Rosenqvist, A. 
  
estimation, since the signals saturate at low biomass levels. Although C-band sensitivity to biomass up to almost 100 t/ha in 
certain circumstances may be achieved by dual-pass interferometric technique (Askne et al. 1997, Santoro et al. 1999), the 
accuracy is largely dependent on factors such as the baseline distance and surface conditions, and more R&D is required 
before the technique may become operational with single band data. s 
L-band SAR, with a biomass saturation level of 60-100 tons/ha (Dobson 1992, Imhoff 1995), may be useful for coarse 
biomass estimates in regeneration areas (A and R components). There are currently no L-band SAR systems in orbit 
(JERS-1 failed 1998), but a polarimetric L-band system is planned for the ALOS satellite (due for launch in 2002) and 
could be well suited to address biomass issues in the context of the Kyoto Protocol. Interferometric coherence by L-band 
SAR is yet to be investigated. 
While the biomass levels approachable by L-band SAR are still way below those of mature forests, which vary 
between 100 - 600 t/ha, longer wavelengths, together with polarimetric and/or interferometric techniques, can be used to 
push the biomass saturation levels forward and to improve accuracy (Dobson et al. 1992, Imhoff 1995). Aircraft based radar 
sensors having full multi-band, polarimetric, and interferometric capabilities currently exist and have proven capable of 
detecting biomass in a wide range of forests up to 200 tons/ha dry above ground biomass (Ranson et al. 1997), primarily 
due to the long wavelength P-band channel. 
No space-borne P-band SAR system has been launched up-to-date, mainly due to unresolved ionospheric effects 
associated with low frequency radars. These effects, which are functions of the total electron content (TEC) in the 
ionosphere, result in deformation and polarimetric rotation of the signal (Ishimaru et al. 1999). It has however recently 
been shown that polarimetric (Faraday) rotation may be corrected for by using a fully polarimetric system (Freeman et dl. 
1998), while other ionospheric artefacts may be reduced to acceptable levels by accurate timing of the data acquisition 
(early dawn) when TEC is at minimum. Physical constraints in instrument design, relating to minimum antenna dimensions, 
can be by-passed by using a lower (50-100 m) ground resolution (Freeman ef al. 1999). 
Lower frequency SAR systems (VHF and UHF bands) may also prove very useful for biomass mapping. Recent 
studies have shown that these frequencies are capable of accurately measuring biomass above 100 tons/ha (dry above 
ground biomass). Results from the aircraft based CARABAS (20-90 MHz; Luanda et al. 1998), deployed in temperate and 
boreal forests in Europe, and the bios system (80-120 MHz, Imhoff et al. 2000), deployed in the neo-tropics, have shown 
that biomass measures can be accurately derived (within + 10% of field measures) for forests between 100 and 500 tons/ha 
(actual saturation levels for the bios and CARABAS systems have yet to be determined). These systems show great promise 
for local to regional applications using aircraft. However, the deployment of space-based VHF/UHF sensors may not be 
technically feasible due to ionospheric interference with the signal and has yet to be explored. 
The possibility of using UHF and VHF radar for routine forest biomass measurement is very real. Technological 
advancements are eliminating many of the obstacles that previously limited the development of UHF and VHF systems and 
orbital systems may be possible in the near future. In order to take advantage of the potential of these systems in the future, 
the scientific community needs to make the appropriate frequency allocation requirements known to the International 
Telecommunications Union (ITU) and the World Radio Conference (WCR) so that some part of the spectrum can be set 
aside for remote sensing purposes. 
Active Optical Systems (LIDAR 
Combined with allometric models (models linking biomass to measurable parameters such as tree height) the data 
collected by VCL should be capable of helping it make accurate measures of above ground biomass based on vegetation 
structure and canopy height measures. Combined with spatially extensive data, such as optical or SAR imagery, 
interpolation of biomass estimates between VCL sample points could be used to provide local, specific site, biomass 
estimates. As mentioned previously, it remains to be seen how such data will be applied over large areas. 
2.1.5 Mapping and monitoring of certain sources of anthropogenic CH, 
Article 3:1 of the Kyoto Protocol states that " The Parties included in Annex I shall, individually or jointly, ensure that 
their aggregate anthropogenic carbon dioxide equivalent emissions of the greenhouse gases listed in Annex A do not exceed 
their assigned amounts..." , "with a view to reducing their overall emissions of such gases by at least 5 per cent below 1990 
levels in the commitment period 2008 to 2012". Hence, the Kyoto Protocol relates to all six greenhouse gases listed in 
Annex A, including CH4, which is the second of the two greenhouse gases, apart from CO, considered relevant in the 
context of this report. : 
Although it may well be included in the paragraphs above, mapping and monitoring of certain sources of 
anthropogenic CHy is here listed separately, as it is not generally related to forestry or to forest change. Apart from livestock 
management - which is not considered feasible to monitor by remote sensing - CH, is also emitted as a result of anaerobic 
conditions in open water bodies following extended inundation. Typical anthropogenic sources of CH, include irrigated rice 
paddies, aquaculture (e.g. fish- and shrimp cultivation) and hydroelectric reservoirs. 
  
1282 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000.