Posted by: on

Technical REDD+ workshop organized jointly by FCPF and FAO, 3-4 April.

Accurate and consistent information on forest area and forest area change is a requirement if a country is to access results-based payments for its REDD+ activities. The goal is to obtain as accurate area estimates as possible but what is ‘as accurate as possible’?

‘What is the level of precision a country can achieve when estimating its area of forest and how the area is increasing or decreasing over time’?

These questions can help clarifying the complex issue of area estimation but do not have straight-forward answers. The accuracy depends on many factors which experts from FAO, FCPF, GFOI, USFS, JRC of EC, WWF, academia and the governments of Norway and Germany discussed at FAO Headquarters in April.

This technical exchange lasted for two days and was an opportunity to synthesize experiences from support provided to developing countries on accuracy of activity data in the REDD+ context.

At this point in time the stock-taking helped explore the bottlenecks so far and outlined actions needed to improve as countries are advancing further.  It was shown that the publication Good practices for estimating area and assessing accuracy of land change by Prof. Pontus Olofsson, as well as FAO’s technical guide released last year, Map Accuracy Assessment and Area Estimation: A Practical Guide, with the contribution by UN-REDD Programme experts, assisted countries and stakeholders in practice. Both of these resources have been serving as references and have facilitated the provision of streamlined support to countries while taking into account each country’s specific context. A focused discussion ensued on the application of these guidance tools and how to deal with the gaps identified since they were released.area estimae publ

Along with countries’ progress, new challenges have emerged. Different UN-REDD Programme and FCPF partner country examples were reviewed and indicated that the forest area estimates in their FREL submissions in 2016 and 2017 — both to UNFCCC and FCPF — had a quite high levels of uncertainty. Having large confidence intervals may make payments for results quite challenging for donors. 

While the FAO Practical Guide has assisted in methodology developments including hands-on implementation of area estimation of land cover as well as change — the experts identified the necessity for further technical guidance on issues addressing sampling design and inference, response design and categorization, and the steps to be taken after the accuracy assessment.

The aim is to ensure that countries have the information and tools needed to make the most precise estimates possible and be aware of implications of imprecise estimates as well as implications for monitoring. FAO’s collaboration with FCPF and GFOI has led to the catalysation of expert gatherings in order to guide countries in their forest area estimation.

The GFOI GOFC-GOLD (2]  organized a follow-up expert meeting in Norway in June with the goal of developing concrete country examples and experiences into more specific guidance. A detailed guidance documents is scheduled for publication in October.

 

Area accuracy

Based on high resolution data and the FAO Open Foris CE tool, this is an example of a sampling exercise on assessment of forest area using different sampling points at different points in time.

 

For more information, please contact Inge Jonckheere, Forestry Officer, Remote Sensing Lead in REDD+ Team, FAO  (E-mail: )

  

Useful resources:

Map Accuracy Assessment and Area Estimation: A Practical Guide (Accessible here)

Olofsson, P., Foody, G. M., Herold, M., Stehman, S. V., Woodcock, C. E., Wulder, M. A. 2014. Good practices for estimating area and assessing accuracy of land change. Remote Sensing of Environment, 148:42–57. (Accessible here)

 

Available software applications that can be freely used include QGIS, R, SEPAL and Open Foris (Geospatial Toolkit and Collect Earth). See more information here. There is a specific application on SEPAL for the stratified sampling for area estimation.

 

 


 

Tags:
Last modified by: on