This wonderful chart, put up by Veronique de Rugy over at The American, is even better than the chart that I’ve used off and on for months to show the staggering deficit projections resulting from the Obama Administration unsustainable economic policies. As per de Rugy’s explanations, here’s what the color bars mean:

• the orange bars (up through 2008) show the actual deficits from the last year of the Clinton Administration and the first seven years of the Bush Administration (kind of makes you wistful, doesn’t it?)
• the purple bars show the deficit projections as of September 2008, two months prior to the Presidential election
• the black bars show the deficit projections as of January 2009, after the bailout efforts of the outgoing Bush Administration
• the red bars show the deficit projections as of this August, resulting entirely from the actions of the Obama Administration and the current Congress

In other words, you can’t blame Bush for anything in red. As de Rugy points out (emphasis mine):

. . . Obama is right to note that he inherited a large deficit in fiscal 2009. But as we can see here, he is responsible for growing the deficit beyond expectations in fiscal 2009 and thereafter. In fact, in its January 2009 projections, the CBO built in most of the Bush-era policy spending, including the TARP bailout (which President Obama voted for as a senator) and the takeovers of Fannie Mae and Freddie Mac. In spite of his rhetoric, President Obama bears most of the responsibility for the red part of the bar in fiscal 2009, which includes, among other things, some auto bailouts and $31 billion of additional funding for the omnibus bill, the share of the stimulus funding spent in that fiscal year.

. . . Obama’s deficits are frightening but they promise to get worse. Each month that goes by the president adds spending to the deficit. The August 2009 projections for instance, do not include any of the president’s healthcare reform spending and they assume that the “temporary” stimulus spending will not be prolonged past fiscal 2011. Finally, they also assume that the economy will recover soon and that it will grow enough to generate increasing tax revenue, in spite of the president’s plan to impose new taxes and regulations on the private sector. In other words, the deficit will likely continue to deteriorate beyond the current projections.

Good thing that the Obama Stimulus package has done so much for unemployment and the economy, eh?  ..bruce w..

Hat tip to Mother, May I Sleep with Treacher? ..bruce w..

Word via io9 that Purdue has put together a map of the US showing their guess of CO₂ emission density.

They call the System “Vulcan” and it’s interesting because there is no sensor grid for CO₂. Instead Vulcan derives a guess on CO₂ density from local air pollution data from the EPA sensor grid. (here as well)

To create the Vulcan maps, the research team developed a method to extract the CO₂ information by transforming data on local air pollution, such as carbon monoxide and nitrous oxide emissions, which are tracked by the Environmental Protection Agency, the U.S. Department of Energy and other governmental agencies.

While it’s fun that they are trying to show this data, the science behind it is (in my opinion) quite flawed. First off, they do not have the means of directly measuring CO₂, so they take a guess based on other air pollutants. One thing that drives me insane is opaque scientific data. If you are confident in your approach, share the source data and let us “trust but verify”. Who knows how they cooked up their guess?

A preliminary analysis of the Vulcan data suggests that previous maps of U.S. fossil fuel emissions were inadequate for current scientific and policy-making needs, Gurney says.

“When you compare the old inventories to Vulcan, the new data show atmospheric CO2 differences that are as large as five parts per million in some U.S. regions in the late winter,” he says. “The levels in the global atmosphere only rise one and a half part per million every year, so this is the equivalent of three years of global emissions in the atmosphere that isn’t where we thought it was. This will be important for policy-makers and is enormous from a scientific point of view. It’s shocking.”

Anyone surprised that they are finding data that shows things are more severe than they thought? Give me a break people, how about some scientific honesty here?

Gurney is Kevin Gurney of the Purdue Climate Change Research Center (PCCRC). He is also the author of “Mending the Ozone Hole“, which was another big scare in the 90′s that caused massive misallocation of capital for something that turned out not to be a real problem after all.

I am surprised not to see massive CO₂ “red spot” in eastern central California; an area of heavy volcanic emission of CO₂, Carbon Monoxide and other “green house gasses”. This omission alone calls the accuracy of this visualization into question.

Professor Gurney – how about CO₂ uptake as well? Did you take vegetation into account in your models? The sad truth is we just don’t know do we? We get the “cooked” data and have to trust someone who clearly is pushing an agenda. Advocacy journalism is bad, advocacy science is much worse (almost as bad as advocacy intelligence).

If I could wish for something here, it would be for the Global Warming shills on the take from “Big Green” would be more transparent in their sources and methods, and allow everyone who is skeptical to either convince themselves or point out the flaws.

A system such as Vulcan will be an important tool to understanding our lack of understanding of emissions and what impact (if any) they have on the environment. But it will require transparent methods for data collection, aggregation and display to be credible in what has become a politically driven, agenda subscribing sector of atmospheric science.

Anyone who has been following the financial news knows that banks and other large financial institutions are under a growing amount of strain given the misguided investments in speculative ventures over the past 5 years, including mortgages and exotic credit / investment packages.

While chasing down what state a company like Merrill-Lynch might be tough, banks are required to report data to the FDIC. The bad news is you end up with a massive pile of tabular text data that makes for potent nap fuel. Readers will recall that my team has been working on new techniques for finding and presenting massive information sets, a technology based on mashups and the semantic web we call Boomerang.

We decided to apply the Boomerang approach to this mass of data, adding in some ideas from the groundbreaking gapmider.org effort to create a dynamic overview of just how tough it’s getting for banks and their mortgage portfolios.

Charts and data after the jump.

[Update - Henderson: Fixed some problems with the Flash charts and Internet Explorer]

First lets look at the top 8 banks and their mortgages that are 90+ days late. Below is a flash charting system, feel free to use the controls and experiment. We chart the total assets of the bank along the horizontal axis, the value of loans that go 90+ days late on the vertical, and the size of the circles represent the total loan portfolio for that bank. You can set the charts in motion by hitting the “Play” button and stop them at any time. Hovering over a circle will show you the value for that data point.

Our charts step forward in time for Q1-2002 one quarter at a time, reading directly from the bank’s own FDIC reports.

Bank Portfolios – 90+ Days Late

Some fascinating detail here, most notably Wells Fargo (my bank) seems to be the leader in holding onto 90+ Day late mortgages. Bank of America seems to be keeping things under control, while Washington Mutual seems to be having problems creeping up as well. Citibank takes a healthy jump in Q4 2005 and then keeps increasing slowly.

The take away from this chart is that most of the banks that we are showing are keeping their 90+ Day lates under control, at least up until the end of Q4 2007 (The last date we have data for thus far).

The picture is quite different if we take a look at loans that are charged off by these banks (recently referred to as “Write Downs”). Some notes to look out for: JPMorgan goes through several steep increase / decrease cycles as they buy smaller, weaker banks and clean up their balance sheets.

Again, feel free to use the controls and experiment. We chart the total assets of the bank along the horizontal axis, the number of loans that are written off as non-performing on the vertical, and the size of the circles represent the total loan portfolio for that bank. You can set the charts in motion by hitting the “Play” button and stop them at any time. Hovering over a circle will show you the value for that data point.

Bank Portfolios – Total Charged Off Mortgages

With the exception of JPMorgan’s action buying out weaker banks, everything stays calm until Q2 2007, when everything starts decaying in a hurry. Over the last half of 2007 Citibank, JPMorgan (Chase) and WaMu start exploding with non-performing mortgages. Bank of America seems to be keeping things reasonably under control, but you can clearly see the strain on the balance sheets for all the banks we are tracking.

The Bottom Line

The banks are under a growing amount of stress, but unless you are a hard core accountant or analyst, it can be tough to wade through the public data to figure it out. Tools like the trend animator provide a means to convey complex data sets in a manner that lets anyone recognize what is happening.

What is clear is the strain becomes quite serious in 2007, and is increasing rapidly towards the end of the year. Being a big bank does not seem to isolate you from crisis right now, as even some of the largest are facing increasing trouble.

Just where this ends is anyone’s guess. We are facing problems on multiple economic fronts at the same time, and our banks are likely to be the ones to show pain first. To be fair this chart does not show companies that are strict mortgage lenders, as they have no accountability to FDIC. Countrywide is an exception because they have their own bank.

Keep watching this space, we will work to bring you an update once the Q1 data is published.

Special thanks to Nathan Redding of the Boomerang team who coded the Flash charts

[Update - Henderson] If you really like these charts, you can work with a more robust version over at the Boomerang mothership – we have data for more banks, more kinds of trends we can show and lots of variations on the theme.

[Update 02/19/08 - 0838 PST]Updated plot graphics can be found here

There has been a lot of news that a poor broken US intelligence satellite is going to be a practice target for a missile defense test (press write up can be found here)

From a personal standpoint I am very upset about this. The satellite in question is referred to in public as USA 193. It is a National Reconnaissance Office bird that was first launched in December of 2006 from Vandenberg AFB in California into a polar orbit. Right from the start it had problems, and there is no unclassified explanation of what happened.

We do know that it was an expensive failure. These kinds of platforms tend to cost hundreds of millions of dollars including the cost of the launch vehicle. Prime contractor Lockheed (along with Raytheon most likely) are going to have to reimburse the government for most of the cost of this expensive dud. In addition I am fairly certain the intelligence community needed this asset in place to replace a bird that has probably come up to the end of its life time.

That aside, I was curious about how the shoot down would work. Word from the Pentagon was that a US Navy Aegis cruiser would be taking the shot using a juiced up AM-3 missile. This same missile has been very successful in tests shooting down ballistic missiles. Initial reports were that they intended to “take the shot” from somewhere near the coast of Alaska.

In addition NASA wanted to get the shuttle on the ground by Wednesday to open the window for the shoot down, and ensure the shuttle was not still in space when this test took place.

Lastly I would assume that this test is going to be the kind of thing that the DoD and its contractors for missile defense want a lot of data on. That means ideally they would take the shot after Wednesday the 20th, when the satellite is near the west coast of the US, but not over it. As a bonus if you could make it so the target was illuminated by sun while the ground was dark or in twilight so that the equipment they will use to film and photograph the impact will have maximum contrast.

Plugging all of that into the orbital calculators gives us something like this:

Which is only good if you are a space junkie like me. But being handy with maps (as some of you may recall from the San Diego Fires), I thought we could make a much nicer view:

Please note this represents my “best guess” on when they would do it. This could be very far from correct, but I would think it is the confluence of a lot of “wants” and represents at least one optimum for the shot.

The green blob off Alaska is where they will likely park the Aegis cruiser, the yellow arc represents the orbital track USA 193 will be on as the sun is setting on the 21st on the west coast. Impact in the red circle would ensure that it happened away from the US mainland, but close enough to be captured by ground based instruments that will be very valuable in judging how well the SM-3 and the Aegis targeting system worked.

Orbital track for USA 193 Thursday here: http://www.calsky.com/csrender.cgi?&object=Satellite&number=3&sat=29651&tracker=

Special thanks to Milcom Monitoring Post for very nice orbital data and background information.

Jerry also pointed out a National Academies report chaired by Calit2′s Director, Ramesh Rao, entitled Improving Disaster Management: The Role of IT in Mitigation, Preparedness, Response, and Recovery (you can order it at the link just given).

Jerry and Calit2 are putting together a site with copies of web-based content (such as ours) that provided information during the fires. If some of the rest of you who were running sites during the fire would be willing to have copies of your fire-related posts on that site, drop me a line, and I’ll give you Jerry’s contact information.

On a vaguely related note, I’ve moved the San Diego Fires blogroll below the regular blogroll, since the fires appear to be good and out. I’ll leave on the blog for the time being.  ..bruce w..

One of the founding concepts behind my new software venture is that the current trend in IT is reaching a point of diminishing returns. One of the wonders of the past 20 years has been the power of the US economy, largely thanks to insane levels of worker productivity in this country. Much of that is directly attributable to the adoption of information technology by American companies to improve the efficiency of almost any task they do more than twice. The concept behind Boomerang is that the next big jump will come from making people smarter, rather than making it easier to order paper for the printer.

This means being able to discover and extract information from many different sources automatically, and have back end machines and programs that can shape this information into a form where it can be combined and re-combined based on topics of interest. For those readers with a background in the Intelligence Community, it’s the classic All-Source problem and how do you let everyone know what is going on without swamping them with a million details so that the message is lost.

We have built several really interesting systems using this approach, known more widely as “Enterprise Mashups“, and many of the ideas of how it works and why it matters stem directly from earlier work on DoD projects such as DCGS (opens a PDF).

On Sunday October 21st, our first clue that something was going on came with the wind that was blowing strongly from the east. The slight but unmistakable smell of a brush fire. The first question on everyone’s mind is – where is it and where is it going? Flipping between the local stations yielded a lot of football and no information.

We needed to decide quickly if the horses in the eastern hills near Valley Center would be threatened or not by this fire. Checking the local news paper’s web site resulted in little or no information. Checking the California Department of Forestry was scant to none. Waiting to be told there was a problem was asking for trouble. What was needed was “Situational Awareness” of the threat now and the projected threat a few hours out, and we were low on data to feed that decision.

We had heard the fire was in Ramona, and we knew how the wind was blowing. We also knew some information about the terrain. Using a combination of Google Earth, physical maps and our knowledge of the burn pattern in 2003, we started laying out what we knew, what we suspected and what we though the fire would do next. We made the decision there was a very real threat to the stables and mobilized people to evacuate the horses, several hours ahead of the official word to move out from the County.

Throughout the first 2 days of this fire, information resources could not handle the stress of people trying to access them over the web. The entire information structure started to buckle. It was at this point I decided that based on what we knew from listening to AM radio, scanners and what news we could get that it should be possible to display where the fire was and where it was headed. I did what any intel guy would do, I started making maps. Just by whim I published one to the web site, and people were drawn to it. So much so that we got our web site temporarily knocked off line.

It cannot be stressed enough the importance of getting accurate, condensed information in the hands of everyone as quickly and directly as possible. The web is an outstanding resource for doing that, but the tools to do it quickly and easily are lacking.

Some of the best source of data that were used:

Pacific Southwest Research Station – These are the guys who were suppling the infra-red data that I was overlying on Google Earth with such great results. If I were President or Governor I would buy them a pair of Predator UAVs to boost their capability. The proved they could deliver results in a time of need, and they are the best bet (in my book) to move forward.

San Diego Union Tribune – They were a bit slow on the up-take, but quickly got in gear putting out information. Specifically there were some false starts with their Fireblog, that they eventually got off of their network and onto a high capacity server.

Cal Fire – This web site folded like a lawn chair for the first 2 days under the traffic, and did not have much useful detailed information until the 2nd or 3rd day. Suggest California think about whether this is fulfilling its intended mission. It’s saving grace is it gave me the rough parameters of the Witch Creek Fire late Sunday night after trying to load it for over an hour.

San Diego County Emergency – Another web site that quickly imploded under the network traffic. They even started publishing detailed PDF maps towards the end of the second day. I am hoping they are still smarting after their failure for the first 36 hours and someone in power will build that site out to what it should be in terms of content generation and capacity.

Google Maps – Three cheers for Google Maps and the community around it. Early on all kinds of people got just as frustrated as I was about what was going on and the lack of information around it. We started making maps on Google and sharing them. The software allows you to combine multiple maps in one view. This made it easy for people like me to combine knowledge from many people at once and create a common view. The only down side is you sort of had to know what you were doing to get really good results out of it.

KPBS – Knocked off the air early when the fire took out their broadcast antenna, they made some of the more detailed Google Maps of this event. Well done.

Tools that handled the load:

Google Earth – Decent visualization tool for 3D terrain and geo-spatial data. KML file format was very helpful in swapping data with other people watching the fire.

Mars Edit – Blog (for the Mac) posting tool was my front end to the web site. Proved itself as a huge help in quickly creating posts with a large amount of graphical data easily.

Word Press – The Blog software that runs this site. A few strange quirks under load but did ok.

OmniGraffle – Mac based drawing tool that I used to annotate the maps and visualizations. Good stuff all around.

Thanks again for reading this blog and supporting us.

Bruce

Some of the maps and visualizations we produced…

October 22nd

October 23rd

This image was created in Google Earth, looking South / Southwest. You can clearly see the fire still working it’s way down the mountain.

The Poomacha fire, which has been burning parallel to the Witch Creek fire across North County, has containment lines around almost half of its 43,000 acres. It has merged again with the Witch Creek at its northern and southern flanks, and flames have crested Palomar Mountain. Full containment is expected by Monday; no ETA yet on control. About 2,000 homes in the vicinity remain threatened. The fire currently is burning toward Lake Henshaw and Sunshine Summit. Areas still of concern include Rincon and Valley Center, Hidden Meadows and Pauma Valley. Residents of those areas need to monitor fire conditions and information closely.

Below is data from Pacific Southwest Research Station overlaid onto Google Earth. Markups and annotations are mine.

Poomacha Pauma Reservation Area – Oct 26 2000 PDT

This view is looking east at the lower shoulders of Mount Palomar’s north side. You can clearly see a very hot area of the fire threatening the Pauma reservation area. The fire is burning through the canyons near Gomez Trail.

Poomacha Palomar Summit Area – Oct 26 2000 PDT

This view is looking at the summit and north west. The fire is now burning up to Doane Valley Road and is still hugging Palomar Divide Truck Trail. The burn is closer to the Palomar Observatory, but is still a safe distance away.

Poomacha South / La Jolla Amago Area – Oct 26 2000 PDT

The south end of the Poomacha fire is burning in the canyons on the flanks of Mount Palomar near Highway 76. This view is looking South East.

According to news site, the majority of aircraft fighting fires in San Diego county are now tasked to the Poomacha Fire, this includes 1 helitanker, 11 fixed-wing tankers and 13 helicopters.

Poomacha Looking West – Oct 25 2000 PDT

We can clearly see the fire attempting to push to the eastern side of the summit in this image. It should be noted the fire is burning now on the North East side of Palomar, and the majority of the population is well South (left side) of this map,

Poomacha Looking North – Oct 25 2000 PDT

This image shows the northern portions of Palomar burning under the Poomacha fire. Note that the populated areas near Crestline are well South (behind) this view. The fire seems to be burning strongly near Palomar Divide Truck Trail.

Poomacha Looking South – Oct 25 2000 PDT

This view is looking South towards the populated areas. In response to questions from readers, I do not have any word on Mother’s Kitchen, one of my favorite places too.