# MacKay Minutes 2016/06/16

## How did we do on our todo list?

• Andy
• Begin writing up our results so far.
• like Madison, will think about how to turn air temperatures into soil temperatures
• The Journal of Hydrometeorology presents us with a possible strategy based on "inverse problems" -- they use phenology to forecast climate; Climate = f(phenology). If we can invert that, we can find phenology = f - 1(Climate)....
• usually f isn't even a function, however: it's something more general, a relation; we're missing crucial information, and f isn't invertible.
• Laura:
• will try to create surrogate variables and see how they regress against the coordinates of the 10 sites over time.
• Ensemble Mayflower
• Talk to geologists and biologists
• longitudinal strongest
• will think about how to turn air temperatures into soil temperatures
• will use the SVD approach to create the small multiples.
• will call a climate specialist
• will check with geologist about soils of Nova Scotia, and why there's a longitudinal effect.
• was going to check on a strange value in one of the data sets -- was anything said in the text about it? Should we look back at the raw data for that year, to see if we can figure out what went wrong? Better not to just throw out a value, when we might be able to explain it.... If we doubt, however, we could try estimating it using our SVD-based approach.
• Anonymous?:)
• sensitivity of results using SVD estimation
• writing optional arguments into Mathematica....
• Steve
• Would think about the SVD technique, and try to make sense of it.

## Moving forward

• We need to begin writing up our results (it's best to add them to the paper as they are discovered -- it's always easier to throw things out at the end than to remember where something was, or how you did it). I've begun, but we all need to start adding our results.
• Climate and first appearance: what is our model?
• Do atmospheric temp and precipitation constitute "climate"? How about ocean currents?
• Soil temperature as a function of climate -- we've got a model to start with (Chang, et al.).
• Can we use the rank one version of our five favorite species, to reduce noise? We create one new variable from the top five, and follow it through time and space....

## In the News

• For Madison!:( Revealed: first mammal species wiped out by human-induced climate change: scientists find no trace of the Bramble Cay melomys, a small rodent that was the only mammal endemic to Great Barrier Reef
• A Simple Idea Could Help Wildlife Survive Climate Change: Global warming is chasing plants and animals, forcing them to head uphill or north to find suitable habitat. Scientists have considered migration corridors — restored, healthy natural areas that connect current habitats with likely landing spots — as a way to help plants and animals stay a step ahead of climate change.
• Carbon dioxide levels in atmosphere forecast to shatter milestone: Scientists warn that global warming target will be overshot within two decades, as annual concentrations of CO2 set to pass 400 parts per million in 2016
• “We could be passing above 450ppm in roughly 20 years,” Betts said. “If we start to reduce our global emissions now, we could delay that moment but it is still looking like a challenge to stay below 450ppm. If we carry on as we are going, we could pass 450ppm even sooner than 20 years, according to the IPCC scenarios.”