Awesome work, have you looked into using Racket's fork of Chez? It supports some additional extensions and architectures from base Chez that would seem to beneficial as a compiler target, at least that was the case a couple years ago.
I agree I don't think it will always be the case but in the 1-5 year term the economics dictate that is what you get, full battery backup to run something like a data center is expensive even if technically feasible. It is not 1:1 backup but significant additional marginal gas production is going to come online with battery and solar build outs, that is what you see in grid queues across the US.
You are ignoring the fact that batteries like many other flex grid assets have diminishing marginal returns to arbitrage power. That is why you see them in composition with other flexible base load resources in a market system. We could reform the system to have other objectives.
During the Biden administration almost 40 trillion ft^3 of natural gas was produced annually breaking all records. I do agree the policies of energy were more all the above during Biden and renewable build outs were incentivized additional to record gas production.
I think the biggest material factor between the two administrations is coal capacity factors on the grid are increasing again which would not happen under an alternative admin. We were on a good path towards phasing out coal completely with natural gas and an increasingly significant amount of renewable energy.
As much as I wish it were different I think this would be the outcome under any administration. Natural gas in the USA is just stupidly cheap inflation adjusted, is energy dense, power plants require few land resources, pipeline infrastructure exists, robust and deep supply chain, etc. I don't hold much hope of the USA kicking the habit as it is such a competitive advantage. If we don't burn it locally it will be exported at a huge margin for geopolitical advantage. 1/3 of the CO2 reductions in the US power sector are due to renewables, 2/3 are from natural gas. Renewables would need to be backed up by additional gas capacity, although the average capacity factor would be reduced with a more robust renewable build out.
Some of the largest proposed data centers in the US are in Pennsylvania, West Virginia, and Ohio. The electricity infrastructure is important so is cheap natural gas from the Marcellus shale, natural gas pipeline infrastructure and relaxed regulations.
Any advice will really depend on where you live and the dominant source of local error. These are all good options for the US and if the one main source of error is near surface winds around complex terrain then you can also look into WindNinja from the National Forrest Service.
When you say runs that is an ensemble model forecast which is different than a deterministic model forecast. Any type of weather model sensitive to initial input state errors will have outlier model runs.
Traditional physics based weather models also rely heavily on physical parameterization for sub grid scale processes (think clouds, microphysics of rain sleet snow, etc) so even the deterministic physics models are learned approximations from data.
ECMWF has an experimental AIFS direct observational prediction model (AIFS-DOP) that has become competitive with their physics based IFS model on certain metrics just in the past year.
Heat domes are already killing off cattle and reducing crop yields, even in western countries? There was significant cattle die off in the US during 2022/2023. Europe just experienced 10,000 excess deaths due to heat impacts and a reduced grain yield by almost 8-10%. There have not been correlated heat dome events over multiple regional bread baskets (yet) but those are going to get increasingly more probable. We have a lot of adaptability on the food side though as reducing meat consumption is an easy behavioral change. Climate change impacts is one reason why beef is currently so expensive in the US.
Its a bit more complicated than that, you have a period during the spring called the spring persistence barrier that makes skillful prediction of a developing el nino difficult. What is impressive is that multi month model predictions before this spring barrier had a historic el nino developing which turned out to be correct (an out of sample prediction).
Its regionally complicated because increased tree cover can lower albedo (reflectivity of radiation) which offsets the co2 removal benefits. Forrests can tip from co2 sinks to co2 sources which is happening in different regions of the world currently.
Being next to a nuclear power plant but attached to a regional grid is not that relevant in terms of total additive emissions from a data center build out. You have to look at the change in emissions of the grid as a whole. Companies are incentivized to care about disclosing a narrow boundary view of scope 1 & 2 emissions but we live in the real world not a spreadsheet.
US electrical emissions YOY increased in part due to data center build out and energy demand.
The nuclear generating station is part of a larger system. Say you build enough data centers in your local area to use up all the station power so that 80% is no longer exported, what is getting built somewhere else to make up for that missing generation the data centers now use? Its not new nuclear. When you add data center load to a grid how the additional generation is supplied is really what matters in terms of impacts.
It is much more than the Colossus sites, I would look at the capacity additions in the regional grids you are apart of and what is being funded to meet these increases in demand from data centers. The majority of it is natural gas generation and a sizeable but minority amount of battery storage. Just outbidding people for a relatively fixed amount of clean nuclear ignores the second order effects of adding large loads like these. What happens in your area does have larger impacts.
The amount of nuclear generation is roughly fixed (minus the refurbishment of three mile) in that region. If you add additional large load to the grid and outbid other demand for that power you are just shifting the load you replaced to other sources, which in PJM region would be mostly gas (new or existing) and delaying the decommissioning of existing base load coal plants. Renewables in that region are unfortunately a small percent of electrical generation.
I do agree that other demands like water consumption are overblown and could be largely regulated to enforce best practices. What infrastructure we are building as a society to meet this load demand is going to be the lasting impact of this generational infrastructure investment and it's looking like that will be mostly fossil fuel based in the near to mid term.
But in PJM they are almost entirely being powered by natural gas and coal? Even if you contract out power from a nuclear plant some other plant on the grid is now enjoying a higher capacity factor, at the margin natural gas.
The data center in question in Utah was marketed as a 9GW full build out natural gas facility more than twice the electrical generation of the entire state. Coal electrical production in the US increased 13% last year.
Of course it is, to be healthy you need to be born Mississippi is double the OCED average for infant mortality. Then you can track obesity, diabetes, heart diseases, etc. throughout life. The only area where all this supposed American exceptional wealth confers an advantage in healthcare is post op outcomes where higher penetration of CT / MRI scans can make a measurable difference. Even then the difference is not large. You don't get a DECADE of life expectancy difference without an objective difference in health.
Emigration of Germans to the USA is only 10,00-15,000 people per year since 2000 and is trending down especially since 2016 to lowest levels ever. Emigration of US citizens to Germany started from almost nothing around the same time and is now trending up to almost 1,500 people a year. We are talking about a small number of decidedly non median people in either case but even then the rate of change is clear, neglecting the fact that Germans learn English in public school while Americans typically don't have public schooling in German. My family is descended from German immigrants who came during the 1930s to escape right wing authoritarianism.
Why is the median considered "poor"? The true poor in Mississippi are way worse off than those in Germany. If you neglect everything it takes to live a good life like public capital, education, healthcare, time off with family, a retirement, total years on this earth and ignore the insane inequality in Mississippi then sure the median numbers are not that far off.
median household income is not higher in missisippi vs germany, especially not true if you adjust for the value of healthcare, education, and social benefits including time off work.
These are typically representative of cost performance per watt of one part of a more complex deployed energy system. Things like the aluminum / steal for the container / framing, copper / aluminum for the transmission and wiring, land and labor for installation decline at much less aggressive rates or increase over time.
In almost all pareto optimal least cost energy system models that I've seen, high penetration of solar, wind, batteries plus some minority amount of (clean) baseload power is the most capital efficient energy system.
The water at these temperature / depths has a lot of dissolved salts and minerals so it's not (human / ag) usable. Modern designs are closed loop systems where production wells bringing the hot water to the surface go through a heat exchanger to a different working fluid to drive the turbine and then is re-injected back into the reservoir. There is consumptive water use for fracking the reservoirs in these types of enhanced geothermal systems, but beyond that it's more water redistribution in the area around the well systems where re-injection and production lead to different pressurization from pumping / natural ground water replenishment rates.
Newberry Volcano is too good to be true in that there are few (outside of Yellowstone) equivalent sources of geothermal awesomeness at similar depths in the USA. Good for research bad for generalization of drilling costs to hit similar temperatures. There are federal protections for geothermal drilling anywhere near Yellowstone.
It does work technically I think it is still an open question if it can work economically. There are issues of commercially viable flow rates / thermal decline rates that are harder physical limits you run up against and the pilot design doesn't address. In human timescale terms it's more like heat mining rather than renewable heat due to thermal depletion rate vs replenishment rate. These systems have a targeted lifetime of ~20-30 years and net power will decline over this timespan.
The core breakthroughs were working with partners to develop PDC bits that enable high rates of penetration in drilling out these horizontal wells in high temp granitic rock and then demonstrating plug / perf fracture networks that have a high engineered permeability in these source rocks to support economical flow rates and heat transfer. These were considerable advances over previous efforts.
There will be other learning by doing advances in how you structure your power plant design to take advantage of these to make practical long term power production possible (well spacing and injection / production placement / flow rate and temperature decline management).
In traditional fault hosted (not magmatic) geothermal the convection of the water up the fault brings the thermal energy closer to the surface where drilling depths are economical. This convection heats the surrounding rock and over hundred thousand - million of years brings the background temperature around a large volume at depth surrounding these systems considerably above traditional background geothermal gradients. By drilling into a much larger volume of impermeable hot rock surrounding a very small permeable fault hosted section you can considerably enhance the power potential of a traditional fault hosted geothermal system (the E in EGS). That is what Fervo is doing and why their projects are situated right next to traditional geothermal power plants.
The assumption is that if you can increase drilling efficiencies enough then you don't even need a fault hosted or similar system to bring that energy close to the surface, you can just drill down deep enough to get at similar temperatures. That is a big assumption in the economics.