Lessons Learned
59
BBOP Pilot Project Case Study – Ambatovy Project
11.1.2 Averaging
Averaging of attribute values was necessary in the habitat hectares calculation in order to avoid overly
conservative results and to reflect the reality of conditions on the ground. However, it is commonly recognised
that by calculating averages of vegetation parameters within a large area, habitat heterogeneity and its
variations, which are the key elements for microclimate and ecological processes, are masked (Whittaker
et
al.
1974; Parkes
et al.
2003). This approach thus limits the impact of variability on the calculation. While the
variation is small-scale, considering its effects is important.
11.2 Recommendations
1. Integrate a temporal parameter to the post-Project impact in the habitat hectares calculation in order to
show the Project’s success through time in reducing habitat hectares loss. This aspect is important for time
discounting (i.e. delay of offset creation and delivery of ultimate habitat hectares gain, see Step 7).
2. Averaging of contiguous
HABITAT TYPES
as far as habitat classes is optimal as it best reflects the reality of
conditions in the impact area.
3. Tools are based merely on numerical data that might not always have a biological and / or ecosystem
health significance. It is important to assess how to integrate biodiversity qualitative data (like
ENDEMISM
,
only presence / absence etc.) with the habitat hectares calculation.
4. There is a need for a ‘Weighting Guideline’ to guide the operator in weighting attributes, since this will
ultimately affect the scoring.
5.
SPECIES DIVERSITY
and other fauna attributes should be used (e.g., lemur and herpetofauna) as often as
possible to better reflect true biodiversity values (as opposed to forest structural
SURROGATES
).
6. The Project should design its data acquisition programs at the ESIA stage (baseline collection) to ensure it
meets the habitat hectares calculation requirements in an optimal fashion (e.g. fauna, flora, aquatics
sample sites linked with all main vegetation types likely to be impacted, as well as their condition classes –
for example, good / disturbed).
