Uso de imagens de sensoriamento remoto para estratificação do cerrado em inventários florestais

• Autor(a) principal: Silva, Sérgio Teixeira da
• Data de Publicação: 2020
• Outros Autores: Mello, José Marcio de, Acerbi Junior, Fausto Weimar, Reis, Aliny Aparecida dos, Raimundo, Marcel Régis, Silva, Iasmim Louriene Gouveia, Scolforo, José Roberto Soares
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Institucional da UFLA
• Texto Completo: http://repositorio.ufla.br/jspui/handle/1/38621
• topic: Amostragem sistemática; Amostragem casual estratificada; Interpretação visual de imagens; Sensoriamento remoto; Inventários florestais; Systematic sampling; Stratified casual sampling; Visual interpretation of images; Remote sensing; Forest inventories

Remote sensing imagery can be a very useful auxiliary tool for native forests inventory. Thus, the objective of this study was to evaluate the stratification of a cerrado (Brazilian savanna) patch based on visual image interpretation techniques as well as to compare the errors from two sampling designs, the stratified random sampling (SRS) and the systematic sampling (SS).The study area corresponds to a cerrado sensu stricto patch located in the municipality of Papagaios, Minas Gerais State, Brazil. The cerrado wood volumes were obtained from a forest inventory field campaign where 32 plots were measured systematically. The study area was stratified based on a visual interpretation of a Landsat 5 TM image, and the strata formed were: “strata I”, “strata II”, “strata III”, water and riparian forests. There was a reduction of 43% on the inventory errors using the SS estimators compared to the inventory errors using the SRS estimators. We concluded that the stratification based on image interpretation techniques was efficient since there was a reduction on the cerrado inventory errors.