6
severe environmental impacts, necessitating targeted
mitigation efforts to conserve biodiversity. Kamoto mine
(Figure 6) uniquely shows a slight positive correlation with
mean NDVI which could be attributed to the significantly
less cobalt ore mining activity between 2000 and 2017,
hence a possible less impact on vegetation health (consider-
ing the aggregated impact of this time-series analysis).
Analysis of the estimated correlation between cobalt
production and the standard deviation of NDVI (R2) sug-
gests some evidence of variability in vegetation health in
relation to cobalt production. For example, at Etoile and
Mutanda mines, as production increases, vegetation health
or patterns become more variable and fragmented. Unlike
the mines at Etoile and Mutanda, Kasulo and Ruashi
exhibit little to no change in variability, suggesting uniform
impacts on vegetation across the mining landscape, as pre-
sented in Table 1.
Understanding these correlations is critical for cobalt
production with environmental conservation. By integrat-
ing NDVI insights into mineral resource extraction and
rehabilitation effort, companies can strategically mitigate
potential impacts induced by land use while maintaining
production volumes. For example, NDVI insights could
offer a cost-effective means to track the impact at Mutanda,
Etoile, Kasulo, and Ruashi mines where land use impact
on vegetation health is intense, helping mining compa-
nies identify areas needing intervention. Evidence shown
here suggests mine sites with high negative R1 correlations
should prioritize restoration efforts to conserve vegetation.
SUMMARY AND CONCLUSIONS
This study aimed to establish a framework for assess-
ing and quantifying land use impacts based on interdecadal
Earth observation data and mathematical algorithms that
leverage NDVI as a proxy for vegetation health by correlat-
ing it with cobalt production in the Democratic Republic
of Congo. By integrating satellite-derived environmental
indicators (NDVI) with ground-based production metrics
(cobalt ore production), we identified key spectral signa-
tures indicative of vegetation loss or negative impacts on
biodiversity induced by the intensity of cobalt ore produc-
tion. This approach offers a cost-effective means to track
vegetation health in comparison with existing land use
impact assessment methods that are often bundled within
impact assessments categories, which are frequently devel-
oped with characterization factors known to be liable to
high uncertainties. The study also highlighted inconsisten-
cies in current land use impact assessment methodologies
within life cycle assessment frameworks, such as differing
results among ReCiPe and IMPACT World+. This suggests
Figure 5. Time series comparison of cobalt production data for Mutanda mine with the mean and standard deviation of the
NDVI
severe environmental impacts, necessitating targeted
mitigation efforts to conserve biodiversity. Kamoto mine
(Figure 6) uniquely shows a slight positive correlation with
mean NDVI which could be attributed to the significantly
less cobalt ore mining activity between 2000 and 2017,
hence a possible less impact on vegetation health (consider-
ing the aggregated impact of this time-series analysis).
Analysis of the estimated correlation between cobalt
production and the standard deviation of NDVI (R2) sug-
gests some evidence of variability in vegetation health in
relation to cobalt production. For example, at Etoile and
Mutanda mines, as production increases, vegetation health
or patterns become more variable and fragmented. Unlike
the mines at Etoile and Mutanda, Kasulo and Ruashi
exhibit little to no change in variability, suggesting uniform
impacts on vegetation across the mining landscape, as pre-
sented in Table 1.
Understanding these correlations is critical for cobalt
production with environmental conservation. By integrat-
ing NDVI insights into mineral resource extraction and
rehabilitation effort, companies can strategically mitigate
potential impacts induced by land use while maintaining
production volumes. For example, NDVI insights could
offer a cost-effective means to track the impact at Mutanda,
Etoile, Kasulo, and Ruashi mines where land use impact
on vegetation health is intense, helping mining compa-
nies identify areas needing intervention. Evidence shown
here suggests mine sites with high negative R1 correlations
should prioritize restoration efforts to conserve vegetation.
SUMMARY AND CONCLUSIONS
This study aimed to establish a framework for assess-
ing and quantifying land use impacts based on interdecadal
Earth observation data and mathematical algorithms that
leverage NDVI as a proxy for vegetation health by correlat-
ing it with cobalt production in the Democratic Republic
of Congo. By integrating satellite-derived environmental
indicators (NDVI) with ground-based production metrics
(cobalt ore production), we identified key spectral signa-
tures indicative of vegetation loss or negative impacts on
biodiversity induced by the intensity of cobalt ore produc-
tion. This approach offers a cost-effective means to track
vegetation health in comparison with existing land use
impact assessment methods that are often bundled within
impact assessments categories, which are frequently devel-
oped with characterization factors known to be liable to
high uncertainties. The study also highlighted inconsisten-
cies in current land use impact assessment methodologies
within life cycle assessment frameworks, such as differing
results among ReCiPe and IMPACT World+. This suggests
Figure 5. Time series comparison of cobalt production data for Mutanda mine with the mean and standard deviation of the
NDVI