Skip to content
Fundación Loros

Research and monitoring · Conceptual model

The first
72 hours

Conceptual simulator of the events an ex-pet Amazon parrot faces after being released into the tropical dry forest. Pedagogical model — not predictive.

Run a simulation

What the simulator makes visible

Every release is a cascade of variables that multiply each other.

A single deficit is rarely fatal. What kills — in the field and in the model — are the deficits that compound: low muscle mass plus high cortisol plus poor spatial memory plus hunger.

Moving a single variable immediately reveals how the others reconfigure: a bird with poor flight cannot move far from the release site to find water, its satiety drops, it tries to approach the village where it can be recaptured. The simulator makes that logic transparent so anyone — not only a biologist — can understand why release is hard and why method matters.

Three dimensions

What the bird brings, what we did for it, where we release it

  1. Individual factors

    What the bird IS at intake: body condition, human affinity, prior forest experience, temperament, years in captivity. The developmental imprint does not erase — it is compensated for.

  2. Quality of prior process

    What WE DO for the bird before releasing it: acclimatization, flight reconditioning, body recovery, wild diet recognition, reduction of human imprinting. This is where the sanctuary intervenes.

  3. Site quality

    WHERE the bird is released: hydration and feeding stations, flock presence, low predator density, natural water, distance from human settlements. Geography is not decoration — it is outcome.

Cómo leer los indicadores

El simulador rastrea diez variables hora a hora durante las 72 horas posteriores a la liberación. No son lecturas independientes — están acopladas:

  • Cortisol es el multiplicador. Cuando sube, suprime el forrajeo, degrada la memoria espacial y exagera la respuesta de miedo.
  • Miedo determina si el ave se congela (freezing) o huye descontroladamente. Las dos respuestas queman energía sin resolver la situación.
  • Hidratación es el factor limitante crítico en el bosque seco. Cae rápido en horas de calor; sólo se recupera si el ave encuentra una fuente.
  • Energía y saciedad dependen de saber comer: un ave que no reconoce el alimento del bosque puede pasar hambre rodeada de fruta (neofobia).
  • Capacidad de vuelo es la variable mecánica — si la pectoral está atrofiada, el ave no puede alejarse, escapar de rapaces ni buscar recursos.
  • Memoria del territorio es lo que el ave aprende del lugar durante los primeros días — degradada por estrés alto.
  • Integración a bandada da vigilancia colectiva y reduce ansiedad nocturna. Solo cuenta si hay congéneres en el sitio.

La variable más útil para leer el modelo es movilidad: es vuelo × energía. Un ave puede tener un excelente entrenamiento de vuelo, pero si llega agotada al final del día no puede escaparse de un depredador. La movilidad es el cuello de botella real.

Interactive model

Release a virtual bird

Adjust the individual, prior-process and site factors. Each run is a different parrot under the same conditions. Click "Run new simulation" to see other outcomes, or estimate average survival with 400 runs.

Individual factors

Weight, health and useful reserves at intake (not lipidosis fat).

Imprinting: high = actively seeks people (risk near villages).

Was it a wild bird before captivity? Helps foraging and orientation.

0 = fearful/reactive · 100 = bold/proactive; modulates stress.

Depth of the developmental imprint; worsens stress calibration.

Quality of prior process

Perches, foliage, canopy; improves spatial map and reduces neophobia.

Reverses pectoral atrophy; efficient flight to escape.

Reverse deficiencies and lipidosis before release.

Learn to identify and open forest food.

Wilder temperament; less attraction to humans.

Site quality

Run outcome

Dies · Starvation / exhaustion

Survived until D2 02:00 · 18 h post-release

seed 938551972
prep 30% · site 0%

Parrot diary · first-person narration

Generated with Claude from this run's log.

Click «Narrate this story» so the parrot can tell, in first person, how it lived through these three days according to what happened in the simulation.

Indicators · 72 h evolution

Cortisol (stress)8

§3·§4 cascade multiplier

Fear / neophobia0

§4.2 suppresses foraging and exploration

Hydration32

§2.3 critical limiting factor

Energy / condition0

§2.4bis useful reserves

Satiety4

§2.4bis depends on knowing how to eat

Flight capacity45

§2.4 pectoral atrophy

Territory memory49

§3.3 degraded by stress

Flock integration0

§2.5 collective vigilance

Mobility (flight + energy)29

Displacement range per turn

Distance traveled4.9

Far from release site only if it can fly

Emotional timeline · hour by hour

CalmExploringAlertFearPanicExhausted· faint = night ☾

Event log · hour by hour

D1 08:00
Agua oportunista (rocío/fruto/bromelia)
D1 14:00
Neofobia: rodeada de comida pero no la reconoce
D1 15:00
Agua oportunista (rocío/fruto/bromelia)
D1 16:00
Neofobia: rodeada de comida pero no la reconoce
D1 18:00 ☾
Neofobia: rodeada de comida pero no la reconoce
D1 19:00 ☾
Agua oportunista (rocío/fruto/bromelia)
Neofobia: rodeada de comida pero no la reconoce
Teme al follaje; pernocta en estructura humana (expuesta)
D1 20:00 ☾
Neofobia: rodeada de comida pero no la reconoce
Hambre/sed: logra desplazarse y busca recursos en el pueblo
D1 21:00 ☾
Neofobia: rodeada de comida pero no la reconoce
D1 22:00 ☾
Duerme en percha baja/expuesta (mayor riesgo)
D1 23:00 ☾
Agua oportunista (rocío/fruto/bromelia)
Teme al follaje; pernocta en estructura humana (expuesta)
D2 00:00 ☾
Agua oportunista (rocío/fruto/bromelia)
Hambre/sed: logra desplazarse y busca recursos en el pueblo
D2 01:00 ☾
Hambre/sed: logra desplazarse y busca recursos en el pueblo
D2 02:00 ☾
Duerme en percha baja/expuesta (mayor riesgo)
Colapso energético (no pudo desplazarse a buscar recursos)
faint = night

Conceptual model, not predictive. Values and probabilities are illustrative and do not represent measured empirical rates. Should not be used for operational release decisions.

Lo que el simulador no modela

Para ser útil pedagógicamente, este simulador deliberadamente excluye variables que importan en una liberación real pero que harían el modelo opaco:

  • Enfermedad subclínica — un ave que parece sana puede llegar al campo con parásitos, infecciones respiratorias o problemas hepáticos no detectados. En la práctica eso cambia todo el desenlace.
  • Eventos climáticos extremos — una tormenta el día 1 puede matar a un ave perfectamente preparada. El modelo asume condiciones promedio del bosque seco.
  • Variabilidad genética entre individuos — algunas aves nacen con mejor calibración del estrés, otras con peor. El simulador trata el temperamento como una sola variable.
  • Cazadores específicos — la presencia o ausencia de un cazador local con escopeta no se modela. La variable «alejado de asentamientos» captura parte del riesgo pero no su variabilidad.
  • Interacciones sociales complejas — peleas territoriales, jerarquías dentro de la bandada, parejas que se forman o se rompen.

Las probabilidades del modelo son ilustrativas, no empíricas: representan tendencias razonables según el marco teórico, pero los números exactos no derivan de tasas medidas en campo. Este simulador no debe usarse para tomar decisiones operativas sobre una liberación real. Para eso —en Fundación Loros y en cualquier otro proyecto— se usan los protocolos del método, el seguimiento por radiotelemetría y el criterio del equipo técnico.

Frequently asked questions

Is it predictive? Can I use the result to decide whether to release a parrot?+
No. It is a conceptual model. The simulator's probabilities represent reasonable tendencies within the theoretical framework, but the numbers do not come from field-measured rates. A real release decision requires clinical protocols, veterinary evaluation, knowledge of the specific site, and the technical team's judgment — none of which is in the simulator. Use this tool to understand why release is hard, not to substitute fieldwork.
What does the «parrot diary» generate?+
A first-person narration of the bird built with Claude (Anthropic's language model) from the exact events that occurred in your run. The diary respects the individual traits and outcome: if the bird has low human affinity, it does not write nostalgically about its caretaker; if it died on day 2, it does not write day 3. It is a pedagogical device to emotionally connect with what the numerical model makes visible. It is not the real bird's voice — it is literary interpretation informed by the simulation.
Why is mobility so important?+
Because almost everything else depends on being able to move: finding water, finding wild food, escaping a raptor, joining a flock, moving away from human settlements. Mobility is flight × energy: even if you train flight heavily, an exhausted bird at the end of the day is still easy prey. In the model, the cascade that kills most often starts with low mobility → cannot move away from the release site → cannot find resources → approaches the village → capture.
Why do results vary between runs if the parameters are the same?+
Because each run uses a different random seed. Two birds with the same traits do not have the same fate: one may find water on day one and another not, one may be detected by a raptor and another escape by seconds. To understand the system's average behavior, click "Estimate survival · 400 runs" — that smooths the noise and reveals the structural pattern.
How does this simulator relate to Fundación Loros' actual method?+
The simulator is a pedagogical distillation of the technical framework of the method. The five rehabilitation stages we practice — acclimatization, body condition, flight reconditioning, wild diet recognition, deprinting — are exactly the «prior process» sliders. The operational question is always the same: how do we move each slider upward before releasing? The simulator does not replace the method; it makes it explainable to non-biologists.

Dive deeper into what the simulator summarizes

The simulator is an entry point, not an endpoint. If you want to understand how we move each slider in practice, read the method. If you want to see which trees of the territory sustain which wildlife, explore the living atlas.

Read the methodLiving flora-fauna atlas