Friday, October 2, 2026

CULAP Data Center — Friday SportsCenter-Style Post-Game Report

For Friday, October 2, 2026
CULAP Data Center and Education Group | Grades 6–12
Feature: CULAPSE Darkaggriculture — Kaggriculture Season 0

Welcome to the CULAP Data Center

Good afternoon, student analysts. Welcome to our Friday Post-Game Data Report, where we take the excitement of a sports post-game show and apply it to computer science, mathematics, agriculture, data science, and artificial intelligence.

This week's matchup wasn't played on a basketball court or football field. It happened on a 720-turn virtual farm.

Our competitors were three versions of the CULAPSEDataFarm autonomous farming agent, developed during Kaggriculture Season 0. Each agent had to make decisions about production, maintenance, harvesting, selling, capital and reinvestment.

The final question wasn't simply:

“Who scored the most?”

For the CULAP Data Center, the better question is:

What can the numbers teach us about why the agents performed differently?

The Scoreboard

Here are the scores displayed on our Kaggle submissions page at the end of our Season 0 analysis period:

AgentStrategyScore
Agent V0 / v0.6DWHEAT baseline199.4
Agent v0.1 FPP-07Six-tile WHEAT production191.4
Agent v0.1-RC3Validated autonomous WHEAT agent193.7

That gives our Grades 6–12 analysts an immediate mathematical problem.

Agent V0 exceeded RC3 by 5.7 points and FPP-07 by 8.0 points.

But don't stop at the scoreboard.

Play of the Game: The Baseline

The most interesting result is Agent V0.

We developed additional logic after the baseline, yet the original V0 currently shows the highest score of the three submissions.

In sports, adding more plays to a playbook doesn't guarantee more points. The same principle can appear in computer programming. A more complicated algorithm isn't automatically a more effective algorithm.

That's our first Computer Science Stat of the Week:

Complexity ≠ Performance

Students can now investigate whether additional decision rules helped the later agents—or caused them to spend resources or act differently at critical moments.

By the Numbers

Suppose we call Agent V0's score V=199.4 and RC3's score R=193.7.

Their difference is:

199.4 − 193.7 = 5.7 points

For middle-school students, this becomes practice with decimals, subtraction and interpreting data.

For Algebra students, we can represent performance differences using variables and functions.

For Statistics students, one score isn't enough to establish why an agent performed differently. We need repeated observations, episode data and measures such as mean, range and variability.

And for Computer Science students, we can examine the actual agent policies responsible for producing those observations.

One scoreboard—multiple levels of mathematics.

Film Room: The 720-Turn Replay



Sports analysts study game film. CULAP student researchers study simulation data.

Our Kaggriculture episode replay becomes the equivalent of the film room.

Instead of asking only whether an agent won or lost, students can examine:

BUY SEED → PLANT → WATER → MAINTAIN → GROW → HARVEST → SELL → REINVEST

Every decision becomes an observable event.

When did the agent plant?

Did it maintain its crops?

How long did capital remain tied up?

When did crops become harvestable?

When did the agent sell?

Did reinvestment occur early enough to produce another crop cycle?

These are questions about algorithms and systems, not just gaming.

Wednesday, July 22, 2026

Exploring the Living Laboratory: A Research Introduction to Caldwell Preserves for CPS Students

1. Caldwell Preserves as a Premiere Research Site

As CULAP researchers, we view Caldwell Preserves not merely as a park, but as a critical urban ecology "living laboratory." For Chicago Public Schools students, this site offers a front-row seat to the complex struggle for survival at the intersection of wildlife behavior, human recreation, and environmental health. By studying how animals adapt to our city’s infrastructure—from the North Branch Trail to the picnic groves—we generate the data necessary to protect Chicago’s biodiversity. Our mission is to move beyond passive observation and apply scientific rigor to understand the pressures of urban life on our local fauna.

Research Question: How are wildlife abundance and vigilance associated with air quality, sound, human activity, and habitat conditions at Caldwell Preserves?

Before you step into the field, you must master the standardized terminology and metrics that allow us to transform a walk in the woods into a high-quality scientific data set.


2. Decoding the Metrics: Defining Wildlife Abundance and Vigilance

Standardized metrics are the foundation of scientific integrity. By using the same codes and definitions, we can reliably compare data from Caldwell_Grove1 to other sites across the CULAP network. This consistency is what allows us to identify city-wide trends in animal behavior.

In your research, you will distinguish between Wildlife Abundance and Vigilance. It is important to note that your survey form asks for a "Total Animal Count." However, for our research, we convert this raw count into a population density metric by factoring in your observation duration and the specific area of your station. Vigilance is a survival metric; it represents the energy an animal "spends" staying alert to threats. Every minute an animal spends in a vigilant state is a minute lost for foraging or resting, which can significantly impact its health.

You will use the following Primary Behavior Codes to document your observations:

Behavior Code
Researcher Definition/Criteria
F
Foraging: Actively searching for or consuming food.
V
Vigilant: Alert, head up, scanning the environment for threats.
M
Moving: Traveling or changing location within the site.
R
Resting: Stationary, not engaged in active movement or feeding.
H
Human Alert: Vigilance specifically directed toward human presence (e.g., staring at a hiker).
FL
Fleeing: Rapidly leaving the area in response to a disturbance.
G
Grazing: Specifically feeding on grass or low-lying vegetation.
W
Drinking: Consuming water from local sources.

These definitions ensure that when you record an animal as "H," another researcher knows exactly what that behavior looked like in the field.


3. Field Snapshot: Initial Findings at Station Caldwell_Grove1

We often begin our research at Caldwell_Grove1. This is a strategic choice; as a picnic area near major trails, it serves as a high-interaction baseline to measure how wildlife responds to frequent human proximity.

In a recent session at this station, our researchers analyzed the following:

  • Location ID: Caldwell_Grove1 (Picnic Grove Area).
  • Species Count: White-tailed Deer were the dominant species, with a recorded count of 2 in a single group.
  • Habitat Description: While the station encompasses the picnic area, the deer were spotted in the adjacent Open Meadow, a habitat characterized by lower vegetation density compared to the wooded interior.
This snapshot suggests that larger mammals may utilize the meadow for foraging while remaining close enough to the grove for quick cover, a trend we will continue to monitor.


4. The "So What?" Layer: Correlating Behavior with Human and Environmental Disturbance

In urban ecology, we must account for "energy budgets." When a squirrel or deer is forced into a vigilant state by a passing cyclist, it stops eating. If this happens too often, the animal may suffer from chronic stress or malnutrition.

Based on previous observer notes at Caldwell_Grove1 and the North Branch Trail, we have identified several trends that we want you to test as hypotheses:

  • Cycling and Running: These high-speed activities on the North Branch Trail are frequently correlated with the Fleeing (FL) code, suggesting that rapid human movement is a high-stress trigger.
  • Picnic and Stationary Activity: Observations near the picnic tables often show higher counts of Human Alert (H) behavior. Animals may stay in the area for potential food scraps but remain highly vigilant of human proximity.
  • The "Meadow Refuge" Hypothesis: Wildlife in the Open Meadow appears to exhibit more Grazing (G) and Foraging (F) with lower vigilance scores, suggesting the meadow may act as a lower-stress refuge compared to the high-traffic trail corridors.
  • Additional Stressors: We are also tracking the presence of Dogs and Aircraft Overhead. As CULAP researchers, you must note if these auditory and predatory cues cause an immediate shift from resting to vigilance.
Disturbance vs. Behavior Hypotheses:

  • Cycling/Running: Predicted Response — Fleeing (FL).
  • Dogs Present: Predicted Response — High Vigilance (V) or Fleeing (FL).
  • Picnic Activity: Predicted Response — Human Alert (H).
  • Aircraft Noise: Predicted Response — Scanning/Vigilance (V).

5. Invisible Influences: Integrating Air Quality and Soundscape Data

Human activity isn't the only disturbance. Invisible factors like air quality and noise pollution are critical to our research. We use the EPA’s AQI (Air Quality Index) to categorize our research days.

According to the EPA Technical Document, an AQI of 101-150 is categorized as Orange (Unhealthy for Sensitive Groups). This is a critical threshold; particulate matter (PM) at these levels can "impair or halt" the built-in, self-cleaning mechanisms of animal respiratory tracts. To visualize this, we use the "Bird’s Eye View" methodology. By spreading petroleum jelly on "bird wings" (constructed from stockings and hangers), we can physically trap the microscopic particles that wildlife is forced to breathe. If the wings are dirty, it is a clear indicator that the animals' lungs are being exposed to pollutants that bypass their natural defenses.

Research Tip: When recording the "Average Noise Level (dB)," you must check the Bus or Traffic Noise box if vehicles are audible. This is particularly important at Caldwell_Grove1, which is located near road infrastructure. Chronic traffic noise can mask predator sounds, forcing wildlife into a state of permanent, exhausting vigilance.


6. Conclusion: Join the Research Mission at Caldwell Preserves

Your role as a citizen scientist is vital. By completing the Caldwell Preserves Weekly Wildlife Observation Survey, you provide the data Chicago needs to manage its urban green spaces effectively. Every GPS coordinate and behavior code you record helps us advocate for a healthier environment for both humans and animals.

Researcher’s Checklist for a Successful Field Visit

  1. Identity & Station: Enter your observer name and select the correct Station ID (e.g., Caldwell_Grove1).
  2. Precise Location: Record GPS Latitude and GPS Longitude as two separate entries to ensure data integrity.
  3. Atmospheric Data: Check the AQI Color Scale and record the Average Noise Level (dB), noting if aircraft or traffic are present.
  4. Habitat Assessment: Rate both Vegetation Density and your Visibility Score on a scale of 1 (Very dense/poor) to 5 (Very open/clear).
  5. Behavioral Analysis: Select your Focal Animal Species and record a Vigilance Score (0-5), where 0 represents no vigilance and 5 represents constant vigilance.

Your contributions today are the building blocks of tomorrow’s urban ecology. Step into the laboratory and help us protect the wild heart of Chicago.

Saturday, July 18, 2026

Catching the Invisible: How a "Bird’s Eye View" Connects Air Quality Data to Pre-Health Science Education

Hey educators and data enthusiasts! Welcome back to the blog. Today, we are diving into a fantastic curriculum transformation. We are taking the EPA's classic "Bird’s Eye View" Lesson Plan and showing you how we adapt it for the Chicago Urban Living Animals and Plants (CULAP) Data Camp.

For our Pre-Health Science students, learning biology isn't just about cell structures under a sterile microscope; it’s about understanding environmental health determinants right in their own backyards. Let’s look at how a simple field experiment becomes a powerful dataset for urban health analysis.

The Original Inspiration: A Bird’s Eye View

The original EPA lesson plan is geared toward grades K-5, using physical models to teach students about particulate matter (PM) and air pollution. Here’s the setup:

  • The Model: Students construct "birds" using wire coat-hangers and stockings.

  • The Collector: The wings are coated with petroleum jelly using cotton swabs.

  • The Experiment: These "birds" are hung outside for one full week to collect air samples.

  • The Analysis: A week later, students examine the wings with a magnifying glass to see the dust, pollen, and man-made particles trapped in the jelly.

It concludes with a critical discussion: If the air makes these simulated birds this dirty, what is it doing to our lungs?

The CULAP Data Camp Upgrade: From Observation to Urban Health Analytics

At CULAP, we take this core hands-on methodology and scale it up into an interdisciplinary data-science and pre-health experience for older students. In our version of the camp, we bridge the gap between field collection, clinical health implications, and digital mapping.

1. Standardized Spatial Sampling Across Chicago

Instead of hanging birds randomly around a school yard, CULAP students deploy their sticky collectors at standardized micro-locations across Cook County—spanning from high-traffic urban corridors near busy roadways to open green spaces like the Cook County Forest Preserves. This allows students to establish clear environmental variables (e.g., proximity to public transit lines vs. canopy cover).

2. Digitizing the Data

Instead of just saying "this wing looks dirtier than that one," our data campers digitize their samples:

  • Students use basic smartphone macro-photography to capture high-resolution grids of the trapped particulate matter.

  • They log their empirical observations into our custom data collection platform, tagging each sample with localized metadata (GPS coordinates, average daily temperature, humidity, and proximity to major highways).

3. The Pre-Health Connection: Respiratory Tract Dynamics

For our pre-health science track, the discussion moves from simple awareness to medical implications. The microscopic elements trapped—like animal dander, pollen, bacteria, and combustion soot—serve as a proxy for what bypasses the body's natural respiratory defenses. Students calculate estimated inhalation volumes and discuss how deep penetration of fine particulate matter (PM2.5) contributes to asthma flare-ups, chronic respiratory disease, and systemic inflammation in urban populations.

Bringing It All Together in the Classroom

If you want to replicate this hybrid field-and-data approach in your own program, here is the blueprint:

  1. Build & Deploy: Follow the classic EPA manual—grab your hangers, stretch the stockings, apply the petroleum jelly, and let them sit for 7 days.

  2. Quantify: Have students count the particles per square centimeter on different areas of the wing.

  3. Map It: Drop the counts into a shared spreadsheet or mapping tool. Correlate high-particle zones with local traffic patterns or industrial zoning.

  4. Prescribe Solutions: Conclude the unit by acting as public health advocates. What preventative measures or urban infrastructure designs (like expanding urban forestry) can protect community lungs?

By transforming a classic science experiment into a data-driven inquiry, we don't just teach students about the environment—we train the next generation of data-savvy medical and public health professionals to treat the systemic causes of urban disease.

Have you tried tracking air quality with your students? Drop your favorite field hacks in the comments below!

CULAP Data Center — Friday SportsCenter-Style Post-Game Report

For Friday, October 2, 2026 CULAP Data Center and Education Group | Grades 6–12 Feature: CULAPSE Darkaggriculture — Kaggriculture Season 0 W...