LLMs

[castlebravo92]

Overall, not that bad. Some of the dates need tweaking (but that's more of an issue with my prompt than anything else). Definitely useful for pushing through writer's block (by way of comparison, this was the paragraph I created before I decided to throw an LLM at it):

American cities in the Twilight:2000 are quite different from what they were just a few years prior. No city escaped unscathed. Many were targeted with nuclear strikes, but even the ones that were not targeted suffered greatly from civil unrest, famine, and epidemics that came after Federal, state, and local governments collapsed, organized food distribution ceased, and access to clean drinking water became a thing of the past. Even in areas where elements of the Federal or state governments managed to maintain order from late 1997 to 2001 (Salt Lake City, Denver, or Minneapolis, for example), the cities themselves are a pale shadow of their former selves, with only 1/10th to 1/3rd of the amount of people living there as before the war.

[3catcircus]

This is good stuff. A few more prompts should allow it to generate actual battle narratives and current unit status, as well as expanding to world-wide events.

[HaplessOperator]

One imagines that that 15-25% of people left in most cities are going to die off pretty quickly once there stops being any water to drink and once nobody's delivering bottled stuff.

Take Seattle. The entire remaining populations of Seattle, Bellevue, Renton, and Kirkland would have to essentially converge around Lake Washington to have any hope of having water to drink. There's a reservoir farther out into King County towards Enumclaw, but - like Lake Washington - its viability wouldn't last long if there turned out to be 3/4 of the populations of the actual cities in the area running out and pissing and shitting and throwing garbage into it, and the people that actually need it to survive would probably quite literally start shooting people to death over it.

Hell, literally first thing I'd have done if I was up there and something like this kicked off was start blowing those bridges past Renton, Maple Valley, and Black Diamond to deliberately keep most of the populations you're talking about from getting out and to where we were. You'd only have to keep most of them out for a week or two until the mass die-offs began, and the vast majority of people in Seattle would do literally nothing to contribute to anyone's survival once the lights went out, and would - to most of the rural populations out there - represent nothing more than an encroachment and vast consumption of any resources they might already have.

You're talking about hundreds of thousands of people who have literally no life skills beyond a service economy and that don't even know you have to shit downriver. I had the misfortune of moving up there for about eight years from a lifetime in rural WV, and the prognosis of any given person in Seattle surviving past the lights going out and the pantry going empty... suffice to say I think your LLM is being incredibly generous.

Most of these larger cities have populations that have willingly and deliberately spent entire generations shaping themselves - mentally and physically - into people who are abjectly incapable of surviving the conditions we're talking about, and no amount of will can overcome a lack of training, preparedness, or knowledge.

There's a multitude of reasons that the Congressional commissions on EMP damage estimated a 90% die-off of the population in the first year, and that doesn't even count the first hard winter.

[castlebravo92]

So, Seattle has an advantage that it actually gets rainfall. Some urban areas like Phoenix, it's going to be tough going.

Also, LLMs don't do math THAT great yet. At their core, they are glorified "what's the next probable word" regression algorithms, so a number that sounds somewhat realistic is impressive enough (to me, but I'm easily impressed).

Also, with a limited strike, yes, the national power grid goes down for good, but in intact cities, water service would run for a while off portable generators (until the gas ran out).

I expect there would be a panic and evacuation of at least 50% of the population in any cities that weren't hit out of the expectation that "we're next". The other 50% would be either be people who couldn't leave (old, sick, young, taking care of old, sick, young, or like in any weather disaster, staying behind to loot).

The immediate die off would be from people who need meds to stay alive. Total wild ass guess, but I figure 10% of the pop.

For cities part of the critical defense industrial base, the military would be sent in to secure production facilities and workers (congrats, you're drafted into the "army").

In the T2K v1 timeline, the water and the military in the cities (probably) lasts until about Aug-Sep 1998 when the fall harvest fails and the military unceremoniously pulls back to defensible cantonments. Incidentally, the US keeps 1 to 2 years of grain on hand, so this accelerated famine timeline only makes sense if combined with something like the US depleting its grains stores by sending it to war torn Europe to keep the flagging war effort alive.

If you take a city like Dallas, it has a land area of ~385 square miles, which translates into 245,000 acres (right now, it has about 90,000 undeveloped acres). It couldn't self-feed a million people (city limits pop in 1997), but 90,000 acres could give 200,000 people a fair go of it. Of course, it wouldn't be a "city" at that point so much as multiple communities, neighborhoods, or camps defending a piece of arable land or a salvage area. 200,000 gets you to 760 people per square mile or what I consider the max rule of thumb density without trade (~2 to 2.5 people per acre).

Long winded way of saying an 80% population reduction of an "intact" (non-nuked) urban area is a decent rule of thumb in areas with agricultural potential. 90-95% reduction for areas with lower food production potential, more violence, etc.

I think the EMP / Carrington 90% die-off projects are a bit off.

[castlebravo92]

Prompt: estimate the carrying capacity of Dallas, TX, assuming a general loss of electricity production aside from small portable generators, and the loss of refined petroleum products. Assume no trade with outside parties.

Grok's answer:
Estimating the carrying capacity of Dallas, Texas, under the conditions you’ve described—a loss of electricity production (except for small portable generators) and the loss of refined petroleum products, with no trade from outside parties—is a complex task. It requires assessing how these disruptions would affect the population’s ability to meet basic needs like food, water, shelter, and sanitation, while relying solely on local resources. Let’s break this down step-by-step, making reasonable assumptions based on Dallas’s geography, population, and resources.

Dallas has a population of about 1.3 million people in the city proper, with the broader metro area exceeding 7 million. For this estimate, I’ll focus on the city itself, as the "no trade" condition isolates it from the surrounding region. Carrying capacity in this scenario is the maximum number of people the area can sustainably support given the constraints.

Key Factors

1. Food Supply
   Without electricity or refined petroleum, industrial agriculture collapses. No fuel for tractors, no refrigeration, no powered irrigation pumps. Small generators might run limited equipment, but fuel for them would be scarce and finite.
   Dallas sits in the Blackland Prairie, a fertile region historically good for farming. Pre-industrial methods (manual labor, animal power) could be used, but the transition would be slow, and most urban land isn’t currently farmland.
   Local crops like corn, wheat, and sorghum could be grown, alongside cattle (Texas is a cattle state). However, only about 1-2% of Dallas County is farmland today—roughly 5,000-10,000 acres. A sustainable yield might be 50-100 bushels of corn per acre annually with primitive methods, providing 1-2 million calories per acre. For 10,000 acres, that’s 10-20 billion calories yearly, enough to feed 10,000-20,000 people at 2,000 calories/day, assuming all land is converted and harvested efficiently.
   Hunting (deer, rabbits) and fishing (Trinity River, local lakes) could supplement this, but game would deplete quickly under pressure from 1.3 million people.
   
2. Water Supply
   Dallas relies on reservoirs like Lake Ray Hubbard and Lake Lewisville, fed by the Trinity River. Without electric pumps, water treatment and distribution fail. People could manually fetch water from rivers or lakes, but contamination risks rise without purification (small generators might power basic filtration for a tiny fraction of the population).
   Annual rainfall in Dallas is about 37 inches. Rainwater harvesting could help, but with no infrastructure, most would be lost. A household of 4 might collect 20,000 gallons/year with effort, enough for drinking and minimal use, but scaling this to 1.3 million is impractical without preparation.
   
3. Shelter and Heating
   Existing housing would remain, but no electricity means no air conditioning in brutal Texas summers or heating in winter. Wood burning could work for heat, but local forests (e.g., Great Trinity Forest) would be stripped fast. Small generators might power fans or heaters for a few, but fuel runs out.
   
4. Sanitation and Health
   No electricity or fuel means no sewage systems or waste removal. Disease would spike—cholera, dysentery—especially with a dense population and limited clean water. Medical care would revert to pre-industrial levels, reliant on local knowledge and scavenged supplies.
   
5. Energy and Transport
   Portable generators provide minimal power, but fuel (gasoline, diesel) isn’t replenished. Horses or human labor replace vehicles, drastically slowing movement of goods and people. This reinforces isolation and limits resource gathering.

Estimation
In this scenario, Dallas’s carrying capacity hinges on food and water, as they’re the most immediate limits. Pre-industrial societies often supported 50-100 people per square mile in fertile areas without trade or technology. Dallas County is 873 square miles, but much is urban, not arable. If 5-10% of land (50-100 square miles) could be farmed effectively, that suggests a capacity of 2,500-10,000 people at subsistence levels. Water access might support more initially, but sanitation and disease would cut numbers fast.

Realistically, the initial 1.3 million would face mass starvation, disease, and conflict within months, reducing the population dramatically. After a chaotic collapse, a steady-state carrying capacity might settle around 5,000-15,000 people, relying on local farming, river water, and scavenged resources. This assumes survivors adapt quickly to primitive methods and avoid total societal breakdown.

This is a rough estimate—actual outcomes depend on human ingenuity, weather, and how fast order unravels. What do you think about this scenario? Want to tweak any assumptions?

[castlebravo92]

Modification / conversation:

assume a fair percentage of yards and parks are convertible to gardens or hand cultivated small farms and adjust estimates accordingly

Alright, let’s adjust the carrying capacity estimate for Dallas, Texas, by assuming a fair percentage of yards and parks can be converted into gardens or small hand-cultivated farms. This change significantly boosts local food production, so I’ll rework the numbers while keeping the other constraints—no electricity beyond small portable generators, no refined petroleum products, and no trade with outside parties—in place.

Revised Assumptions
Dallas has an area of about 340 square miles (city proper). Much of this is urban—residential, commercial, and industrial—but yards, lawns, and parks offer untapped potential. Let’s estimate how much land could be converted:

• Residential Yards: Roughly 40-50% of Dallas’s land is residential. Assuming an average lot size of 0.2 acres (typical for urban/suburban homes) and half of each lot (0.1 acres) being convertible (front/back yards minus structures), with 150,000-200,000 households, that’s 15,000-20,000 acres potentially farmable.
• Parks and Green Spaces: Dallas has over 400 parks covering about 20,000 acres total (e.g., White Rock Lake Park, Kiest Park). If 50-75% of this—say, 10,000-15,000 acres—can be turned into cultivated land (excluding dense forest or steep terrain), that adds significantly.
• Vacant Lots and Medians: Urban vacant land and roadside strips might contribute another 5,000 acres if aggressively repurposed.

Summing this up, a "fair percentage" could mean 30,000-40,000 acres of Dallas’s 217,600 total acres (340 square miles) become gardens or small farms—about 14-18% of the city’s area. This is optimistic but plausible with widespread effort.

Food Production
Using pre-industrial farming methods:

• Yield: Hand cultivation might produce 50-100 bushels of corn per acre (1,500-3,000 lbs), or 1-2 million calories per acre annually, depending on soil, water, and skill. Let’s average at 1.5 million calories/acre.
• Total Calories: For 30,000 acres, that’s 45 billion calories/year; for 40,000 acres, 60 billion calories/year.
• Population Supported: At 2,000 calories/day per person (730,000 calories/year), 45 billion calories supports ~61,600 people, and 60 billion supports ~82,200 people.

This assumes all land is used efficiently, with crops like corn, beans, and squash (calorie-dense and viable in Texas). Gardens could also grow vegetables and fruits, adding nutrition but less caloric density. Chickens or rabbits raised in yards might supplement protein, though feed would limit scaling.

Water, Sanitation, and Other Limits

• Water: The Trinity River and reservoirs remain accessible by hand, and rainwater harvesting from rooftops could support more people with garden-scale irrigation. Still, purification stays a bottleneck without power.
• Sanitation: More people mean more waste. With 60,000-80,000 spread across the city, composting toilets and manual waste management could work if organized, but disease risk grows.
• Energy: Small generators might power tools or pumps for a few, but most labor is manual. Wood from parks and yards fuels cooking until depleted.

Adjusted Carrying Capacity
With 30,000-40,000 acres of gardens and farms, Dallas could sustain 60,000-80,000 people long-term, assuming survivors adapt to hand cultivation, preserve seeds, and manage water and waste. This is a huge leap from the 5,000-15,000 in the prior estimate, reflecting how urban land conversion mitigates food scarcity. However:

• Transition Chaos: The initial 1.3 million would still face a brutal die-off before stabilizing at this level, as converting land and growing food takes time (6-12 months for a first harvest).
• Density: At 60,000-80,000, population density drops to 175-235 people/square mile, low enough to avoid immediate resource wars but high enough to strain sanitation and wild game.

So, with yards and parks turned into small farms, Dallas’s carrying capacity rises to 60,000-80,000 people. This assumes a best-case scenario of cooperation, skill-sharing, and no major climate disruptions. Does this align with what you had in mind, or want to explore further adjustments?

[castlebravo92]

chatgpt o1:

estimate the carrying capacity of Dallas, TX, assuming the loss of electricity generation and the availability of petroleum products. Some electricity generation would be possible using portable generators. Assume yards, parks, and greenbelts would be convertible to gardens for food production, but otherwise assume no food is imported or traded for from surrounding areas.

Short Answer:
A rough but defensible estimate is that under extreme “grow‐your‐own” conditions with no outside food imports, Dallas might only support on the order of 100,000–300,000 people long-term. In other words, the current population (over 1 million) would be far greater than the city’s carrying capacity if it had to rely solely on backyard and greenbelt agriculture plus minimal generator power.

[HR]

Key Factors in Estimating Carrying Capacity

1. Land Area and Usable Growing Space
   • The City of Dallas covers about 385 square miles, or roughly 246,000 acres.
   • Not all that area can be converted into productive farmland: roads, buildings, parking lots, etc.
   • After subtracting infrastructure, a generous upper bound might be 10–30% of city land converted for growing food (i.e., 25,000–75,000 acres).
2. Yield per Acre Under Intensive Cultivation
   • Conventional agriculture might require 0.5 to 2 acres to feed one person for a year.
   • Highly intensive “biointensive” methods can theoretically feed one person on about 0.1 acres, or about 4,000–5,000 square feet.
   • In practice, urban gardening is less efficient due to soil, water, and shading issues, so a realistic average might be 5–10 people per cultivated acre.
3. Water Availability
   • North Texas experiences hot summers and periodic droughts.
   • Irrigation and water storage become critical; without reliable electricity, pumping water is difficult.
   • Water constraints may sharply limit the number of acres that can be effectively cultivated.
4. Population vs. Potential Food Production
   • If 10% of Dallas’s area (~25,000 acres) were intensively farmed at 5–10 people per acre, that might feed 125,000–250,000 people.
   • If 30% of land (~75,000 acres) were farmed with similar yields, one might theoretically reach 375,000–750,000.
   • Real‐world limitations (uneven soil, water shortages, etc.) suggest a much lower, more realistic threshold.
5. Energy and Resource Constraints
   • Loss of large‐scale electricity generation compromises city services (water treatment, transport, etc.).
   • Portable generators help, but require petroleum products, which may be scarce without imports.
   • Resource bottlenecks reduce overall agricultural and logistical efficiency.

Putting It All Together

• Best‐case (highly optimistic): Several hundred thousand people could survive if the city reorganized around intensive urban agriculture with ample water and expertise.
• Moderate/realistic scenario: Likely 100,000–300,000 inhabitants, assuming a concerted effort to convert yards and parks into productive gardens.
• Compared to Dallas’s actual population (~1.3 million): Without imported food, Dallas is currently far above the sustainable threshold.

Although urban farming can produce surprising yields, true self-sufficiency in a modern city is extremely challenging. A realistic carrying capacity for Dallas—absent outside food sources—would likely be only a fraction of its current population, on the order of a few hundred thousand at best.