Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery
The leaf area index (LAI) provides information about the amount of photosynthetic area in relation to the total surface of an ecosystem and it is related to vital processes such as photosynthesis, respiration, and productivity. Thus, it is important to have information about the spatial distribution...
| Päätekijät: | , , , |
|---|---|
| Aineistotyyppi: | Online |
| Kieli: | spa |
| Julkaistu: |
Instituto de Ecología, A.C.
2018
|
| Linkit: | https://myb.ojs.inecol.mx/index.php/myb/article/view/e2431666 |
| _version_ | 1799769350166544384 |
|---|---|
| author | Nafarrate-Hecht, Ana Cristina Dupuy-Rada, Juan Manuel George-Chacon, Stephanie P. Hernández-Stefanoni, José Luis |
| author_facet | Nafarrate-Hecht, Ana Cristina Dupuy-Rada, Juan Manuel George-Chacon, Stephanie P. Hernández-Stefanoni, José Luis |
| author_sort | Nafarrate-Hecht, Ana Cristina |
| collection | MYB |
| description | The leaf area index (LAI) provides information about the amount of photosynthetic area in relation to the total surface of an ecosystem and it is related to vital processes such as photosynthesis, respiration, and productivity. Thus, it is important to have information about the spatial distribution of LAI at the landscape level. One of the most used methods for estimating LAI from satellite images is to associate it with spectral characteristics of the image and vegetation indices. However, these indices have a strong limitation due to saturation problems, which reduces the possibility of generating accurate LAI maps, particularly in forests with high levels of biomass. Here, we obtained regression models to map LAI in a tropical dry forest of Yucatan, during the rainy and dry seasons from high resolution satellite imagery. We used regression analysis combined with kriging, as this procedure integrates the relationship between LAI and both spectral and texture information of the imagery, as well as the spatial dependence of the observations. LAI values were obtained in the field using hemispheric photographs. The results show that LAI values differ significantly between seasons, with mean values of 3.37 in the rainy season and 2.49 in the dry season. The R2adj values of the regression analysis were 0.58 and 0.63 for the rainy and dry season respectively. Overall, our results demonstrate that by using texture measures, we are able to obtain accurate estimations of LAI in tropical dry forests with high levels of biomass. |
| format | Online |
| id | oai:oai.myb.ojs.inecol.mx:article-1666 |
| institution | Madera y Bosques |
| language | spa |
| publishDate | 2018 |
| publisher | Instituto de Ecología, A.C. |
| record_format | ojs |
| spelling | oai:oai.myb.ojs.inecol.mx:article-16662022-11-29T22:54:43Z Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery Modelización y mapeo estacional del índice de área foliar en un bosque tropical seco usando imágenes de satélite de alta resolución Nafarrate-Hecht, Ana Cristina Dupuy-Rada, Juan Manuel George-Chacon, Stephanie P. Hernández-Stefanoni, José Luis spectral data dry season vegetation indices rain texture metrics kriging regression datos espectrales estiaje índices de vegetación lluvias métricas de textura regresión con kriging The leaf area index (LAI) provides information about the amount of photosynthetic area in relation to the total surface of an ecosystem and it is related to vital processes such as photosynthesis, respiration, and productivity. Thus, it is important to have information about the spatial distribution of LAI at the landscape level. One of the most used methods for estimating LAI from satellite images is to associate it with spectral characteristics of the image and vegetation indices. However, these indices have a strong limitation due to saturation problems, which reduces the possibility of generating accurate LAI maps, particularly in forests with high levels of biomass. Here, we obtained regression models to map LAI in a tropical dry forest of Yucatan, during the rainy and dry seasons from high resolution satellite imagery. We used regression analysis combined with kriging, as this procedure integrates the relationship between LAI and both spectral and texture information of the imagery, as well as the spatial dependence of the observations. LAI values were obtained in the field using hemispheric photographs. The results show that LAI values differ significantly between seasons, with mean values of 3.37 in the rainy season and 2.49 in the dry season. The R2adj values of the regression analysis were 0.58 and 0.63 for the rainy and dry season respectively. Overall, our results demonstrate that by using texture measures, we are able to obtain accurate estimations of LAI in tropical dry forests with high levels of biomass. El índice de área foliar (IAF) proporciona información acerca de la cantidad de superficie fotosintética que existe en relación con la superficie total del ecosistema y se relaciona con procesos vitales como la fotosíntesis, la respiración y la productividad. Por lo tanto, es importante contar con información sobre la distribución espacial del IAF a escala de paisaje. El método indirecto más utilizado para la estimación del IAF se basa en imágenes de satélite y consiste en asociarlo con características espectrales e índices de vegetación. Sin embargo, estos índices tienen una fuerte limitación debido a problemas de saturación, lo cual restringe la posibilidad de generar mapas precisos de IAF, particularmente en bosques con altos niveles de biomasa. En el presente trabajo se obtuvieron modelos para mapear el IAF en un bosque tropical seco de Yucatán durante las estaciones de lluvia y estiaje a partir de imágenes de alta resolución, utilizando un procedimiento de regresión combinado con kriging. Este procedimiento integra la relación del IAF, tanto con datos espectrales y de textura de las imágenes, como con la dependencia espacial de los residuales. Se obtuvieron valores de IAF por medio de fotografías hemisféricas con una precisión aceptable y valores medios significativamente diferentes entre la temporada de lluvias (3.37) y la de estiaje (2.49). Los valores de R2aj de los modelos de regresión múltiple fueron de 0.58 y 0.63 para la temporada de lluvias y estiaje, respectivamente. En general, los resultados demuestran que, al utilizar el análisis de textura, se pueden generar modelos aceptables para la estimación del IAF en bosques tropicales secos con altos niveles de biomasa. Instituto de Ecología, A.C. 2018-10-11 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Artículo evaluado por pares application/pdf application/xml https://myb.ojs.inecol.mx/index.php/myb/article/view/e2431666 10.21829/myb.2018.2431666 Madera y Bosques; Vol. 24 No. 3 (2018): Otoño 2018 Madera y Bosques; Vol. 24 Núm. 3 (2018): Otoño 2018 2448-7597 1405-0471 spa https://myb.ojs.inecol.mx/index.php/myb/article/view/e2431666/1806 https://myb.ojs.inecol.mx/index.php/myb/article/view/e2431666/1898 Derechos de autor 2018 Madera y Bosques |
| spellingShingle | Nafarrate-Hecht, Ana Cristina Dupuy-Rada, Juan Manuel George-Chacon, Stephanie P. Hernández-Stefanoni, José Luis Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title | Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title_full | Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title_fullStr | Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title_full_unstemmed | Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title_short | Modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| title_sort | modeling of seasonal leaf area index values in a tropical dry forest using high resolution satellite imagery |
| url | https://myb.ojs.inecol.mx/index.php/myb/article/view/e2431666 |
| work_keys_str_mv | AT nafarratehechtanacristina modelingofseasonalleafareaindexvaluesinatropicaldryforestusinghighresolutionsatelliteimagery AT dupuyradajuanmanuel modelingofseasonalleafareaindexvaluesinatropicaldryforestusinghighresolutionsatelliteimagery AT georgechaconstephaniep modelingofseasonalleafareaindexvaluesinatropicaldryforestusinghighresolutionsatelliteimagery AT hernandezstefanonijoseluis modelingofseasonalleafareaindexvaluesinatropicaldryforestusinghighresolutionsatelliteimagery AT nafarratehechtanacristina modelizacionymapeoestacionaldelindicedeareafoliarenunbosquetropicalsecousandoimagenesdesatelitedealtaresolucion AT dupuyradajuanmanuel modelizacionymapeoestacionaldelindicedeareafoliarenunbosquetropicalsecousandoimagenesdesatelitedealtaresolucion AT georgechaconstephaniep modelizacionymapeoestacionaldelindicedeareafoliarenunbosquetropicalsecousandoimagenesdesatelitedealtaresolucion AT hernandezstefanonijoseluis modelizacionymapeoestacionaldelindicedeareafoliarenunbosquetropicalsecousandoimagenesdesatelitedealtaresolucion |
