Your first query

FRED takes plain-English questions. The phrasing matters less than the content. This page covers what to include so FRED can answer well, and what common phrasings work best.

#Starting out

Open FRED chat and type. There is no command syntax. You can write the way you would describe the problem to a colleague. FRED will ask follow-up questions if it needs more information.

The fastest way to get a useful answer is to include three things in your first message: a location, a scale, and a preference.

#Good questions to ask

Here are examples that work well, paired with the kind of answer FRED will return.

NOTE
"Find biomass for a 20MW facility near Redding."
→ returns a recommended site, supply curve, and projected LCOE
NOTE
"How much sustainable biomass is available within 50km of Burney, CA?"
→ returns total BDT/year, number of contributing clusters, and a map overlay
NOTE
"Compare procurement cost between a Quincy and Susanville facility, both at 15MW."
→ returns side-by-side cost and supply for both candidate sites
NOTE
"What's the leverage ratio for fire risk reduction in Shasta County?"
→ returns the cost-vs-fire Pareto curve with the knee identified
NOTE
"Show me cluster supply over 10 years if I run a 25MW plant near Auburn."
→ returns a year-by-year supply curve with depletion modeling

#What to include

FRED works best when it has these inputs. You can leave any of them out and FRED will ask, but giving them up-front skips a turn.

  • Location. A town, county, watershed, or coordinate pair all work. "near Redding", "in Plumas County", and "around 40.5, -121.6" all resolve fine.
  • Facility scale. Either as MW capacity ("a 20MW plant") or as biomass demand ("about 40,000 BDT per year"). FRED knows the standard conversions between common technology types.
  • Preference. What you want to optimize for. Cost-only is the default. If you want fire-risk weighting say something like "prioritize fire risk" or "weight fire at 0.05". If you want multi-year, say "over 10 years".

#Following up

FRED remembers the conversation. After an initial query, you can refine without restating everything:

  • "What about with a 60km radius instead?"
  • "How does that look at α = 0.1?"
  • "Project this over 15 years."
  • "What if I move the facility 10km north?"
  • "Show me only the clusters in the very low fire risk tier."

FRED keeps track of the location, scale, and preferences you established. Follow-ups generally come back in 1-2 seconds because much of the underlying data is already in cache.

#Common mistakes

A few patterns that confuse FRED or produce less useful answers:

  • Asking about places outside California. FRED is currently scoped to California. "biomass near Portland, Oregon" will return an out-of-coverage notice.
  • Asking about specific cluster IDs. Cluster numbers are internal identifiers; phrase questions in terms of location and scale instead.
  • Asking for absolute lowest cost without realism constraints. The cheapest cluster in California is isolated and supply-limited. Always include a scale; "as cheap as possible" without context is not actionable.
  • Confusing total biomass with annual supply. When FRED says a region has 38,000 BDT/year of supply, that's the annual sustainable yield given current prescriptions, not a one-time inventory.
TIP
If FRED's answer doesn't match your intuition, ask "why?" or "what drove this?". FRED can show its work: which clusters contributed, what their predicted costs were, what the haul distances looked like.