TY - JOUR
T1 - Lidar-derived environmental drivers of epiphytic bryophyte biomass in tropical montane cloud forests
AU - Lai, Guan Yu
AU - Liu, Hung Chi
AU - Chung, Chih Hsin
AU - Wang, Chi Kuei
AU - Huang, Cho ying
N1 - Funding Information:
We appreciate the field assistance provided by colleagues and friends, especially Sheng-Fong Tsai and Zhang Jun, and Land Administration of Taiwan and Central Geological Survey of Taiwan for the airborne lidar data. We also thank the efforts of the editors Randolph Wynne and Jing. M. Chen, and the editorial team handling this manuscript. Comments from two anonymous reviewers greatly improved the quality and clarity of the work. This work was supported by the Ministry of Science and Technology (MOST) ( 105-2633-M-002-003- ), National Taiwan University (NTU) EcoNTU project ( 106R104516 ), and the NTU Research Center for Future Earth from the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education in Taiwan.
Publisher Copyright:
© 2020 Elsevier Inc.
PY - 2021/2
Y1 - 2021/2
N2 - Epiphytic bryophytes (EBs) are commonly found in tropical montane cloud forests (TMCFs), and they play significant roles in ecological functioning. Field sampling to assess the abundance of EB is challenging because of their “epiphytic” habitat, which makes large-scale quantifications impractical. The abundance of EBs is highly related to forest structure, physical environment and microclimate. These characteristics may permit landscape-scale assessments using a synoptic sensing approach. In this study, we investigated the relationship between the plot-scale EB biomass density (kg ha−1) and a comprehensive set of field and airborne light detection and ranging (lidar)-derived forest biophysical, topographic and bioclimatic attributes (factors), and assessed the feasibility of landscape-scale mapping of EB biomass in TMCFs. The study was carried out in 16,773 ha of TMCFs on Chilan Mountain in northeastern Taiwan. The relationship between EB biomass density data from 21 plots (30 × 30 m) and 39 field or 1-m gridded lidar data-derived forest structural, topographic and bioclimatic factors was investigated. We applied a partial least squares regression (PLSR) model to minimize the effects of multicollinearity among those 39 factors, and selected latent variables (LVs) explaining the majority of data variation for landscape-scale EB biomass mapping. The first four LVs explained 92% of the data variation, and the performance of the PLSR was satisfactory (R2 = 0.92, p < 0.001). The majority (35 out of 39) of the selected forest structural, topographic and bioclimatic factors were significantly related to one or more LVs, and most (37 out of 39) could be directly derived or were indirectly related to lidar metrics, thereby permitting the landscape-scale mapping of EB biomass density. We estimated that the mean (± standard deviation) EB biomass density was 296.5 ± 373.1 kg ha−1 and that the total EB biomass of the TMCF of Chilan Mountain was 4973.9 Mg. This study demonstrates that the proposed approach may be feasible for landscape-scale EB biomass mapping, thereby advancing our understanding of the role of EBs in the hydrological and nutrient cycles of TMCFs. The outcomes of the PLSR may elucidate the physiological mechanisms underpinning EB abundance in TMCFs and guide ecological management under future climate scenarios.
AB - Epiphytic bryophytes (EBs) are commonly found in tropical montane cloud forests (TMCFs), and they play significant roles in ecological functioning. Field sampling to assess the abundance of EB is challenging because of their “epiphytic” habitat, which makes large-scale quantifications impractical. The abundance of EBs is highly related to forest structure, physical environment and microclimate. These characteristics may permit landscape-scale assessments using a synoptic sensing approach. In this study, we investigated the relationship between the plot-scale EB biomass density (kg ha−1) and a comprehensive set of field and airborne light detection and ranging (lidar)-derived forest biophysical, topographic and bioclimatic attributes (factors), and assessed the feasibility of landscape-scale mapping of EB biomass in TMCFs. The study was carried out in 16,773 ha of TMCFs on Chilan Mountain in northeastern Taiwan. The relationship between EB biomass density data from 21 plots (30 × 30 m) and 39 field or 1-m gridded lidar data-derived forest structural, topographic and bioclimatic factors was investigated. We applied a partial least squares regression (PLSR) model to minimize the effects of multicollinearity among those 39 factors, and selected latent variables (LVs) explaining the majority of data variation for landscape-scale EB biomass mapping. The first four LVs explained 92% of the data variation, and the performance of the PLSR was satisfactory (R2 = 0.92, p < 0.001). The majority (35 out of 39) of the selected forest structural, topographic and bioclimatic factors were significantly related to one or more LVs, and most (37 out of 39) could be directly derived or were indirectly related to lidar metrics, thereby permitting the landscape-scale mapping of EB biomass density. We estimated that the mean (± standard deviation) EB biomass density was 296.5 ± 373.1 kg ha−1 and that the total EB biomass of the TMCF of Chilan Mountain was 4973.9 Mg. This study demonstrates that the proposed approach may be feasible for landscape-scale EB biomass mapping, thereby advancing our understanding of the role of EBs in the hydrological and nutrient cycles of TMCFs. The outcomes of the PLSR may elucidate the physiological mechanisms underpinning EB abundance in TMCFs and guide ecological management under future climate scenarios.
UR - http://www.scopus.com/inward/record.url?scp=85095734027&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85095734027&partnerID=8YFLogxK
U2 - 10.1016/j.rse.2020.112166
DO - 10.1016/j.rse.2020.112166
M3 - Article
AN - SCOPUS:85095734027
SN - 0034-4257
VL - 253
JO - Remote Sensing of Environment
JF - Remote Sensing of Environment
M1 - 112166
ER -