Quantifying spatial distributions and temporal trends of livestock populations across pastoral agroecosystems at high resolution

CONTEXT

Global agroecosystems, predominantly pastoral and rangelands, are crucial for food, fibre, carbon sequestration, and biodiversity, covering about 40-50 % of Earth's land. Yet, they are also sources of greenhouse gas emissions, soil erosion, and land degradation. In New Zealand, nearly half the land supports grazing ruminants, significantly impacting the environment and contributing to methane and nitrous oxide emissions. Accurate livestock population data are essential to understand and mitigate these impacts.

OBJECTIVES

This study aims to create a high-resolution, farm-scale dataset of livestock densities in New Zealand, using a novel methodological framework that integrates agricultural production surveys, geospatial products and data derived from remote sensing. This framework is designed to improve the accuracy of environmental assessments, inform national and global livestock inventories, and guide sustainable land-use and conservation efforts.

METHODS

The study used a combination of geodatabases, data derived from remote sensing, regional statistics, and surveys to generate high-resolution gridded data products of livestock densities. Livestock counts were refined using land cover and farm-type data to exclude non-pastoral lands and accurately estimate grazing densities. The approach integrated data cleaning, processing, and spatial analysis within Python and qGIS, providing time-series analyses and regional validations against standard survey data to ensure accuracy and reliability. The outputs were further compared with global-scale livestock data to validate the accuracy and bias in global data being used to inform international modeling efforts.

RESULTS & CONCLUSIONS

We generated detailed maps showing spatial and temporal trends of sheep, beef, and dairy cattle across New Zealand. This high-resolution national dataset provides a more accurate national data source than previous global estimates for New Zealand and identifies biases/overestimations in global livestock data. The study also offers insights into the environmental pressures of livestock grazing, particularly regarding greenhouse gas emissions and soil erosion. The research presents a significant advance in our ability to quantify livestock populations at farm scale across national extents, providing a basis for more precise environmental and policy-making decisions. It underscores the need for high-resolution, locally validated data to inform global databases and supports targeted interventions to mitigate environmental impacts.

SIGNIFICANCE

The study's findings are crucial for managing agroecosystems sustainably, enhancing greenhouse gas inventories, and improving land and water quality management. They represent a step forward in our ability to quantify and manage the ecological footprint of livestock farming, with implications for policy and land management both nationally in New Zealand and potentially globally.