Fig. 5. Trends of low-cloud fraction (Δ CF) for Tropical
Mountain Cloud Forests (TMCFs) between 1997 and 2020 among countries.
Grey points represent the average for each nation, while color points
each TMCFs according to their distribution among biogeographical realms.
Country acronyms represent the ISO 3166 country codes.
The altered balance of ecosystems resulting from reduced cloud cover can
also affect water supply, which not only impacts resident species but
also disrupts downstream water resources, thereby affecting human
settlements and industries. For example, the hydropower industry might
be affected as many dams rely on water recharge from TMCFs. As such, the
integration of ecosystem services that recognize TMCFs as economical
assets, beyond carbon sequestration, is crucial for future conservation.
Initiatives such as the Cloud Forest Blue Energy Mechanism(Narvaez et al., 2017) or Cloud Forest Bonds (Litovsky et al.,
2022) which consider the economic perspective of TMCFs may serve as
financial instruments to support the conservation and restoration of
these ecosystems while generating financial returns.
Protecting future TMCFs may depend on our ability to accurately observe
and project changes in cloudiness. Despite our findings document
declines in low-clouds on most of TMCFs, our results rely on the
accuracy of ER5’s low-cloud to observed clouds. A study by Dommo et al.
(2022) suggests that ERA5’s low-cloud product can capture the spatial
distribution of low-clouds across Western Central Africa compared with
other satellite products (e.g., MODIS). However, to our knowledge, no
studies have assessed temporal uncertainties of ERA5’s low-cloud.
Evaluations of diurnal cycles and long-term trends of ERA5’s total-cloud
product (i.e., the total atmospheric column) appears to have a coherent
variation with satellite imagery (Himawari-8) (Lei et al., 2020).
Likewise, temporal trends of surface temperature from ERA5 have shown to
be congruent as well with trends from meteorological stations (Yilmaz,
2023). If ERA5 low-clouds do not exhibit strong temporal biases, we
could imply that our estimated trends are valid. This could be
particularly true for observations from recent decades as ERA5 and its
improved estimates leverage in available remote sensing and
meteorological observations (Yilmaz, 2023). However, a temporal
assessment of ERA5’s low-cloud uncertainties require further
investigation. Therefore, future studies should utilize tangible cloud
immersion observations to assess temporal patterns such as time-lapse
photos or visibility data. These cloud immersion observations should
also evaluate the spatial and temporal uncertainties of cloud
observation products — such as ERA5 low-clouds — and explore their
reliability at a large scale. Unfortunately, due to limited availability
of local cloud observations in these remote ecosystems, it is unlikely
that such assessments will occur in the near future. Consequently, our
study emphasizes the need for the development of global networks for
cloudiness observation in TMCFs. These networks should be conducted in
partnership with countries, conservation agencies, and industries as the
observation and projection of clouds can have implications in several
sectors.