For Forest resource assessment, FAO , with the help of experts associated with, now a days,  use RS &GIS in most of the countries. However, the depth & quality depends on availability of  expertise for ground truth & analysis.

Please can you provide a specific example with link or reference? Does India report forest statistics for the FRA to FAO based on RS inventories?

Generally, FAO is not carrying out inventories for the FRAs. Instead FAO Forestry mails a questionnaire to all member states and compiles the completed and returned questionnaires. Member states decide individually how to obtain the information to complete its FRA questionnaire. Rwanda (see attached), Italy, Netherlands, for example, do not use RS for their FRA reporting.

Article A comparison of fine resolution census and image-based natio...

The USDA Forest Service Forest Inventory and Analysis (FIA) unit is responsible for the census of America's forests.  As I understand, they can update the inventory annually. Satellite imagery is used to enhance the plots and data. For more information on the FIA unit see Forest Inventory and Analysis National Program http://www.fia.fs.fed.us/.

I found this article by Erik et al. from FAO related to your question. But may be you already know about it. Here it is anyway

https://www.researchgate.net/publication/220473294_The_global_Landsat_imagery_database_for_the_FAO_FRA_remote_sensing_survey

I was wondering whether there is any similar 'independent assessment of forest cover and change, a global remote sensing survey (RSS)' organized like the FAO FRA 2010 for the 2015 FRA?

Article The global Landsat imagery database for the FAO FRA remote s...

Thanks Gyde and James for useful info; from your info it seems to me that RS is an ancillary tool in FRA; not the main, as we demonstrated to be feasible and effective for Rwanda and as another FAO branch produced with satellite imagery for Africa (Africover).

James, in our Rwanda article we refer to another global independent forest assessment (UNFCCC). For the countries I know best, the use of RS  imagery is also ancillary at best. But unlike the FRA, UNFCCC does independent peer review.

In the case of Mexico, the National Forest Commission (CONAFOR) is the agency responsible to prepare the forest statistics for each 5 year report to the FAO.  Those stastistics are prepared based on the Vegetation and Land Use spatial dataset that is produced by the National Institute of Statistics and Geography (INEGI), which is the National Statistics and Mapping agency for Mexico.  At the INEGI, we update the Vegetation map in several year cycles, using satellite imagery and field work.  The imagery is visually interpreted, and the polygon vectors for the Vegetation categories are updated. (about 60 vegetation types).  We have produced so far 5 versions (1985, 1993, 2000, 2007, 2011 as reference dates), and are now working in version 6 (with 2014 imagery).  We used aerial photographs for the 1st version (1985), and Landsat images for the others, except 2007, where we used SPOT multispectral imagery.

The dataset is delivered to the CONAFOR, and then they group the vegetation classes according to the FAO land cover classes and calculate the forest extent.

We are also working with the National Biodiversity Commision (CONABIO) and the CONAFOR, to develop Land Cover datasets using supervised classification of Landsat and Geoye images.  Once that is operational it could be used as the base for the National reports to the FAO.  We will then have two datasets: the Vegetation and Land use Map in a 3 - 5 year cycle, and a more frequent and more spatially detailed Land Cover (but with much less classes) hopefully in an annual basis.

Finally an example with RS imagery at the base of FRA and actual forest cover mapped rather than forestland/forest reserves (use). However, the visual interpretation and the transformation of forest types into FAO forest classes (varying over the years) are potential sources of uncertainties. 

Beyond the above, I am interested in the categories/class boundaries of the visual interpretation; as these need to be comparable over decades, the class boundaries are presumably prescribed (one of the LUCCs?).

Yes, there are some advantages and disadvantages with the visual interpretation.  On the positive side, we are able to identify and map vegetation types; not only for forests (temperate and tropical) but other plant formations as shrublands, grasslands, etc, as well as some succesional status characterization (primary, or secondary growth).  Some of the forest types are: pine, fir, pine-oak, oak-pine,  oak, cloud, juniper, etc (temperate), and among the tropical: high  evergreen tropical forest, medium semievergreen to low, deciduous (and some between them).  Basically the same classification system has been used through the whole versions.  Now, on the negative side: the subjective component in interpretation, and the different imagery sources; with varied characteristics and geometric properties. This makes hard to overlap precisely and so to compare different versions/dates . Another limitation is the source scale data, 1:250,000 it seems presently is not a detailed enough information.  On the other hand, the stats derived from each of the maps for the whole country seem consistent.  The LUCC is rather a vegetation classification system we use for Mexico, a crosswalk table is used to convert from the vegetation types to the categories in the FAO - LUCC.  The CONAFOR makes the conversions and area estimates, but they discuss with us the results before sending them to the FAO. As the responsible agency, they might consider some other issues in the final figures, but so far, the vegetation data sets from INEGI are the main base.  As mentioned in the previous post, eventually the land cover datasets obtained from the processing of Landsat - Geoeye imagery could be the source for the reports.  To have a more complete info, I can contact the CONAFOR.

Hi,

1. Forestry Department in Brunei Darussalam do not use RS as a mean to update their raport for FRA. They simply extrapolate the national forest inventory done 30+ years ago using a very 'optimistic' forest depletion model (some -0.2%yr-1).

2. Poland also do not report to FRA based on RS.

Best Regards,

Thanks, Kazimierz.

I agree, -0.2 % seems optimistic (see below).

What are the type of data used for FRA reporting in Poland?

regards

Article What Drives Conversion of Tropical Forest in Carrasco Provin...

Dear Hein,

The forest depletion (= deforestation + degradation) rate for Borneo is approx. -3%yr-1. This includes -2.2%yr-1 of deforestation. So, your Bolivian findings are consistent with Borneo data.

 We don't have illegal logging problem in Poland. Forests almost entirely belong to the state. The data for GFRA report are prepared based on data provided by the forest district officers and are reliable.

 Will you be interested to do a mutual project on forest dynamics? (kbecek"gmail.com)

 Here are my paper related to Borneo/Brunei.

 Best Regards,

Article Identification and assessment of factors affecting forest de...

Conference Paper Impact of Global Warming on Borneo Forest – the Worst Possib...

Somehow deforestation and spontaneous reforestation (attached Majella paper) rates at landscape level show a rather narrow range. Any idea why?

My experience on forest extent data provided by forest district officers is that these represent forest land (forest use; forest reserves) rather than forest cover (attached Rwanda paper). Also data from Indonesia often refer to forest land rather than forest cover in my experience. No problem, unless you compare between countries and over time within countries.   

Article Forecasting the pattern and pace of Fagus forest expansion i...

Article A comparison of fine resolution census and image-based natio...