Predictive models for assessing 'Spartina argentinensis' biomass

Vast areas of Argentina are covered by rangelands, which provide important ecosystem services. Traditionally, livestock farming has been the main productive activity in these environments, with the use of fire as a common management practice. Although burning stimulates the regrowth of grasses with...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Jozami, Emiliano, Di Leo, Nestor, Barbona, Ivana, Feldman, Susana
Formato: Artículo revista
Lenguaje:Español
Publicado: Facultad de Ciencias Agronómicas - UNR 2025
Materias:
Pastizales naturales
Estimación de biomasa
Drones
Rangelands
Biomass estimation
Pastagens
Estimativa de biomassa
Acceso en línea:https://cienciasagronomicas.unr.edu.ar/index.php/agro/article/view/98
Aporte de:
Ciencias Agronómicas de Universidad Nacional de Rosario
id I15-R223-article-98
record_format ojs
institution Universidad Nacional de Rosario
institution_str I-15
repository_str R-223
container_title_str Ciencias Agronómicas
language Español
format Artículo revista
topic Pastizales naturales
Estimación de biomasa
Drones
Rangelands
Biomass estimation
Drones
Pastagens
Estimativa de biomassa
Drones
spellingShingle Pastizales naturales
Estimación de biomasa
Drones
Rangelands
Biomass estimation
Drones
Pastagens
Estimativa de biomassa
Drones
Jozami, Emiliano
Di Leo, Nestor
Barbona, Ivana
Feldman, Susana
Predictive models for assessing 'Spartina argentinensis' biomass
topic_facet Pastizales naturales
Estimación de biomasa
Drones
Rangelands
Biomass estimation
Drones
Pastagens
Estimativa de biomassa
Drones
author Jozami, Emiliano
Di Leo, Nestor
Barbona, Ivana
Feldman, Susana
author_facet Jozami, Emiliano
Di Leo, Nestor
Barbona, Ivana
Feldman, Susana
author_sort Jozami, Emiliano
title Predictive models for assessing 'Spartina argentinensis' biomass
title_short Predictive models for assessing 'Spartina argentinensis' biomass
title_full Predictive models for assessing 'Spartina argentinensis' biomass
title_fullStr Predictive models for assessing 'Spartina argentinensis' biomass
title_full_unstemmed Predictive models for assessing 'Spartina argentinensis' biomass
title_sort predictive models for assessing 'spartina argentinensis' biomass
description Vast areas of Argentina are covered by rangelands, which provide important ecosystem services. Traditionally, livestock farming has been the main productive activity in these environments, with the use of fire as a common management practice. Although burning stimulates the regrowth of grasses with better forage quality, it also poses environmental challenges due to the emission of CO2eq into the atmosphere. In response to the growing demand for renewable energy sources, rangelands under livestock production systems with frequent burning present a sustainable alternative for bioenergy production. In the province of Santa Fe, rangelands dominated by Spartina argentinensis in the Bajos Submeridionales region, which covers more than two million hectares, have high potential for bioenergy production without compromising the existing biodiversity. This work focuses on the development and evaluation of predictive models for S. argentinensis biomass using spectral images obtained by drones. Multiple linear regression and classification models were developed, considering spectral variables, site, and seasons, to predict total biomass and its green and senescent fractions. The models obtained to estimate the total biomass of Spartina explained up to 62% of its variability. The green and senescent biomass fractions were predicted with greater accuracy, showing an R² of 66% for both. These findings highlight the potential of remote sensing technology to optimize the planning and sustainable management of biomass resources in Argentine grasslands.
publisher Facultad de Ciencias Agronómicas - UNR
publishDate 2025
url https://cienciasagronomicas.unr.edu.ar/index.php/agro/article/view/98
work_keys_str_mv AT jozamiemiliano predictivemodelsforassessingspartinaargentinensisbiomass
AT dileonestor predictivemodelsforassessingspartinaargentinensisbiomass
AT barbonaivana predictivemodelsforassessingspartinaargentinensisbiomass
AT feldmansusana predictivemodelsforassessingspartinaargentinensisbiomass
AT jozamiemiliano modelospredictivosparalaestimaciondebiomasadespartinaargentinensis
AT dileonestor modelospredictivosparalaestimaciondebiomasadespartinaargentinensis
AT barbonaivana modelospredictivosparalaestimaciondebiomasadespartinaargentinensis
AT feldmansusana modelospredictivosparalaestimaciondebiomasadespartinaargentinensis
AT jozamiemiliano modelospreditivosparaestimarabiomassadespartinaargentinensis
AT dileonestor modelospreditivosparaestimarabiomassadespartinaargentinensis
AT barbonaivana modelospreditivosparaestimarabiomassadespartinaargentinensis
AT feldmansusana modelospreditivosparaestimarabiomassadespartinaargentinensis
first_indexed 2025-09-04T05:07:18Z
last_indexed 2025-12-04T05:10:48Z
_version_ 1850553189789597696
spelling I15-R223-article-982025-11-17T13:05:56Z Predictive models for assessing 'Spartina argentinensis' biomass Modelos predictivos para la estimación de biomasa de 'Spartina argentinensis' Modelos preditivos para estimar a biomassa de 'Spartina argentinensis' Jozami, Emiliano Di Leo, Nestor Barbona, Ivana Feldman, Susana Pastizales naturales Estimación de biomasa Drones Rangelands Biomass estimation Drones Pastagens Estimativa de biomassa Drones Vast areas of Argentina are covered by rangelands, which provide important ecosystem services. Traditionally, livestock farming has been the main productive activity in these environments, with the use of fire as a common management practice. Although burning stimulates the regrowth of grasses with better forage quality, it also poses environmental challenges due to the emission of CO2eq into the atmosphere. In response to the growing demand for renewable energy sources, rangelands under livestock production systems with frequent burning present a sustainable alternative for bioenergy production. In the province of Santa Fe, rangelands dominated by Spartina argentinensis in the Bajos Submeridionales region, which covers more than two million hectares, have high potential for bioenergy production without compromising the existing biodiversity. This work focuses on the development and evaluation of predictive models for S. argentinensis biomass using spectral images obtained by drones. Multiple linear regression and classification models were developed, considering spectral variables, site, and seasons, to predict total biomass and its green and senescent fractions. The models obtained to estimate the total biomass of Spartina explained up to 62% of its variability. The green and senescent biomass fractions were predicted with greater accuracy, showing an R² of 66% for both. These findings highlight the potential of remote sensing technology to optimize the planning and sustainable management of biomass resources in Argentine grasslands. En Argentina, los pastizales naturales cubren vastas áreas, prestando importantes servicios ecosistémicos. Tradicionalmente, la ganadería ha sido la principal actividad productiva en estos ambientes, con el uso del fuego como práctica de manejo habitual. Aunque las quemas estimulan el rebrote de pastos con mejor calidad forrajera, también resultan en la emisión de CO2eq a la atmósfera, lo que plantea desafíos ambientales. Ante la creciente demanda de fuentes de energía renovable, los pastizales naturales, manejados en sistemas ganaderos con quemas frecuentes, se presentan como una alternativa sustentable para la producción de bioenergía. En la provincia de Santa Fe, los espartillares dominados por Spartina argentinensis en los Bajos Submeridionales, abarcando más de dos millones de hectáreas, tienen un elevado potencial para la producción de bioenergía sin comprometer la biodiversidad existente. Este trabajo se centra en el desarrollo y evaluación de modelos predictivos de biomasa de S. argentinensis utilizando imágenes espectrales obtenidas mediante drones. Se desarrollaron modelos de regresión lineal múltiple y de clasificación, considerando variables espectrales, sitio y estaciones del año, para predecir la biomasa total y sus fracciones verdes y senescentes. Los modelos obtenidos para estimar la biomasa total del espartillo permitieron explicar hasta el 62% de su variabilidad. Las fracciones de la biomasa verde y senescente pudieron ser predichas con mayor precisión, presentando un R2 de 66% para cada una de ellas. Estos hallazgos destacan el potencial de la tecnología de teledetección para optimizar la planificación y manejo sostenible de recursos biomásicos en los pastizales argentinos. Na Argentina, as pastagens naturais cobrem vastas áreas, prestando importantes serviços ecossistêmicos. Tradicionalmente, a pecuária tem sido a principal atividade produtiva nesses ambientes, com o uso do fogo como prática comum de manejo. Embora as queimadas estimulem o rebrote de gramíneas com melhor qualidade forrageira, elas também resultam na emissão de CO2eq na atmosfera, o que apresenta desafios ambientais. Diante da crescente demanda por fontes de energia renovável, as pastagens naturais, manejadas em sistemas pecuários com queimadas frequentes, apresentam-se como uma alternativa sustentável para a produção de bioenergia. Na província de Santa Fe, as pastagens de esparto dominadas por Spartina argentinensis nos Baixos Submeridionais, cobrindo mais de dois milhões de hectares, têm um elevado potencial para a produção de bioenergia sem comprometer a biodiversidade existente. Este trabalho foca no desenvolvimento e avaliação de modelos preditivos de biomassa de S. argentinensis utilizando imagens espectrais obtidas por drones. Foram desenvolvidos modelos de regressão linear múltipla e de classificação, considerando variáveis espectrais, locais e sazonais, para predizer a biomassa total e suas frações verdes e senescentes. Os modelos obtidos para estimar a biomassa total de Spartina permitiram explicar até 62% de sua variabilidade. As frações de biomassa verde e senescente puderam ser previstas com maior precisão, apresentando um R² de 66% para cada uma. Essas descobertas destacam o potencial da tecnologia de sensoriamento remoto para otimizar o planejamento e manejo sustentável dos recursos de biomassa nas pastagens argentinas. Facultad de Ciencias Agronómicas - UNR 2025-06-30 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artículo evaluado por pares text/html application/pdf https://cienciasagronomicas.unr.edu.ar/index.php/agro/article/view/98 10.35305/agro45.e047 Ciencias Agronómicas; Núm. 45 (25): 2025; e047 2250-8872 1853-4333 spa https://cienciasagronomicas.unr.edu.ar/index.php/agro/article/view/98/113 https://cienciasagronomicas.unr.edu.ar/index.php/agro/article/view/98/114 Derechos de autor 2025 Emiliano Jozami, Nestor Di Leo, Ivana Barbona, Susana Feldman https://creativecommons.org/licenses/by-nc-sa/4.0

Ejemplares similares