Using the latest GIS and AI technology, AHA’s independent assessment of deforestation risk is based on the USDA National Agricultural Statistics Service (NASS) Cropland Data Layer (CDL). The CDL is an annual geo-referenced, crop-specific land cover data layer produced using satellite imagery and extensive agricultural ground reference data. Annual review of the CDL data, which has a spatial resolution of 30 meters, allows the rate and immediate drivers of deforestation (in terms of the crops or other land uses that replace forests) to be readily identified.
The CDL has been updated annually since 2008 with new data released, usually in January each year, reflecting the crop and forest types occupying land during the previous growing season. The CDL draws on satellite data from the NASA/USGS Landsat Program and the European Space Agency’s Copernicus Sentinel-2 mission. This allows trends to be analysed over time, while removing the need for additional time-consuming and costly work by AHA to compile satellite data and classify land uses. The AHA deforestation-risk assessment for each year is updated as soon as possible after the CDL annual update becomes available.
To enhance the accuracy of the assessment, AHA is developing a procedure to bring expert human eyes to bear on any sites flagged by AI analysis of satellite data as potentially at risk of deforestation. Even the best algorithms can struggle to accurately distinguish between sites subject to sustainable timber harvesting operations; or that are damaged by pests, fires or windthrow; or that are undergoing conversion. In practice, some sites must be scrutinised by experts with knowledge of local forestry practices using higher-resolution satellite data, or even by putting feet on the ground. Over time, the results of this “truthing” exercise will be used to refine the AI analytical procedures.
To facilitate broad acceptance in the global market, the AHA assessment of deforestation risk is based, as far as possible, on internationally recognised definitions by deforestation by the United Nations Food and Agriculture Organisation (UN FAO). It also takes account of the definitions used in the EU Deforestation Regulation (EUDR). As the focus is on the risk of commodity-driven deforestation of hardwood stands, the analysis quantifies the extent of conversion of deciduous and mixed deciduous/coniferous forest to agricultural land.
The annual average percentages of forest area converted to another use can be broken down according to the classification of the newly converted land (e.g. specific agricultural crops, urban development etc). This allows insights into the drivers of deforestation and, specifically, to assess the risk of deforestation due to demand for different commodities in trade.
Drawing on the analysis and in consultation with stakeholders, AHA will determine a threshold level for deforestation. If deforestation falls below this level, the county level of risk will be considered negligible, requiring no further risk mitigation measures by U.S. hardwood operators.
AHA’s annual analysis of CDL data will be combined with an analysis of deforestation risks and drivers, prepared by the U.S. Forest Service, using the extensive sample-based inventory data collected across all the U.S., as part of the Forest Inventory and Analysis (FIA) Program. The U.S. Forest Service is currently implementing a strategic plan for enhanced integration of remote sensing information into the FIA Program, with special emphasis on so-called “small area estimation” (SAE) techniques. This effort will engage a broad range of stakeholders to assist the U.S. Forest Service in meeting a Congressional directive to “implement procedures to improve the statistical precision of estimates at the sub-State level”. As in AHA’s analysis of CDL data, the U.S. Forest Service is focused on improving the level of precision and access to data at county level, to support regulatory compliance and environmental claims in markets for forest products and agricultural commodities.
Sign up to the AHA newsletter to keep up to date