I think it is time to upgrade the technology I have used for 14 years to monitor the temperature of my compost pile.
The graph below is live, as long as I keep going out and reading the temperature and coming back in and updating the Google spreadsheet. In 1998 I inherited the old YSI thermometer device and found a 7.5 volt mercury battery for it. The battery is still good, and I even have a spare (also 14 years old). I hope that the participants at LEAFFEST can help me find a replacement technology for the beast.
The featured pile was built on September 10 from a three month accumulation of kitchen, garden, and yard waste, some old hay mulch, freshly cut alfalfa, 100 chopped corn plants, a quart of organic fertilizer (5-3-4), and 20 gallons of spent tomatoes, cucumbers, peppers and cabbage. The pile was turned on October 7.
Twice this year I walked around for more than an hour flying a kite-lofted camera that I thought was taking photos, when in fact it was just fooling me into getting exercise. On the first occasion the shutter controller battery died, and the other time the camera just got confused and stopped shooting. I think I sent radio commands to shoot too frequently and it decided to take a nap. This happened another time as well but I happened to reel everything down after only 20 minutes of not taking photos while I was jogging here and there. I have modified the controller so the battery lasts much longer, and I have learned to be patient when sending radio commands, but I am certain that the rather fragile KAP rig electronics will surprise me again with their unexpected inactivity. So I have been looking for a way to monitor the camera operation from the ground. I have been brainstorming with some friends about how to do this, and Don Blair, a physics graduate student at UMass has been prototyping an alert system based on a $20 pair of walkie-talkies. I couldn’t stand hearing about all the fun he was having in Amherst using MOSFETs to remotely push the walkie-talkie PTT button, so I decided to build my own solution.
It has been 14 years since we set up our first study plots south of the Alaska Range. I just returned to Fairbanks from a week-long trip to re-census the plots located above the upper forest limit at the Canyon Creek and Monahan Flats sites. There were a few scattered spruce trees and a total of a few dozen spruce seedlings marked with numbered tags in the quarter-hectare plots in 1998. The trees have not changed much, and only 8 to 10% of the seedlings have died. The surprise was the number of new spruce seedlings that have established in 14 years.
The NIR modified and normal A495 cameras with AuRiCo controller. Synchronous, vertical photos were taken every 16 seconds.
Seedling density increased by a factor of 2.9x at Canyon Creek and 2.2x at Monahan Flats. The similarity between the two sites suggests that the seedling invasion may be widespread. These seedlings have a lot to endure before they become trees, but if similar numbers of seedlings continue to establish and many of them become trees, these plots will no longer be above treeline in 50 years.
I was able to get some aerial photos of most of the plots after I had marked each seedling with pink flagging. I flew a dual camera KAP rig with normal and infrared-modified cameras lofted by a kite. I will be experimenting with false color IR (NRG) and NDVI images which can be created with information from both cameras.
I flipped the joystick 1200 times to trigger the shutter (the 4 GB SD card filled up 10 minutes before I pulled it down).
The S95 captured 1170 photos in 1.5 hours while the Fled flew over the relic Gold Dredge #3 of the Fairbanks Exploration Company (F.E. Company). It was built in place in 1927-1928 and dug its own pond which it moved around until 1958. Hundreds of acres of the Chatanika River valley were turned into these concentric ridges of tailings as the placer gravels were devoured by the floating dredge and disgorged by the systematic arcs of its conveyor arm.
A month ago, on May 10, I noticed a wad of dry grass in the birch tree outside my home office window. It was obvious what it was, and a noisy pair of Baltimore orioles soon confirmed that a nest was being constructed. Four days later the nest building seemed to be mostly completed, and I stopped taking photos (click them to enlarge).
May 11, 2012. Female Baltimore oriole weaving a nest in a paper birch tree
On May 5 my infrared KAP rig made its first photographic flight. The shutters of two Powershot A495 cameras, one modified to record only near-infrared (NIR) light, were triggered every 10 seconds by an AuRiCo KAP controller. The Levitation delta kite hoisted the rig over the northern edge of the Salisbury Swamp, VT. In about 1.5 hours, 420 synchronous pairs of normal and NIR photos were captured. It has taken a while to process just a fraction of these images, but the results are promising. First, some kite porn:
Note the Henry's Handle hanging from the belt. The kite line is cleated to that while I attach the Picavet and run through the pre-launch checklist.
The huge kame terraces in my town support a dry-tolerant forest of oaks, beech, and red maple, and as the new leaves emerge it is easier to tell the species apart at a distance. So I have been hoping to fly a kite over the terraces in early May and see if the trees delineated the terraces. The wind was gusty today, but it was supposed to calm down a bit in the afternoon, so I hiked up the hill and launched the nine foot Levitation delta. It was after 6:00 PM when I first triggered the S95 with the Futaba transmitter, and the light was wonderful even if the wind was not. It mostly cooperated until it slowed down enough to send me running upwind barely fast enough to keep the line out of the trees. But I had 900 feet of line out for a while and the camera was well out of the clearing were it could see the tree tops as I never had.
The PLOTS visible/near-infrared camera tool was ready to fly on Thursday, but the Fled could not find enough wind. So I switched on the two cameras and the timer and walked around the field making photos every 10 seconds.
At midday on Sunday. With a north wind, we looked into the sun a lot that day.
There was good sun and enough wind to loft a camera on both Saturday and Sunday last weekend, and the apples were just starting to bloom. So I knocked on Barney Hodges’ back door and introduced myself. Barney bought a 200 acre dairy farm in 1974 and started planting apple trees. His son runs Sunrise Orchards now, and Barney seems quite content to be in his beautiful old farmhouse in Cornwall surrounded by thousands of apple trees that he planted. He graciously permitted me to fly a kite and take photos, and I did.
Grass fire along the Lemon Fair River, April 18, 2012. Photo by Preston Turner.
I thought I had finally done everything right. The Flowform 16 was flying steadily in a 15 mph breeze (with gusts to 20) and easily lofting the full KAP rig with Canon S95. My plan was to take vertical photos for combining into an orthophoto image of a hayfield that had been partly burned two days earlier by the farmer. Preston had posted some dramatic photos of the evening blaze advancing across the field, and the burn pattern looked impressive when I drove by the next day looking for way to get closer to it. On my return a day later, I avoided asking permission for farm road access, and just parked two miles away and walked along the Lemon Fair River. Because the good wind provided plenty of lift, I opted for the full KAP rig and radio control instead of just the camera pointed down and shooting every ten seconds. That gave me the option of also taking some oblique shots, and the 5:00 PM shadows of burr oaks and silver maples were already enticing on the new grass. Radio control also allowed me to release the shutter only when the rig was quiet between sways, and to take far fewer photos than the ten second intervalometer would have taken.
Ever since I backed the balloon mapping Kickstarter project, I have been thinking about what to do with the matched pair of visible and infra-red cameras I will receive next month. I’m intrigued by the idea of making NDVI images by combining information from the two cameras. NDVI maps of vegetation can reveal new ecological patterns, but I still have no idea what I might learn with this tool. I have also been thinking about how to deploy the cameras to take aerial photos. The level of my Kickstarter pledge entitles me to a kit with cameras and other things, but it is not yet clear what other things will be included. So when it was 50° a couple of weeks ago and tolerable to work in the garage, I started playing with a piece of one of the aluminum storm window frames that were thrown behind the shed. The goal was to make a cheap, lightweight, self-leveling rig to hold two cameras pointing in the same direction: down.