Cold Winter, Spring Plantings, Bayesian Inversion

Winter may be going out with a bang, but that hasn't stopped us from our first spring planting. The next seven days are all forecast to be more than 20 degrees below average, and Thursday Feb 27 is forecast to have a record low of -20 F. This will easily be the coldest winter since 1979 in Southern Wisconsin, and could well have been the coldest winter since 1935 or even 1901. I'm not sure how much of a "heat island" effect the city of Madison has - I know the temperatures at my house would qualify as the coldest winter since 1901 in Madison, but we've averaged about 1.5 degrees lower than the official site in Madison.

Anyway...as winter winds down we've germinated our first seeds - Datil peppers which is an exceptionally hot pepper - a variety  of the species Capsicum chinense.  From the reading I've done online, Capsicum Chinense varieties of hot peppers have an exceptionally long germination periods (12-25 days). Thus it was with some surprise that I noted more than 25% of the seeds had germinated on day 5. An additional 25% germinated on day 6 - I was going to be happy with 25% total germination in 25 days! For anyone interested, I used the "wet paper towel in a baggie" method and kept the baggie behind my wood stove laying on the bricks. I did go to the trouble of finding a place where the temperature varied between 65 and 90 degrees - most of the day it is between 75 and 85 on the bricks, and it cools down below 70 for only a few hours in the early mornings each day.

I've now transplanted all the seeds into wet soil, and they will remain on the bricks, covered by saran wrap until the first shoots break the soil. Then I will move them under a grow light where they will remain until late May.

Since we bought many of our seeds from a place with mixed reputation this year, I might try the paper towel method on a wide variety of seeds just to see whether I need to re-order any seeds. This is where a Bayesian Inversion spreadsheet would come in handy! If I only need to be 95% sure that 35% or more of the seeds will germinate, than I can do a trial on just two seeds - if they both germinate, I can be 95% sure that 36.8% or more of the seeds will germinate. Likewise if 3 of 4 germinate I can be similarly confident of a 34%+ germination rate, and even 3 of 5 gives me 95% confidence that 27%+ will germinate.

Update on lake ice...

Lake Wingra froze on November 27th according to the UW AOS website. We went down there on the 28th and the ice near the shore was about 1.5 inches thick but it was gorgeous - clear, dense, and flawless - perfect for skating. Probably strong enough for a dog but definitely not for a human. Now I'm just hoping the snow holds off until the ice gets thick enough to go skating on; since we have another arctic front moving in late this week it could be as earlier as Saturday or Sunday.

As far as Monona and Mendota ice, I'm moving my best estimates back a few days: December 7th to December 12th for Monona and December 13th to December 31st for Mendota. My best guesses for the two lakes are December 8th and December 17th.

Predicting Madison Lake Ice 2013

I made a model to predict when the lakes in Madison, WI would freeze this year. This could be an interesting calculation, particularly if you like ice fishing or ice skating:)

Given current temperatures and weather forecasts I predict Lake Wingra to freeze sometime next week, between Saturday, November 23rd and Thanksgiving (Nov 28th). My range of uncertainty is much greater for Monona (Dec 8- Dec 25th) and for lake Mendota (Dec 15-Jan 11th). My best guess for the three lakes are 11/24, 12/13, and 12/23 respectively.

You can see surface temps of lakes here: http://www.infosyahara.org/

I built the model quickly so my level of confidence is therefore low but it I am pretty confident we can be out fishing about 20 days sooner than last year - freeze dates last year were 12/21, 12/31. and 1/14. I think this years dates will be closer to long term averages.

Figure 1: Red Line - surface water temp, 2013, Blue Line - surface water temp, 2012. Approximations done with calculations based on air temps, not observations.

Lakes act as a thermal mass, and the larger (primarily deeper) the lake the more cold it will require to freeze. Actually it is probably more accurate to think of the lake as heating the air, then the air cooling the lake - but I guess that's more of a philosophical point. In one extreme, a body of water will freeze almost immediately (think of a puddle or a film of water on a cold surface) so a few hours with air temperatures marginally below freezing will suffice. On the other extreme, a large enough body of water in a region where the average annual temperature is above zero would never freeze. Lake Michigan and Lake Superior are examples although even these large bodies of water will freeze on cold years.

The freezing process can be thought of in three phases - the whole lake cooling to 39 degrees, the surface cooling to 32 degrees, and the surface fusion. Cooling only takes about 1.2 % of the energy of fusion (1:80) but the cooling process down to 39 degrees cools the entire lake while the fusion process just freezes the surface. This is because water is most dense at around 39 degrees and so water from the surface of the lake will displace deeper water until the whole lake is at 39 degrees. Thereafter, it is a quick trip down to 32 degrees, and then takes a relatively long time to freeze (fusion).

For deep lakes therefore, the warmth of the summer is an important consideration since it will take longer to get down to 39 degrees. The three Madison lakes are relatively shallow (9 feet, 27 feet, and 41 feet) so the consideration is not great.

Relieving Suffering, Utility Curves of the Rich, and Resource Constraints

There are several concepts in here that I'd like to develop into full-blown posts - I think they are relevant to macro-economic decisions and decisions on the environment and future of humanity. They could also be cast from a "How to Measure Anything" perspective (for example, there might be some common determinants for whether someone is likely to become rich).

The possibility of a low or negative time discount rate could be the most important conceptual problem that we currently face. The implication of a real negative rate but a working real positive rate would be a mis-allocation of resources on a global scale that would result in a major die-off in humans - a sort of Malthusian moment. A negative real interest rate is a conceptual problem because humans are biased toward valuing the present over the future. A self-imposed negative personal discount rate is likely impossible - but we can probably achieve a synthetic negative personal time discount rate by demonstrating that future suffering will be greatly alleviated with low cost decisions made in the present.

Why do we prefer a $ now to a $ in the future?

1.       1) Because there are things I’d like to do with the $ now that will produce a stream of benefits – if I had the  dollar now as opposed to $ a year from now, there would be a year longer of benefits to enjoy.

2.      2) The younger I am, the more acute my senses and so the greater the sense of enjoyment I get.

3.       3)There is a chance I will die before the benefits would come.

If I die, then does the value of a $ drop to zero? Well I think we can safely assume somebody else would receive that $ instead of me – it would come from somewhere so either the entity giving it to me or one of my heirs would receive the $ instead of me, so the value of the dollar doesn’t drop to zero, but it becomes a little bit hard to conceptualize.

So we go from something easy to conceptualize (a $ now versus a $ one year from now) to something difficult to conceptualize ($ now versus $ benefit to the human race/my progeny in the distant future). It might be a little easier to conceptualize if we think of the enjoyment we get from $1,000 – so it would $1,000’s worth of utility now versus $1,000 worth of utility in the future. We could also translate it into some real object – say 10 barrels of oil or 300 gallons of gasoline.

And we might be able to get a sense of utility curves related to after death by asking questions about utility going anonymously to strangers: if nobody was aware of your decision would you rather give $1,000 worth of benefit to 10 strangers or get $1,000 of benefit yourself? What about 100 strangers? What about 1,000 strangers? What about 10,000 strangers? 5 billion strangers? Keep in mind that the $1,000 would probably mean the difference between life and death for some of these strangers.

Additionally we can get some idea of how much people care about giving money to their progeny by comparing $1,000 benefit to self versus $x benefit to children…$x benefit to grandkids…$x benefit to niece/nephew…$ benefit to niece/nephew’s children?

Children = 50% genetic relationship…grandkids = 25% genetic relationship….niece/nephew = 0-50%, BE 25% genetic relationship…niece/nephew kids = 0-25% BE 12.5% genetic relationship

Another interesting experiment would be to see how people changed these answers depending on their perception of who was receiving the $. Are they relieving suffering? Are they increasing suffering? If you give the $1,000 to a self-absorbed consumer addict or a meth addict could you actually be increasing suffering?

And thus I think we arrive at a bit of a problem – there are individuals (self-absorbed consumerist or meth addict) who perhaps increase their suffering by having more resources but who are also the least likely to perceive a value in giving their resources to others who might decrease their suffering with it.

Also, the choice of people to give anonymously to those who are suffering would be surprising unless you consider that people are incorporating how much suffering they are relieving – that is the utility of the recipient into their decision.

One other interesting thought – the rich are often those who can delay gratification, usually through a combination of a lack of consumption and an ability to work hard (i.e. do unpleasant tasks in the present). These are the same people who (generally speaking) would be most likely to intelligently marshal resources, or put another way their marginal propensity to save is high – that is the poor are often those who would immediately increase consumption. In a resource constrained world, a large increase in real spending power for the masses is simply a physical impossibility. 

How (not) to Catch Roosters - running to exhaustion

Well...today is the day we are thinning our chicken flock. We wound up with 8 roosters and 17 hens, and that is about 5 times too many roosters. The roosters had already isolated and terrorized one of their own - a rooster we nicknamed "freedom ranger" as he would hop the electric fence to avoid the other roosters. Poor guy. He also sealed his fate by biting my finger three times - alas because I was kind of fond of the guy. So freedom ranger was the first bird in the cage - partially to keep him from being terrorized by the other roosters, and partially to save our garden since he was happily pecking through our tomato and cucumber crops in his escapades outside the electric fence.

We figured we had to thin an additional 5 roosters, since leaving more than 2 roosters would be cruel to the hens (Rooster = middle school boy, but with unlimited sex drive). So we started noting which roosters gave us attitude (dragging feathers, doing the rooster dance toward us, or mating a hen directly in front of us) and which roosters were good shepherds to the hen - one good shepherd in particular stood out. It is hard to mark a rooster though because you have to catch it and these roosters are fast (note that they are well known for their mousing abilities). So it is really akin to trying to catch a cat on a 1/8 acre course. So, well, we only got paint on one of the remaining 7 - a particularly offensive and aggressive rooster.

We decided we would shut the chickens up in their mobile coop/roost last night and that would make it easier to get the roosters in the morning. So the big morning arrived and we trooped down there as a family to catch the remaining roosters. Unfortunately, we only were able to get 2 of the remaining 7 roosters coming out of their house! Somehow 5 roosters managed to get by 2 humans even though the opening from the coop is only about 18'' by 24''. So this meant we still needed to catch 3 more roosters.

Catching the remaining 3 roosters was a challenge - and among the three we wanted to catch was the attitude laden chicken, Mr. paint tail, who was essentially an Olympic chicken athlete. We struggled for 30-45 minutes chasing the chickens in vain as our kids got increasingly bored and agitated. Finally, I remembered that an essential strategy that humans have historically used with game including big game like deer was to run them to exhaustion. So, picking one rooster I began to focus on just keeping him running for as long as I could. It worked! After 5-6 minutes he cornered himself, exhausted, under our turkey box. In the process I had exhausted myself, but we now had a working strategy. Emily and I then tag-teamed the remaining 2 roosters - taking 1 minute turns keeping the chosen rooster running. Invariably they would become exhausted in 3-6 minutes, although mr. green tail was truly impressive in both his speed and stamina.

So, I think this is really how not to catch roosters. The only advantage this strategy had was that it allowed us to interact with the roosters and pick the best behaved ones - but next time I think I'll just catch them at night when they are sleepy.

Hugelbed and Biochar bed, Part 1

I've been working on a Hugelbed, otherwise known as a woody bed - the basic idea is to put a layer of wood underneath the soil which leads to a carbon rich, aerated, microbial-rich, and hydrophilic garden bed. Traditionally the beds were built with fallen woods that were too soft or punky for other uses. The beds are given a woody core with soil piled on top - with the bed traditionally reaching as much as 6-8 feet high at its center.

These beds have recently become more popular in the United States - with a common variation of first excavating out some soil so that part of the wood is below grade and the final bed only a couple of feet high. The bed I've created is a mixture of the two - I excavated roughly 1-2.5 feet below grade but am building the bed up relatively high so that its center will be 3-5 feet above grade when complete.

This shows the site after excavation but prior to starting construction. Some of the materials (the excavated dirt, wood pile and a good start on the wood chips) are stacked on the upslope

This hugelbed is constructed on contour, meaning the West lip of the bed is a good 12 inches lower than the East lip. The bed is 8-10' wide and about 65 feet long. This hugelbed has the following layers - first, the area below grade is filled with wood chips to the lip of the lower (West) edge. Second, the wood core is stacked on top of the wood chips. Third, a layer of sod is piled on top of the wood. Fourth a layer of old cow manure, old hay and straw, and more wood chips are spread over the sod. Fifth, a layer of hot (fresh) cow manure. And lastly, the soil that was originally excavated from the bed is piled on top.

Here you see the excavated site on the left side oft he picture with the materials piled up on the right side of the picture

Here you see my faithful golden retriever eager to help - for scale, the lip of the West side of the trench is roughly 30 inches high.

Some of the features of the bed are: First, the wood-chips and decaying wood will act as a sponge for water eliminating the need to irrigate the bed even in a dry summer. Since the bed is on a decent slope and is constructed on contour it should also capture some of the water that would otherwise run off the pasture. Second, as the wood decomposes, it will naturally aerate the soil as this wood is less than 1/2 the density of soil - this eliminates the need to till the soil. Third, the wood is a natural host of Mycorrhizal activity. From Wikipedia: Mycorrhizas form a mutualistic relationship with the roots of most plant species. While only a small proportion of all species has been examined, 95% of those plant families are predominantly mycorrhizal.^[3] They are named after their presence in the plant's rhizosphere (root system). More on mycorrhizal benefits below.

Before woodchips...

And after woodchips - we got about 70 yards (2000 cubic feet) of wood mulch from nearby municipality. They delivered it to my driveway for $200. This is a truly monstrous pile of woodchips - we used about 800-1000 cubic feet of them to fill in the base of the bed.

Sugar-water/mineral exchange 

This mutualistic association provides the fungus with relatively constant and direct access to carbohydrates, such as glucose and sucrose. The carbohydrates are translocated from their source (usually leaves) to root tissue and on to the plant's fungal partners. In return, the plant gains the benefits of the mycelium's higher absorptive capacity for water and mineral nutrients due to the comparatively large surface area of mycelium: root ratio, thus improving the plant's mineral absorption capabilities.

Plant roots alone may be incapable of taking up phosphate ions that are demineralized in soils with a basic pH. The mycelium of the mycorrhizal fungus can, however, access these phosphorus sources, and make them available to the plants they colonize. Nature, according to C.Michael Hogan, has adapted to this critical role of phosphate, by allowing many plants to recycle phosphate, without using soil as an intermediary. For example, in some dystrophic forests large amounts of phosphate are taken up by mycorrhizal hyphae acting directly on leaf litter, bypassing the need for soil uptake. Inga alley cropping, proposed as an alternative to slash and burn rainforest destruction, relies upon Mycorrhiza within the Inga Tree root system to prevent the rain from washing phosphorus out of the soil. In some cases, the transport of water, carbon, and nutrients could be done directly from plant to plant trough mycorrhizal networks that are underground hyphal networks created by mycorrhizal fungi that connect individual plants together.

Suillus tomentosus, a fungus, produces specialized structures, known as tuberculate ectomycorrhizae, with its plant host lodgepole pine (Pinus contorta var. latifolia). These structures have in turn been shown to host nitrogen fixing bacteria which contribute a significant amount of nitrogen and allow the pines to colonize nutrient-poor sites.

View of the Hugelbed from the north, after wood-chips. At the far end of the bed we have stacked wood on top of pallets to start a bonfire - we are going to try  to turn the south-most 12 feet of the bed into "biochar."

View of the south end of the bed, with the wood stacked and ready to burn.

View from the South end of the bed looking North.

View of the Hugelbed from the West (downslope) after the wooden core is stacked in the center of the bed.

Now for the fun...

...particularly if you are a pyromaniac

After starting to smother the biochar. We made a crucial mistake of starting to smother it with wood-chips and THEN dirt. That, combined with the fire laying on another pile of wood-chips led to a rather significant problem...

On day 5 the fire is still smoldering. As the burning wood collapses, it creates large air holes, making it very difficult to keep the fire smothered.

By day 7, it is clear the smoldering fire will consume my entire bed - it has already spread an additional 5 feet north of my original plan. I am forced to dig out all of the woodchips and fill it with dirt to create a fire break.

Digging down 2.5 feet next to a smoldering fire is smelly and uncomfortable work. I'm glad when I'm done.

Fire seems to finally be out on day 16. We have now also covered the entire core of the bed with  sod, hay, and straw. View from the South looking North. Note the firebreak about 15-20 feet down the bed.

Now, the bed has been covered by a good layer of old cow manure with lots of hay and straw mixed in which we got from our neighbor.

The next stage will be to add a thick layer (8-12'') of fresh cow manure (from last winter) gathered from our neighbors barn. I will also begin to plant fruit trees into the north side of the bed.

I'm hoping that the Nitrogen from the cow manure will soak into the wood core and wood-chips so that we get even more Nitrogen stored up for future years. I don't have much hope for plantings this summer - but we'll try some vine crops, peas, and otherwise will just do a cover crop like Hairy Vetch.

A pictorial celebration of one of the coldest March/Aprils in the past 100 years in SW Wisconsin. First shot is taken at the beginning of March with 2' snow on the ground, and throughout i'm standing on our well head.  The last shot is ~April 15th after our last dusting of snow of the season.