2017-11-24

Minister emphasizes necessity of unbiased LiDAR mapping



JAKARTA
(foresthints.news) - Following the inclusion of misleading data in the detailed analysis of the Norwegian-funded LiDAR mapping, the implementation of which was coordinated by the World Resources Institute (WRI), foresthints.news managed to get a response on this issue from Indonesian Environment and Forestry Minister Dr Siti Nurbaya.

The minister pointed out that four datasets of the detailed analysis derived from LiDAR mapping performed in four peat hydrological units (KHG), submitted by the Indonesian Peat Restoration Agency (BRG) in August this year, have not been used by the ministry whatsoever.

If it is later deemed necessary to use the four datasets from the LiDAR mapping, she continued, the ministry will still have to review them first, in particular with regard to the scope of the detailed analysis, its mapping methodology and the extent to which ground-based accuracy tests were carried out.

“Obviously I will study the datasets very closely before using the detailed analysis of the LiDAR mapping. To date, the datasets have not been used at all,” Minister Siti Nurbaya, an alumnus of the University of Twente’s Faculty of Geo-Information Science and Earth Observation (ITC) in the Netherlands, explained to foresthints.news (Nov 23) at the ministry building.

Below are photographs (Nov 16) which constitute evidence of the scale of 2015’s burned peat forests in the Kahayan peat hydrological landscape (2°20'17.44”S 114° 4'10.25”E) in Central Kalimantan’s Pulang Pisau regency. This evidence was omitted in the WRI-coordinated LiDAR mapping, a grave error in both a technical and legal sense.





In fact, as demonstrated in the photographs above, of the 336,000 hectares subject to LiDAR mapping in the Kahayan landscape, the Norwegian-funded LiDAR mapping disregarded 141,000 hectares of burned peat areas from 2015, a full 42% of the total LiDAR-mapped landscape.

Unbiased detailed analysis needed

As an expert in the spatial field, Minister Siti Nurbaya explained that LiDAR mapping is one of the most sophisticated mapping tools in existence today, so its level of erroneous measurements should be so small that it can be reliably used to boost government decision-making processes.

“Initially, our expectations were that the results of the LiDAR mapping would not deliver biased detailed analyses,” the minister emphasized.

While talking to foresthints.news, the minister repeatedly underlined that the ministry has not yet used the detailed analysis based on the LiDAR mapping in the four peat hydrological landscapes concerned.

The following photographs (Nov 16) confirm the high level of inaccuracy involved in the WRI-coordinated LiDAR mapping, in that major sections of 2015’s burned peat areas in the Kahayan peat hydrological landscape (2°20'17.44”S 114° 4'10.25”E) - the equivalent of more than 140,000 football fields - were simply ignored.





The high level of inaccuracy in the detailed analysis of the WRI-coordinated LiDAR mapping clearly implies a significant degree of bias, thus rendering the analysis irrelevant as a reference for government institutions, especially given that this bias has more than just technical implications.

Indeed, the extent of the bias involved in the WRI-coordinated LiDAR mapping also has significant legal ramifications, in view of the fact that it led to damaging mistakes in the classification of land cover associated with 2015’s burned peat areas.

The Environment and Forestry Minister cautioned that any mapping tool, whatever its name and including LiDAR, should still refer to land cover changes on a time-series basis while also encompassing a review of the government's legally-defined historical evidence.

The detailed analysis compiled from the WRI-coordinated LiDAR mapping which was submitted by the peat agency to the ministry, among other inaccuracies, failed to cover forestry concessions massively affected by 2015’s peat fires, thereby making the ministry’s decision not to use the mapping results completely sensible.