The high rate of depredation of forest resources causes major climatic changes that affect the economic activities and health of populations, plunging them into poverty and social problems. The state is responsible for dealing with these problems, because it has the technical, legal, and economic power to do so. This research aims to identify the factors that limit the design and formulation of public policies for the conservation of natural resources at the level of local district and provincial governments in Peru. For this study, we used qualitative methodology and non-probabilistic sampling, as well as techniques such as unstructured interviews, focus groups, and documentary review; moreover, for data analysis, we applied the theoretical saturation design in grounded theory. The results show that the conservation of natural resources is not positioned as a priority public policy in municipal administrations; furthermore, the organic units of natural resources suffer financial restrictions, and not because of a lack of budget availability, but because of a lack of will and decision-making capacity of authorities and civil servants, as well as a lack of coordination between the different national governments. It concludes that there are conflicts of interest in public policy making, with abuse of power and corruption predominating. Furthermore, the complexity of addressing sustainability criteria and the inability to confront the environmental crisis mean that international summits and national norms are attenuated in the face of the problems of environmental degradation. Deforestation should be considered a major public priority problem because of its environmental, economic, social, and health impacts. These problems require a holistic approach that combines local, national, and international policies and fosters effective and participatory governance.

Accurate identification of timber species is essential for sustainable forest management and the prevention of illegal logging, particularly in regions such as the Peruvian Amazon. This study evaluated explainable deep learning models for the classification of cedar (Cedrela montana) and cinchona (Cinchona pubescens) using macro-images of transverse wood sections collected in the Amazonas region of northern Peru. A dataset of wood images was expanded through a sliding-window patch extraction strategy and used to train and compare three architectures: a custom Convolutional Neural Network (CNN), MobileNetV2, and EfficientNetB0. Model performance was assessed using accuracy, precision, recall, and F1-score, while Gradient-weighted Class Activation Mapping (Grad-CAM) was applied to interpret the visual regions influencing predictions. All models achieved high classification performance. EfficientNetB0 obtained the best results, reaching 100% accuracy, precision, recall, and F1-score, while MobileNetV2 and the custom CNN achieved near-perfect performance. Training and validation curves indicated stable convergence without significant overfitting. Grad-CAM analysis showed that the custom CNN generated more localized and interpretable activation regions, whereas MobileNetV2 and EfficientNetB0 focused on broader textural patterns, revealing different feature-learning strategies. These findings demonstrate the potential of explainable deep learning for automated wood identification and support applications in forest monitoring, biodiversity conservation, and illegal logging control.

Cinchona spp. is widely distributed in cloud forest ecosystems in different countries in South America (Gómez-Silvera et al. 2016). Likewise, its distribution has expanded to numerous countries in ...