Learning from monitoring networks: Few-large vs. many-small plots and multi-scale analysis

被引:2
|
作者
Tang, Becky [1 ]
Kamakura, Renata P. [2 ]
Barnett, David T. [3 ]
Clark, James S. [2 ]
机构
[1] Middlebury Coll, Dept Math & Stat, Middlebury, VT 05753 USA
[2] Duke Univ, Nicholas Sch Environm, Durham, NC USA
[3] Battelle Mem Inst, Natl Ecol Observ Network, Boulder, CO USA
来源
基金
美国国家科学基金会; 美国国家航空航天局;
关键词
aggregation; clustering; Forest Inventory and Analysis (FIA); parameter uncertainty; prediction performance; sample design; sampling effort; SPECIES DISTRIBUTION MODELS; FOREST INVENTORY; CLIMATE-CHANGE; SPATIAL-RESOLUTION; DESIGN; ECOREGIONS; SIZE; LIFE; ERA;
D O I
10.3389/fevo.2023.1114569
中图分类号
Q14 [生态学(生物生态学)];
学科分类号
071012 ; 0713 ;
摘要
In order to learn about broad scale ecological patterns, data from large-scale surveys must allow us to either estimate the correlations between the environment and an outcome and/or accurately predict ecological patterns. An important part of data collection is the sampling effort used to collect observations, which we decompose into two quantities: the number of observations or plots (n) and the per-observation/plot effort (E; e.g., area per plot). If we want to understand the relationships between predictors and a response variable, then lower model parameter uncertainty is desirable. If the goal is to predict a response variable, then lower prediction error is preferable. We aim to learn if and when aggregating data can help attain these goals. We find that a small sample size coupled with large observation effort coupled (few large) can yield better predictions when compared to a large number of observations with low observation effort (many small). We also show that the combination of the two values (n and E), rather than one alone, has an impact on parameter uncertainty. In an application to Forest Inventory and Analysis (FIA) data, we model the tree density of selected species at various amounts of aggregation using linear regression in order to compare the findings from simulated data to real data. The application supports the theoretical findings that increasing observational effort through aggregation can lead to improved predictions, conditional on the thoughtful aggregation of the observational plots. In particular, aggregations over extremely large and variable covariate space may lead to poor prediction and high parameter uncertainty. Analyses of large-range data can improve with aggregation, with implications for both model evaluation and sampling design: testing model prediction accuracy without an underlying knowledge of the datasets and the scale at which predictor variables operate can obscure meaningful results.
引用
收藏
页数:12
相关论文
共 22 条
  • [1] Multi-scale kronecker-product relation networks for few-shot learning
    Abdelaziz, Mounir
    Zhang, Zuping
    MULTIMEDIA TOOLS AND APPLICATIONS, 2022, 81 (05) : 6703 - 6722
  • [2] Multi-scale kronecker-product relation networks for few-shot learning
    Mounir Abdelaziz
    Zuping Zhang
    Multimedia Tools and Applications, 2022, 81 : 6703 - 6722
  • [3] Dual-Branch Multi-Scale Relation Networks with Tutorial Learning for Few-Shot Learning
    Xu, Chuanyun
    Wang, Hang
    Zhang, Yang
    Zhou, Zheng
    Li, Gang
    APPLIED SCIENCES-BASEL, 2024, 14 (04):
  • [4] Large scale air monitoring: lichen vs. air particulate matter analysis
    Rossbach, M
    Jayasekera, R
    Kniewald, G
    Thang, NH
    SCIENCE OF THE TOTAL ENVIRONMENT, 1999, 232 (1-2) : 59 - 66
  • [5] Multi-scale image analysis for ecological monitoring of heterogeneous, small structured landscapes
    Blaschke, T
    Conradi, M
    Lang, S
    REMOTE SENSING FOR ENVIRONMENTAL MONITORING, GIS APPLICATIONS, AND GEOLOGY, 2002, 4545 : 35 - 44
  • [6] A multi-scale feedback ratio analysis of heartbeat interval series in healthy vs. cardiac patients
    Huo, Chengyu
    Huang, Xiaolin
    Ni, Huangjing
    Liu, Hongxing
    Bian, Chunhua
    Ning, Xinbao
    MEDICAL ENGINEERING & PHYSICS, 2014, 36 (12) : 1693 - 1698
  • [7] Multi-Scale Minero-Chemical Analysis of Biomass Ashes: A Key to Evaluating Their Dangers vs. Benefits
    Comodi, Paola
    Zucchini, Azzurra
    Susta, Umberto
    Cambi, Costanza
    Vivani, Riccardo
    Cavalaglio, Gianluca
    Cotana, Franco
    SUSTAINABILITY, 2021, 13 (11)
  • [8] Multi-scale Recurrence Quantification Analysis of Heartbeat Interval Series in Healthy vs. Heart Failure Subjects
    Huo, Chengyu
    Lu, Yi
    Huang, Xiaolin
    Liu, Hongxing
    Ning, Xinbao
    2014 7TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI 2014), 2014, : 347 - 352
  • [9] MULTI-SCALE EXPLAINABLE FEATURE LEARNING FOR PATHOLOGICAL IMAGE ANALYSIS USING CONVOLUTIONAL NEURAL NETWORKS
    Uehara, Kazuki
    Murakawa, Masahiro
    Nosato, Hirokazu
    Sakanashi, Hidenori
    2020 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2020, : 1931 - 1935
  • [10] Multi-GNSS PPP-RTK: From Large- to Small-Scale Networks
    Nadarajah, Nandakumaran
    Khodabandeh, Amir
    Wang, Kan
    Choudhury, Mazher
    Teunissen, Peter J. G.
    SENSORS, 2018, 18 (04)