tified, legally 
issions to be 
ffset raises a 
sinks at 1990 
ication of the 
exchange of 
mote Sensing 
mote sensing 
th, the Kyoto 
oup and the 
fichigan. The 
and academic 
col; 
1g technology 
with both the 
1g goal and a 
CO, and CH. 
s can be made 
'arbon budget. 
ms, especially 
hanges in land 
yoto Protocol 
prestation and 
  
  
  
Rosenqvist, A. 
2 OVERVIEW OF REMOTE SENSING TECHNOLOGY CAPABILITIES 
2.1 Applying Remote Sensing to the Kyoto Protocol 
A concern of the Kyoto Protocol is the imprecise definition of a "forest", which, in terms of ecosystem type, canopy 
cover, minimum area of interest etc., will have significant implications on the applicability of remote sensing technology to 
the treaty. The IPCC is currently examining the implications of different forest definitions for the Protocol and is also 
evaluating the merits of a more quantitative approach to land cover monitoring which would focus on carbon and biomass 
as a basic unit of measurement. While some of these issues will be addressed at the 6th Conference of the Parties (COP-6) 
in The Hague (NL) in November 2000, there is a need for the remote sensing community to provide a synopsis of what 
Earth observations can do relative to the land cover issues as they are stated now. In this context, five areas were identified 
where remote sensing technology may be applied, partly or fully, toward facilitating the treaty: 
e Provision of systematic observations of relevant land cover (Art. 5, Art. 10); 
Support of the establishment of a 1990 carbon stock baseline (Art. 3); 
Detection and quantification of change in forest area (Art. 3, Art. 12); 
Quantification of above-ground vegetation biomass stocks and associated changes therein (Art. 3, Art. 12); 
Mapping and monitoring of sources of anthropogenic CH, (Art. 3, Art. 5, Art. 10); 
2.1.1 Provision of systematic observations of relevant land cover 
Article 5:1 of the Kyoto Protocol states that " Each Party included in Annex I shall have in place, no later than one 
year prior to the start of the first commitment period, a national system for the estimation of anthropogenic emissions by 
sources and removals by sinks of all greenhouse gases not controlled by the Montreal Protocol." Article 10 (d), in turn, 
states that countries shall "Cooperate in scientific and technical research and promote the maintenance and the development 
of systematic observation systems and development of data archives to reduce uncertainties related to the climate system, 
[and] the adverse impacts of climate change...". 
Providing systematic, repetitive observations of large areas is potentially one of the strengths of remote sensing 
technology, and one where it can provide substantial support to the protocol on a long-term basis. Remote sensing data are 
however generally not acquired in a systematic manner, except locally over specific study sites and regional scale analysis 
of archived data are often complicated by variations in seasonality, sensor characteristics, viewing geometry etc., which 
introduce biases and uncertainties in the interpretation of the results. This is typically valid for most operational sensors, 
both optical and microwave, thereby undermining the usefulness of the data. It is recognized that dedicated and systematic 
acquisition strategies, focusing on obtaining regional coverage on a repetitive basis, would significantly improve the 
usefulness remote sensing data, not only in the context of the Kyoto Protocol, but also in a broader scientific framework. 
It is recognized that a federated approach having common goals and thematic definitions will be required to effectively 
support the Kyoto Protocol. Such an effort is currently underway within the framework of the Global Observations of 
Forest Cover (GOFC) Pilot Project (Ahern et al. 1998), under the auspices of CEOS. 
[Passive] Optical (Multi-spectral and Panchromatic) Systems 
Spaceborne optical systems have been in operation since 1972 and thematic mapping applications are generally past 
their initial research stages. However, while results for numerous land cover mapping applications have been presented over 
the years, they are often site specific and focused on a particular science objective. The feasibility of identifying the 
thematic classes directly applicable to the Kyoto Protocol, as it exists now, remains to be confirmed, and in some cases, 
further investigated. 
While panchromatic systems are of limited use for thematic mapping of vegetation, multi-spectral systems, in 
particular sensors which include mid-infrared bands such as Landsat TM, ETM+ and SPOT HRVIR, are well suited for this 
purpose. High resolution data will be required for the delineation of fragmented land cover and smaller patches of forest. 
Coarser resolution sensors such as NOAA AVHRR and VEGETATION are frequently used in combination with high 
resolution sensors for continental and global scale mapping (Mayaux et al. 1998, Richards et al. 2000). 
Currently available optical systems are generally capable of acquiring data at local, regional, and global scales and in a 
timely and regular manner. Cloud cover, smoke, and haze, however, put limitations on data availability, particularly in the 
tropical zone. The problem can be somewhat overcome with the coarse resolution sensors which have a higher temporal 
repeat-cycle, thereby enabling the creation of weekly or monthly mosaics compiled from cloud free pixels. 
Active Microwave Systems (SAR). M : 
The thematic mapping of vegetation using satellite based radar imaging systems has been less successful than optical 
systems. This is primarily due to the fact that the current suite of orbiting SAR systems all operate with a single-band and a 
single polarization. Classification accuracy with SAR, however, increases notably with the inclusion of additional bands 
  
  
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000. 1279