What do you think about the methodology used in FRA 2020 Remote Sensing Survey (FRA 2020 RSS) to estimate global and regional variables?
I strongly support the FRA 2020 RSS and consider it a technical milestone on global estimates, particularly for the capacity building efforts to the countries. It’s an efficient, sound and solid statically based approach and I value quite much the defined scope on global, regional and GZ results. I much like also the consistence between RSS results and FRA reported results
I wonder if efficiency would result in better index if centroid were represented by at least three unit of 1 ha instead of only 1 single ha. However, I understand the cost implications.
Do you agree on the variables reported by FRA 2020 RSS? Are there any variables you would like to propose to be included in the next FRA RSS Cycle in addition to those proposed on this background paper?
I have no suggestion on this issue.
Do you agree on building on top of the samples already collected by adding additional samples on changes detected between 2018 and 2023?
I agree with maintain sample as stable as possible with an updated stratification, in this regards the option of moving toward Horvitz-Thompson estimators to solve stratification issues seems to me as the best recommendation given:
This estimator is technically unquestionable
I think the current computational capacity make the enormous joint probability calculations feasible.
If I understood well there is a motion to reduce the total number of samples to 200,000 this would reduce calculation to 40x109
Were you involved in the FRA 2020 RSS data collection phase? If yes, do you think that the exercise contributed to the comprehension of the FRA categories (Forest, Other Wooded lands, other) used in the FRA country reporting process
No, I was not involved, although I was involved in the initial task force consultation for FRA RSS
Mr. Carlos Bahamondez
I strongly support the FRA 2020 RSS and consider it a technical milestone on global estimates, particularly for the capacity building efforts to the countries. It’s an efficient, sound and solid statically based approach and I value quite much the defined scope on global, regional and GZ results. I much like also the consistence between RSS results and FRA reported results
I wonder if efficiency would result in better index if centroid were represented by at least three unit of 1 ha instead of only 1 single ha. However, I understand the cost implications.
I have no suggestion on this issue.
I agree with maintain sample as stable as possible with an updated stratification, in this regards the option of moving toward Horvitz-Thompson estimators to solve stratification issues seems to me as the best recommendation given:
No, I was not involved, although I was involved in the initial task force consultation for FRA RSS