Canopy clumping index (CI): A review of methods, characteristics, and applications

https://doi.org/10.1016/j.agrformet.2021.108374Get rights and content

Highlights

  • A review of CI field and remote sensing methods, characteristics, and applications.

  • The LX and CC methods are two standard methods in field CI estimation.

  • Remote sensing CI is mainly derived from the NDHD method and needs more validation.

  • CI shows vertical, scale, directional, and temporal variations.

  • CI is critical in the LAI estimation, canopy reflectance and land surface models.

Abstract

Canopy clumping index (CI) characterizes the spatial distribution of leaves or needles within a vegetation canopy. CI is critical in determining the canopy radiation transfer, photosynthesis, and hydrological processes. This paper reviews the field measurement and remote sensing estimation methods, characteristics, and applications of CI. CI is generally estimated in the field using direct, indirect optical, and allometric methods. The basis for the indirect optical approach is the gap-fraction based method. Remote sensing of CI is carried out using passive optical and active LiDAR technology. Current CI products are mainly derived from an empirical relationship with the normalized difference hotspot and darkspot (NDHD) index. Further CI product validation studies should be conducted using enhanced field measurements. CI typically shows vertical, scale, directional, and temporal variations. The overall CI is calculated as the integration of the directional CI values. CI is a key parameter in leaf area index (LAI) estimation, and canopy reflectance and land surface modeling studies. Future studies should focus on the adoption of automated and wireless measurement methods, development of new remote sensing estimation methods, improving our understanding of the CI characteristics, and accounting for CI in land surface models.

Introduction

The spatial distribution of leaves or needles within a vegetation canopy is critical in determining the canopy radiation transfer (Baldocchi and Harley, 1995; Carrer et al., 2013; Kucharik et al., 1999; Naudts et al., 2015; Nouvellon et al., 2000), photosynthesis (Law et al., 2001; Rambal et al., 2003), and hydrological processes (Chen et al., 2016; Lunka and Patil, 2016). Leaf spatial distribution can be quantified using a canopy clumping index (CI) defined as a ratio of the effective leaf area index (LAIe) to the leaf area index (LAI) (Chen, 2018; Fernandes et al., 2014; Nilson, 1971).ΩLAIe/LAIwhere Ω is the canopy CI. LAI quantifies the amount of live green leaf material present in the canopy per unit ground surface and is an essential climate variable identified by the Global Climate Observing System (GCOS) (2016). LAIe is the LAI value that would produce the same indirect gap measurement as that observed, assuming a simple random foliage distribution (Chen et al., 2005). When the foliage spatial distribution is random, Ω=1. If foliage elements are regularly distributed, Ω >1. When elements are clumped, Ω <1. In the following texts, CI and Ω are used interchangeably, in slightly more descriptive and arithmetic meanings, respectively.

Accounting for the clumping effect is essential for the correct estimation of LAI using the indirect optical approach.LAI=LAIe/Ω

Failure to account for CI would lead to an LAI error of up to 70% (Chen and Cihlar, 1995a; Stenberg, 1996; Woodgate et al., 2015). In land surface modeling, LAI is closely related to canopy photosynthesis, leaf respiration, and litterfall, whereas LAIe governs the light absorption and precipitation interception (Davi et al., 2008). In theory, Ω varies from zero to infinity in the extreme cases when all foliage elements are stacked on top of each other or laid side by side. Remote sensing studies have shown that the global CI values generally vary between 0.3 (very clumped canopies) and 1.0 (randomly distributed foliage elements) (Jiao et al., 2018; Wei et al., 2019).

In many earlier studies, a foliage dispersion parameter was frequently calculated from the contact number using the relative variance approach (Baldocchi and Harley, 1995; Myneni et al., 1989l; Nilson, 1971; Warren Wilson, 1965). The dispersion parameter was then used to calculate the canopy gap fraction with a negative binomial model. A major difference between the foliage dispersion parameter and the current CI is that the former quantifies the ratio of canopy gap fractions under real and random conditions, whereas the latter quantifies the ratio of LAIe and LAI (Eq. (1)). The definition of CI is also different from the aggregation metrics in landscape ecology and spatial statistics. In landscape ecology, the aggregation index and the clumpiness index are calculated as a measure of spatial aggregation for ecological adjacencies and class patches (John et al., 2009; McGarigal, 2015; Rasmussen et al., 2011). Although conceptually similar, the aggregation index and the clumpiness index provide quantitative metrics to quantify the spatial aggregation level of a landscape unit, whereas the CI, defined for the non-randomness of individual leaves or needles, is used to better quantify the transmittance and interception of light and precipitation. Similarly, spatial statistics provide tools to analyze the non-randomness patterns of spatial points (Perry et al., 2006; Szmyt, 2014). However, the spatial pattern metrics derived from statistical methods are not necessarily equal to the CI definition in Eq. (1) and cannot be applied in land surface models (LSMs) for radiation and hydrological modeling within the canopy.

This paper provides a comprehensive review of the canopy CI. Sections 2 and 3 present the CI field and remote sensing methods, respectively. Section 4 discusses the general characteristics of canopy CI. Sections 5 describes the application of CI in LAI estimation, canopy reflectance and land surface models. Section 6 concludes the paper.

Section snippets

Field methods

In field measurements, CI is obtained through direct or indirect methods (Table 1). Direct methods estimate CI by separately estimating LAIe and LAI (Eq. (1)). Indirect methods estimate the distribution of gap fractions or use an allometric relationship with other canopy biophysical variables, e.g., diameter at breast height (DBH) for tree canopies. Both direct and indirect methods are physically consistent with the CI definition (Eq. (1)). Other proxy aggregation parameters have also been

Passive optical methods

CI can be estimated from passive optical remote sensing data through an empirical relationship with vegetation indices or bidirectional reflectance shape indicators, e.g., the commonly used normalized difference hotspot and darkspot (NDHD) index (Chen et al., 2005; Leblanc et al., 2005b).Ω=a(θ)×NDHD+b(θ)where a and b are coefficients that vary with θ, and NDHD is defined asNDHD=ρhρdρh+ρdwhere ρh and ρd are the reflectance values at the hotspot and darkspot in the principal plane, respectively (

Characteristics of CI

It has been known since the 1970s that the CI shows scale, directional, and temporal variations (Nilson, 1971; O'Toole et al., 1979). This section synthesizes the CI variations in these dimensions.

LAI estimation

Because of the coupling relationship between CI, LAIe, and LAI, the most significant application of CI is for the transformation between LAIe and LAI (Eq. (2)). When the measured CI is not available, empirical values from similar vegetation types or from remote sensing products can be used. Pinty et al. (2006) suggested a second-order polynomial relationship between LAI and LAIe, indirectly accounting for the clumping effect. The vertical CI profile derived from the waveform LiDAR can even be

Conclusion

This paper provides a comprehensive review of the field measurement and remote sensing methods to determine the canopy CI and its general characteristics and applications in the LAI estimation, canopy reflectance and land surface models. Indirect optical methods are commonly used to determine the CI in field measurements. The two standard methods, the gap-fraction based (the LX method) and the gap-size based (the CC method) methods, are based on the general rationale to calculate the spatial

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the National Key Research and Development Program of China (2016YFA0600201). The author is thankful for personal communications with Prof. Jing M. Chen of the University of Toronto during the preparation of the manuscript. Prof. Dennis Baldocchi and another anonymous reviewer graciously provided constructive reviews of this work and are gratefully acknowledged. Mr. Yinghui Zhang helped prepare Fig. 1.

References (231)

  • J.M. Chen et al.

    Retrieving leaf area index for boreal conifer forests using Landsat TM images

    Remote Sens. Environ.

    (1996)
  • J.M. Chen

    Leaf area index measurements at Fluxnet-Canada forest sites

    Agric. For. Meteorol.

    (2006)
  • J.M. Chen et al.

    Multi-angular optical remote sensing for assessing vegetation structure and carbon absorption

    Remote Sens. Environ.

    (2003)
  • J.M. Chen et al.

    Global mapping of foliage clumping index using multi-angular satellite data

    Remote Sens. Environ.

    (2005)
  • Q. Chen et al.

    Modeling radiation and photosynthesis of a heterogeneous savanna woodland landscape with a hierarchy of model complexities

    Agric. For. Meteorol.

    (2008)
  • F. Chianucci et al.

    Estimation of leaf area index in understory deciduous trees using digital photography

    Agric. For. Meteorol.

    (2014)
  • F. Chianucci

    Estimation of canopy attributes in beech forests using true colour digital images from a small fixed-wing UAV

    Int. J. Appl. Earth Obs. Geoinf.

    (2016)
  • S. Cohen et al.

    The distribution of leaf area, radiation, photosynthesis and transpiration in a Shamouti orange hedgerow orchard. Part I. Leaf area and radiation

    Agric. For. Meteorol.

    (1987)
  • A. Cutini et al.

    Estimation of leaf area index with the Li-Cor LAI 2000 in deciduous forests

    Forest Ecol. Manag.

    (1998)
  • F.M. Danson

    Developing a dual-wavelength full-waveform terrestrial laser scanner to characterize forest canopy structure

    Agric. For. Meteorol.

    (2014)
  • H. Davi et al.

    Effect of thinning on LAI variance in heterogeneous forests

    Forest Ecol. Manag.

    (2008)
  • A. Deguchi et al.

    The influence of seasonal changes in canopy structure on interception loss: Application of the revised Gash model

    J. Hydrol.

    (2006)
  • V. Demarez et al.

    Estimation of leaf area and clumping indexes of crops with hemispherical photographs

    Agric. For. Meteorol.

    (2008)
  • S. Duthoit et al.

    Assessing the effects of the clumping phenomenon on BRDF of a maize crop based on 3D numerical scenes using DART model

    Agric. For. Meteorol.

    (2008)
  • H. Fang et al.

    Seasonal variation of leaf area index (LAI) over paddy rice fields in NE China: Intercomparison of destructive sampling, LAI-2200, digital hemispherical photography (DHP), and AccuPAR methods

    Agric. For. Meteorol.

    (2014)
  • H. Fang et al.

    Retrieving leaf area index using a genetic algorithm with a canopy radiative transfer model

    Remote Sens. Environ.

    (2003)
  • H. Fang et al.

    Estimation of the directional and whole apparent clumping index (ACI) from indirect optical measurements

    ISPRS J. Photogramm. Remote Sens.

    (2018)
  • H. Fang et al.

    Continuous estimation of canopy leaf area index (LAI) and clumping index over broadleaf crop fields: An investigation of the PASTIS-57 instrument and smartphone applications

    Agric. For. Meteorol.

    (2018)
  • F.J. García-Haro

    Derivation of global vegetation biophysical parameters from EUMETSAT Polar System

    ISPRS J. Photogramm. Remote Sens.

    (2018)
  • M. García

    Canopy clumping appraisal using terrestrial and airborne laser scanning

    Remote Sens. Environ.

    (2015)
  • S. Garrigues

    Intercomparison and sensitivity analysis of leaf area index retrievals from LAI-2000, AccuPAR, and digital hemispherical photography over croplands

    Agric. For. Meteorol.

    (2008)
  • Y. Ge

    Principles and methods of scaling geospatial Earth science data

    Earth Sci. Rev.

    (2019)
  • A. Gonsamo

    Leaf area index retrieval using gap fractions obtained from high resolution satellite data: Comparisons of approaches, scales and atmospheric effects

    Int. J. Appl. Earth Obs. Geoinf.

    (2010)
  • A. Gonsamo et al.

    The computation of foliage clumping index using hemispherical photography

    Agric. For. Meteorol.

    (2009)
  • V. Haverd

    The Canopy Semi-analytic Pgap And Radiative Transfer (CanSPART) model: Formulation and application

    Agric. For. Meteorol.

    (2012)
  • L. He et al.

    Global clumping index map derived from the MODIS BRDF product

    Remote Sens. Environ.

    (2012)
  • L. He

    Inter- and intra-annual variations of clumping index derived from the MODIS BRDF product

    Int. J. Appl. Earth Obs. Geoinf.

    (2016)
  • R. Hernández-Clemente et al.

    Assessing the effects of forest health on sun-induced chlorophyll fluorescence using the FluorFLIGHT 3-D radiative transfer model to account for forest structure

    Remote Sens. Environ.

    (2017)
  • M.J. Hill

    Characterizing vegetation cover in global savannas with an annual foliage clumping index derived from the MODIS BRDF product

    Remote Sens. Environ.

    (2011)
  • R. Houborg et al.

    Intercomparison of a ‘bottom-up’ and ‘top-down’ modeling paradigm for estimating carbon and energy fluxes over a variety of vegetative regimes across the U.S

    Agric. For. Meteorol.

    (2009)
  • D. Huang

    Canopy spectral invariants for remote sensing and model applications

    Remote Sens. Environ.

    (2007)
  • H. Huang et al.

    RAPID: A Radiosity Applicable to Porous IndiviDual Objects for directional reflectance over complex vegetated scenes

    Remote Sens. Environ.

    (2013)
  • S. Jacquemoud

    PROSPECT + SAIL models: a review of use for vegetation characterization

    Remote Sens. Environ.

    (2009)
  • Z. Jiao

    An algorithm for the retrieval of the clumping index (CI) from the MODIS BRDF product using an adjusted version of the kernel-driven BRDF model

    Remote Sens. Environ.

    (2018)
  • Z. Jiao

    A method for improving hotspot directional signatures in BRDF models used for MODIS

    Remote Sens. Environ.

    (2016)
  • I. Jonckheere

    Review of methods for in situ leaf area index determination Part I. Theories, sensors and hemispherical photography

    Agric. For. Meteorol.

    (2004)
  • I. Jonckheere et al.

    A fractal dimension-based modelling approach for studying the effect of leaf distribution on LAI retrieval in forest canopies

    Ecol. Modell.

    (2006)
  • A. Kallel et al.

    Revisiting the vegetation hot spot modeling: Case of Poisson/Binomial leaf distributions

    Remote Sens. Environ.

    (2013)
  • A. Kuusk

    A Markov chain model of canopy reflectance

    Agric. For. Meteorol.

    (1995)
  • A. Kuusk

    A two-layer canopy reflectance model

    J. Quant. Spectrosc. Radiat. Transfer

    (2001)
  • Cited by (42)

    • hemispheR: an R package for fisheye canopy image analysis

      2023, Agricultural and Forest Meteorology
    View all citing articles on Scopus
    View full text