PASCAL (Pre-Annual Shoot Count Assessment Logic)

Intro

UK vineyards currently lack a reliable, cost-effective way to predict shoot counts early, creating uncertainty in yield forecasting, labour planning and profitability—especially for small to medium growers.

The Challenge

The Challenge

Manual shoot counting is slow, weather-dependent and inconsistent, yet shoot count is a key early proxy for yield. Forecast errors drive avoidable cost and operational risk.

The Solution

The Solution

PASCAL will test low-cost, field-deployable sensor clusters (microclimate + multispectral canopy reflectance) to identify early-season signals that correlate with shoot counts, enabling an initial predictive model and farmer-friendly insights through AI forecasting.

Image
PASCAL Vineyard

The Trial

The Trial

Two commercial vineyards (Dunesforde and Jojo’s) will deploy multiple sensor clusters per site from budburst through shoot emergence, alongside GPS-tagged manual counts using a shared protocol. Data uploads to the cloud and AI analysis will produce a paired dataset, practical insights, and a first-pass model for pre-season shoot count prediction and yield.

Website
Dunesforde Vineyard