Comunidades en DSpace
Seleccione una comunidad para explorar sus colecciones.
Envíos recientes
Impact of forest fire severity on soil physical and chemical properties in pine and scrub forests in high Andean zones of Peru
(Agriculture, 2024-12-27) Ocaña Zúñiga,Candy Lisbeth
Agroforestry systems can influence the occurrence and abundance of pests and diseases because integrating crops with trees or other vegetation can create diverse microclimates that may either enhance or inhibit their development. This study analyzes the severity of coffee rust in two agroforestry systems in the provinces of Jaén and San Ignacio in the department of Cajamarca (Peru). This research used a quantitative descriptive approach, and 319 photographs were collected with a professional camera during field trips. The photographs were segmented, classified and analyzed using the deep learning MobileNet and VGG16 transfer learning models with two methods for measuring rust severity from SENASA Peru and SENASICA Mexico. The results reported that grade 1 is the most prevalent rust severity according to the SENASA methodology (1 to 5% of the leaf affected) and SENASICA Mexico (0 to 2% of the leaf affected). Moreover, the proposed MobileNet model presented the best classification accuracy rate of 94% over 50 epochs. This research demonstrates the capacity of machine learning algorithms in disease diagnosis, which could be an alternative to help experts quantify the severity of coffee rust in coffee trees and broadens the field of research for future low-cost computational tools for disease recognition and classification
Current and Future Spatial Distribution of the Aedes aegypti in Peru Based on Topoclimatic Analysis and Climate Change Scenarios
(Insects, 2025-05-02) Ocaña Zúñiga,Candy Lisbeth
Dengue, a febrile disease that has caused epidemics and deaths in South America, especially Peru, is vectored by the Aedes aegypti mosquito. Despite the seriousness of dengue fever, and the expanding range of Ae. aegypti, future distributions of the vector and disease in the context of climate change have not yet been clearly determined. Expanding on previous findings, our study employed bioclimatic and topographic variables to model both the present and future distribution of the Ae. aegypti mosquito using the Maximum Entropy algorithm (MaxEnt). The results indicate that 10.23% (132,053.96 km2) and 23.65% (305,253.82 km2) of Peru’s surface area possess regions with high and moderate distribution probabilities, respectively, predominantly located in the departments of San Martín, Piura, Loreto, Lambayeque, Cajamarca, Amazonas, and Cusco. Moreover, based on projected future climate scenarios, it is anticipated that areas with a high probability of Ae. aegypti distribution will undergo expansion; specifically, the extent of these areas is estimated to increase by 4.47% and 2.99% by the years 2070 and 2100, respectively, under SSP2-4.5 in the HadGEM-GC31-LL model. Given the increasing dengue epidemic in Peru in recent years, our study seeks to identify tools for effectively addressing this pressing public health concern. Consequently, this research serves as a foundational framework for assessing areas with the highest likelihood of Ae. aegypti distribution in response to projected climate change in the second half of the 21st century.
Geospatial Landslide Risk Mapping Using AHP and GIS: A Case Study of the Utcubamba River Basin, Peru
(Applied Sciences, 2025-08-28) Ocaña Zúñiga,Candy Lisbeth
This study examines the use of a spatial multi-criteria approach based on GIS and AHP techniques to model landslide risk in the Utcubamba river basin, Peru. The methodology
consisted of selecting twelve triggering variables: slope angle, geology, precipitation, distance to faults, drainage density, TWI, relative relief, profile curve, land use, elevation,
distance to roads, and distance to population centers. These variables were then analyzed using the AHP method and then integrated into a GIS environment, where the weighted
linear combination (WLC) method was used to map landslide risk. The risk was categorized into five classes, ranging from very low (1) to very high (5). The main results indicate that
32.81% of the area analyzed in the Utcubamba river basin presents a high and very high risk of landslides. The high-risk areas are mainly located in the southern part of the basin and
coincide with areas with steep slopes, high rainfall, and proximity to population centers or communication routes. The model generated was highly accurate (AUC of 0.82), confirming
that the integration of the AHP method with GIS allows for the precise identification of critical areas, which is useful for territorial planning, the prioritization of interventions, and
emergency management, making it a reliable and replicable methodology in other parts of Peru.
Coffee Rust Severity Analysis in Agroforestry Systems Using Deep Learning in Peruvian Tropical Ecosystems
(Agriculture, 2024-12-27) Ocaña Zúñiga,Candy Lisbeth
Agroforestry systems can influence the occurrence and abundance of pests and diseases because integrating crops with trees or other vegetation can create diverse microclimates that may either enhance or inhibit their development. This study analyzes the severity of coffee rust in two agroforestry systems in the provinces of Jaén and San Ignacio in the department of Cajamarca (Peru). This research used a quantitative descriptive approach, and 319 photographs were collected with a professional camera during field trips. The photographs were segmented, classified and analyzed using the deep learning MobileNet and VGG16 transfer learning models with two methods for measuring rust severity from SENASA Peru and SENASICA Mexico. The results reported that grade 1 is the most prevalent rust severity according to the SENASA methodology (1 to 5% of the leaf affected) and SENASICA Mexico (0 to 2% of the leaf affected). Moreover, the proposed MobileNet model presented the best classification accuracy rate of 94% over 50 epochs. This research demonstrates the capacity of machine learning algorithms in disease diagnosis, which could be an alternative to help experts quantify the severity of coffee rust in coffee trees and broadens the field of research for future low-cost computational tools for disease recognition and classification
Non-Destructive Estimation Of Leaf Area In Cinchona Micrantha And Cinchona Pubescens Using Linear Regression Models
(International Information and Engineering Technology Association, 2025-07-25) Huaccha Castillo,Annick Estefany
Accurate quantification of leaf area is essential for ecophysiological, agronomic, and conservation studies, especially in threatened species such as Cinchona micrantha and Cinchona pubescens. This study evaluated simple, quadratic, and composite linear regression models to estimate leaf area non-destructively using morphometric measurements (length and width). A sample of n=800 leaves from 32 individuals was systematically collected and analyzed using a standardized photographic protocol with digital processing in ImageJ. The most robust models were those incorporating composite variables such as the product of length and width (L × W) and the sum squared of both dimensions ((L + W)²), reaching coefficients of determination higher than 0.97. These models consistently outperformed models based on single variables, providing higher accuracy and lower prediction error. High correlations were observed between leaf dimensions and area, and C. pubescens showed greater morphological variability. These findings establish that simple linear models based on L × W are efficient, replicable, and low-cost tools for non-destructive estimation of leaf area, which improves ecological monitoring and supports sustainable forest management, essential for the conservation of these Cinchona species, important from an ecological and medicinal point of view, in tropical ecosystems.