bombed, and the bigwigs tripped over themselves looking for a magic bullet. The only rays of hope that they found were some pretty rough-draft proposals from three district offices recommending some kind of adaptation of SCS’s Range Site methodology as an ecological base to work from.
So – they directed the resource science teams in the then Denver Service Center (DSC) to use those proposals as a starting point and develop the ultimate inventory – and to do so by day before yesterday.
Along came SIM, slow moving SIM, dollar spending SIM – along came slow and not so lanky SIM. The Site Inventory Method (SIM) pulled together everything that anybody would ever want to know about everything. It pulled together gobs of generally acceptable studies under one undoable umbrella, and it failed to incorporate a meaningful basis for stratification, sampling, and classification of site potentials and plant communities.
Everybody revolted. A “management” task force was assembled, they were afforded the advice of a science advisory committee, and some of us on-the-ground peons gained indirect access through the science committee.
Based upon their recommendations, the DSC teams did their magic and unveiled the results at a meeting where we were told to “Sink or SVIM” (John Baker did have a way with words).
So was born the BLM’s Soil Vegetation Inventory Method (SVIM). About $15 million per year was spent in the late 1970’s and first couple of years of the 1980’s to perform a baseline SVIM inventory on tens of millions of acres of BLM administered public lands. Computer programs and linear program models were developed and, during its hay-day, SVIM was the largest single application to ever run on BLM’s mainframe computer.
We didn’t have decent online access and database management programs then, and GIS was still a dream, so the results were batch-processed and printed out on paper reports – CASES OF REPORTS compiling and presenting the data in every form that anybody had been able to think of.
Then, something happened that regularly reoccurs during even numbered years divisible by four. It took them about a year and a half, but Jim Watt’s crew in Interior EMBARGOED SVIM. Reports that were sitting on the loading dock ready for shipment to the Prineville District, for example, were shipped to Denver’s landfill instead – and access to the data (particularly any that had any bearing on grazing capacities) was tightly censored with much of it not even available to the BLM districts that collected it!
BLM rewrote its regulations to prohibit basing grazing allocations on “one-time” inventories alone. SVIM was overhauled to eliminate the calculation of grazing capacities, among other things. A subset of the soil and vegetation methodology became the Ecological Site Inventory (ESI).
That is pretty much the story of “Once Upon a SVIM” except for a couple of addendums:
In the mid 1980’s I attended a BLM Monitoring Coordinators’ Meeting where we were informed that BLM planned to purge all SVIM data so that they could use the disk packs it was stored on for other purposes. Part of their rationale was that the field offices had the reports that had been generated (those that had not been censored!) so didn’t need the raw data anymore.
I strongly objected to simply throwing away what was raw data of potentially historical significance as a benchmark for ecological study, and suggested that they should at least archive it to CD’s rather than just discard it. Others at the meeting agreed.
In 1991 (nearly a decade after the initial SVIM embargo), the power structure had shifted enough that the DSC folks were finally able to produce and distribute a CD with the raw SVIM data on it. That’s right, I said A CD! That mass of data that had brought BLM’s mainframe to its knees, and which was eating up so much off-line storage that they were going to throw it away in the mid 1980’s, all fits on a single CD!
The SVIM CD SHOULD be available (for an appropriate fee) from BLM as a public document, but don’t hold your breath that they will even be able to find a copy of it. The CD’s title is:
“Bureau of Land Management: Soil Vegetation Inventory Method (SVIM) Raw Input Data – Data Element Definitions – Record Formats – J C L – Cobol Source Code. BLM Service Center – 1991”
(As a side note, I will speculate that the “Cobol Source Code” on the disk MAY contain the two linear program models that were developed by Colorado State University under contract to BLM for calculating grazing capacities.)
The CD contains varying amounts and kinds of data for the following inventories
BAK1 – Bodie-Coleville Inventory – Bakersfield CA
BUR1 – Cassia Inventory – Burley ID
BUT1 – Headwaters Inventory – Butte MT
CAC1 – Lahontan Inventory – Carson City NV
CEC1, CEC2 & CEC3 – Pinyon Inventory, Cedar City UT
CNC1 – Trickle Mountain Inventory – Canyon City, CO
CRG1 – Kremmling Inventory – Craig CO
CRG2 – Little Snake Inventory – Craig CO
ELK1 – Saval Ranch Inventory – Elko NV
ELK2 – Wells Inventory – Elko NV
CJT1 – Glenwood Springs Inventory – Grand Jct. CO
CJT2 – ??? Inventory – Grand Jct. CO
CJT3 – ??? Inventory – Grand Jct. CO
IDF1 – Big Lost Mackay Inventory, Idaho Falls ID
LCR1 – Las Cruces Inventory – Las Cruces NM
LCR3 – LC and L MFP – Las Cruces NM
LCR5 – White Sands Inventory – Las Cruces NM
LEW1, LEW2 & LEW3 – Prairie Pot Holes Inventory – Lewistown MT
MED1 – Medford Inventory – Medford OR
MAO1 – Price River Inventory – Moab UT
MON1 – San Juan-San Miguel Inventory – Montrose UT
PRI1 – Brothers Inventory – Prineville OR
RIC1 & RIC2 – Henry Mountain Inventory – Richfield UT
RIC3 – House Range Inventory – Richfield UT
ROS1 – Roswell Antelope Study – Roswell NM
SAL1 – Salmon Inventory, Salmon ID
SLC1, SLC2, SLC3 & SLC4 – Tooele Inventory – Salt Lake City UT
SOC1 – West Socorro Inventory – Socorro NM
SUS1 – Susanville Inventory – Susanville CA
SUS2 – Willow Creek Inventory – Susanville CA
VER1 – Duchesne Inventory – Vernal UT
VER2 – Ashley Creek Inventory – Vernal UT
VER3 – Bookcliffs Inventory – Vernal UT
Please note that the information on the CD is “raw data”. For full utility, you will probably also need “User Requirement Document: Soil Vegetation Inventory Method (SVIM). Bureau of Land Management, Denver Service Center” “User Requirement Document, Soil-Vegetation Inventory Method (SVIM), Bureau of Land Management, Denver Service Center” (reference corrected 11/5/05) and the “Soil -Vegetation Inventory Method” manual which is reprinted in BLM’s “Rangeland Inventory & Monitoring, Supplemental Studies, TR 4400-5, 1992”.
As far as tying any particular data to a specific piece of ground, access to one additional piece of information that can only be found in the local field offices is needed. That is the “SWA Overlay”. The Site Writeup Area (SWA) number is how the individual writeups are tied to specific pieces of ground. These SWA’s are delineated on mylar film overlays to 7 minute USGS quad maps or ortho photos. A few SVIM inventories have been captured in a Geographic Information System (GIS), but most have not.
BLM distributed one copy of the raw data CD to each State Office range monitoring coordinator, and presumably to the Denver and Washington DC offices. As Oregon/Washington BLM’s monitoring coordinator, I had additional copies made which I distributed to our district offices. What other states did, I do not know.
Given this informal distribution, BLM may be hard pressed to find a copy of the raw data CD to copy for you, though the “User Requirements Document” and the SVIM manual should be available on microfiche, and the local field offices should have the SWA overlays.
RangeNet to the rescue! In the interest of historical preservation of this public record data, RangeNet can provide a copy of the CD for the equivalent of the cost of materials and postage. Just send your name, address, and three first class postage stamps to:
RangeNet.org SVIM
2850 SW Cedar Hills Blvd. #251
Beaverton, OR 97005
Addendum 11/5/05:
All files from the CD are now available for downloading online.
That’s all folks, except to note that this little episode of purging the scientists and burning the books did not take place in Nazi Germany of the 1930’s and 40’s – it happened right here in the good old U. S. of A. just a score of years ago! Unfortunately, an encore appears to be in the offing.
Larry Walker
Range Conservationist (retired)