Evaluating the ability of agroforestry systems to mimic forests with remote sensing data in the Amazon

Agroforestry systems (AFSs) are a form of land use capable of sequestering and storing carbon in their perennial structure, while ensuring income distribution and food security in biodiversity-friendly landscapes. Given this, such a productive activity has great potential as an adaptive measure in the face of the ongoing climate change. The logic of this activity is supported by the theoretical framework of biomimetics, even if in an intuitive and non-declared way, where there is a deliberate intention that agroecosystems should imitate natural ecosystems in their structure and functioning. Based on this premise, the objective of this study is to evaluate whether the AFS in the municipality of Tomé-Açu, inserted in the Amazon biome, are capable of mimicking primary forests in terms of their structure and functioning over two decades of observation in a spatially explicit. To achieve this objective, we used the Enhanced Vegetation Index (EVI) calculated on the Google Earth Engine (GEE) platform for primary forests and agroecosystems in the municipality of Tomé-Açu between 2000 and 2020 annually. In addition, we used aboveground biomass (AGB) estimates derived from remote sensors to assess whether there were functional differences in carbon storage between the two evaluated treatments. We observed that using the EVI, there is no statistically significant difference between AFS and primary forests, although there are relevant differences in terms of biomass. Our results indicate that AFS can be structurally similar to mature forests, corroborating studies that point to a similarity in the ability to store carbon. From the results of our methodological approach, we conclude that agroforestry systems can represent a relevant opportunity for mitigating the socio-environmental effects of climate change.