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.