Real-time forest analytics

FAQ

What’s the biomass estimates methodology adopted by biodiversityX?

Bitterlich (1948) described a very rapid method of determining basal area per acre by point sampling or ACS. Imagine a forester using a specialized tool called a prism held at chest height. This tool acts like a window, revealing only a portion of the forest. By counting trees visible within the window and doing calculations, they estimate the basal area per acre. While efficient, it requires specialized tools and knowledge. biodiversityX builds on this, streamlining the process:

Point Selection: Define the forest boundary, and the app randomly selects unbiased sample points - no complex planning needed!

Simplified Capture: Forget the prism! Stand at a point, hold your phone chest-high, and take four photos (one in each direction).

Based on human-inputs, we take care of the math with AI-generated outputs:

Extracting Data: The app uses your phone's camera specs and photo details to determine field of view and photo width. It also leverages location data for tree height estimates from remote sensing.

Smart Counting: Using these details and a concept called Basal Area Factor, the app analyzes photos and counts trees exceeding a specific pixel threshold, contributing to the overall basal area. (Species classification for model calibration coming soon!)

Transparent Results: The app displays: average basal area per unit area (e.g., square feet per acre) based on counted trees and photo analysis, forest volume per unit area (e.g., cubic feet per acre) by multiplying average basal area by average tree height (from remote sensing), inputs used to inform the calculations.

biodiversityX is evolving fast and will release more outputs based on the ground inputs provided by users. There’s a very exciting roadmap ahead, don’t hesitate in sharing your contributions to improve the methodology and the needs you face so we can build what’s most useful. 

Are measurements accurate?

ACS is a credited methodology in the scientific community, recognized for its accuracy and scalability. We have just kicked off comparative analysis of biodiversityX outputs against traditional field plot sampling. results (password is groundfirst). We will make further findings publicly available as they come. We continue to look for testing partners who have extensive ground measurements to test accuracy in different geos, biomes, and settings.

Do you measure plots?

Rather than measuring entire plots, our approach utilizes point sampling. ACS, or plotless sampling, gives us the tools to determine forest parameters like basal area which lead to biomass estimates. This process often involves a prism for tallying trees at chosen points, with trees' inclusion likelihood tied to their basal area size. By leveraging phone camera technology, we recreate the prism's functionality, enabling accurate and widespread measurement capabilities without the need for traditional plots.

How do you handle diverse biomes, seasonal variations, and tree occlusion?

The Angle Count Sampling (ACS) method is designed for various biomes, and its effectiveness across different forests and seasons is to be tested. While ACS provides good estimates, we're investigating if additional calibration with local data and scientific input might be necessary. To address potential tree occlusion in dense forests, we're exploring solutions like increased sampling points or statistical adjustments. We welcome collaboration from the scientific community for data validation and further refinement.

How will we involve local communities?

We embrace inclusivity by catering to both experts and beginners alike, irrespective of their academic or professional backgrounds. This approach makes it significantly easier for local communities to engage with and benefit from our system by participating in data collection, and even earning from it. Utilizing the platform stretches across a broad spectrum of industries catered to businesses and organisations as our primary clientele. These enterprises have the freedom to designate data collection tasks either to their own teams or to local communities. Our goal is an active engagement of these communities in mapping and preserving their natural habitats.

What’s our approach to height?

biodiversityX developed a pixel-level height approach, meaning that the height is not extrapolated, but weighted based on area size for specific selected polygons.