Elsevier

Ecological Indicators

Volume 69, October 2016, Pages 657-666
Ecological Indicators

Can ecological land classification increase the utility of vegetation monitoring data?

https://doi.org/10.1016/j.ecolind.2016.05.030Get rights and content

Highlights

  • Ecological land classification has potential to improve ecosystem trend analysis.

  • Classification utility was tested using a long-term vegetation record.

  • Trajectories of key vegetation attributes varied among ecological land classes.

  • We demonstrate the value of ecological land classification as a monitoring tool.

Abstract

Vegetation dynamics in rangelands and other ecosystems are known to be mediated by topoedaphic properties. Vegetation monitoring programs, however, often do not consider the impact of soils and other sources of landscape heterogeneity on the temporal patterns observed. Ecological sites (ES) comprise a land classification system based on soil, topographic, and climate variations that can be readily applied by land managers to classify topoedaphic properties at monitoring locations. We used a long-term (>40 y) vegetation record from southeastern Arizona, USA to test the utility of an ES classification for refining interpretations of monitoring data in an area of relatively subtle soil differences. We focused on two phenomena important to rangeland management in the southeastern Arizona region: expansion of the native tree velvet mesquite (Prosopis velutina Woot.) and spread of the introduced perennial grass Lehmann lovegrass (Eragrostis lehmanniana Nees). Specifically, we sought to determine if a quantitative, ES-specific analysis of the long-term record would (1) improve detection of changes in plant species having heightened ecological or management importance and (2) further clarify topoedaphic effects on vegetation trajectories. We found that ES class membership was a significant factor explaining spatiotemporal variation in velvet mesquite canopy cover, Lehmann lovegrass basal cover, and Lehmann lovegrass density measurements. In addition, we observed that the potential magnitude of velvet mesquite and Lehmann lovegrass increases varied substantially among ES classes. Our study brings attention to a practical land management tool that might be called upon to increase the effectiveness of vegetation-based indicators of ecosystem change.

Introduction

Vegetation monitoring is one of the principal methods used to assess the ecological consequences of management actions and climate change at local to landscape scales (Herrick et al., 2005). Vegetation dynamics at these scales can vary strongly in response to topoedaphic heterogeneity (Bestelmeyer et al., 2011, Pringle et al., 2006, Wu and Archer, 2005). For example, even relatively subtle variations in soil profile properties, such as the depth to clay- or carbonate-rich horizons in otherwise similar soils, can cause variations in rates of shrub encroachment or grass mortality (Bestelmeyer et al., 2006, Browning et al., 2012). Vegetation monitoring programs, however, often do not consider the impact of topoedaphic heterogeneity on the temporal patterns observed, which can lead to misinterpretation of early warning indicators or the importance of anthropogenic or climatic variables being studied (Pringle et al., 2006).

To address the effects of topoedaphic properties, some authors have recommended that monitoring sites be linked to soil- and climate-based land classification systems (Herrick et al., 2006, Karl and Herrick, 2010) such as the ecological site (ES) classifications used widely in the United States (Brown, 2010, USDA-NRCS, 2013 and similar classifications used worldwide (Blanco et al., 2014, Green and Klinka, 1994, Ray, 2001, van Gool and Moore, 1999). ES classes are subdivisions of a landscape based on soil, topographic, and/or climate properties known to influence vegetation composition and change (Duniway et al., 2010). Each ES class is associated with a state-and-transition model describing the vegetation changes that are likely to occur following specific management actions or natural events (Bestelmeyer et al., 2009, López et al., 2013). Land areas belonging to the same ES class are expected to provide the same general environment for plant establishment and growth. This expectation can give land managers increased confidence that the knowledge they have acquired from a particular vegetation monitoring effort can be effectively applied to other areas belonging to the same ES class (and only cautiously applied to other areas). In addition, the criteria used to differentiate ES classes are in most cases explicitly defined, which enables land managers to assess the degree of similarity between two classes and determine the suitability of applying ecological knowledge across class boundaries. In the United States separate ES classifications are created on a per-region basis, and individual ES classes are typically only utilized in that region they were developed for.

Given the important role of topoedaphic properties in controlling vegetation composition and dynamics, best practices commonly call for the incorporation of topoedaphic strata into vegetation monitoring designs. Use of ES classifications for landscape stratification is likely to increase with official commitment by three prominent US land management agencies − the Natural Resources Conservation Service, Forest Service, and Bureau of Land Management − to utilize ES classifications as a basis for monitoring, assessment, and planning in rangelands (BLM, 2010). ES classifications are already applied to a number of conservation activities and therefore represent a sensible tool for linking monitoring programs to other aspects of land management such as restoration projects and grazing plans. Nevertheless, there has been little empirical study aimed at supporting or refuting the utility of ES classifications with regard to ecosystem monitoring, despite recommendations to further incorporate ES classifications or similar frameworks into vegetation monitoring programs (Bestelmeyer et al., 2009, Herrick et al., 2006, Karl and Herrick, 2010).

We used an uncommonly long (>40 years), well-studied, and spatially extensive monitoring dataset available from the Santa Rita Experimental Range (SRER) to test for differences in vegetation trajectories among ES classes reflecting differences in subsoil properties in sandy soils of piedmont slope landforms. Long-term monitoring of ecological indicators is essential for resolving critical uncertainties in the detection of ecosystem trends, such as whether or not environmental degradation or improvement is taking place in ecosystems, like deserts, that respond slowly or episodically to management or climatic drivers. Increasing the effectiveness of ecological indicators may require addressing topoedaphic variation in a more systematic and detailed way than typically occurred in the past, and ES classification has been identified as one tool that could be used to address topoedaphic variation in such a manner (Bestelmeyer et al., 2009, Herrick et al., 2006). Our study provides a rare, empirical assessment of ES classification utility using an existing long-term monitoring dataset. By associating each SRER monitoring site with an ES class, we sought to determine if the detection of changes in plant species recognized as having heightened ecological or management importance in our study area would be improved. We also sought to determine whether previously unrecognized edaphic effects on vegetation trajectories had the potential to produce erroneous interpretations of vegetation monitoring data and associated indicators of ecosystem change. The ES classes studied here reflect differences in subsoil clay content that would likely go unnoticed by many observers without explicit consideration of ES classes, and earlier published analyses of the SRER long-term monitoring data did not address such soil variations. Finally, our study offered an opportunity to refine interpretations of a high-value long-term dataset and evaluate the need to modify the current ES classification system.

Section snippets

Focal species

We limited our analysis to two plant species having great management significance in the southeastern Arizona region: velvet mesquite (Prosopis velutina Woot.) and Lehmann lovegrass (Eragrostis lehmanniana Nees). Velvet mesquite is a small tree native to portions of Arizona, California, and New Mexico. Historically abundant on the SRER primarily along ephemeral drainages, the species has since colonized most upland areas of the research property (McClaran, 2003, McClaran et al., 2010).

Time series summary statistics

Compared to SLU plots (n = 18), plots assigned to the SLD (n = 24) class exhibited a broader distribution of mean and maximum velvet mesquite canopy cover values (Fig. 2). Median values of these two summary statistics were also higher for the SLD class than for the SLU class. Kruskal-Wallis tests indicated a significant difference in mean velvet mesquite canopy cover between SLD and SLU plots and a nearly significant difference (P = 0.0534) in maximum canopy cover values between these two ES classes (

Discussion

This study utilized a long-term vegetation record from southeastern Arizona to address a simple question: Can a readily-accessed ecological land classification, representing subtle topoedaphic variations that might otherwise go unnoticed by researchers, improve interpretations of vegetation monitoring data? We found that incorporating an ES classification into our analysis of a long-term monitoring dataset helped to explain spatial variations in the magnitude of historical vegetation change,

Summary

ES classifications have been adopted by land management agencies in the United States and other countries as frameworks for developing land unit-specific management recommendations, models of vegetation dynamics (e.g., state-and-transition models), and protocols for assessing ecosystem health. Monitoring activities, in turn, are recognized as an important tool for collecting information that can be used to support the development, testing, and refinement of vegetation change models, management

Acknowledgements

This research was supported in part by USDA CSREES grants 2007-38415-18637 and 2008-51130-19567.

References (49)

  • D.R. López et al.

    Linking ecosystem resistance, resilience, and stability in steppes of North Patagonia

    Ecol. Indic.

    (2013)
  • M.P. McClaran et al.

    Long-term vegetation response to mesquite removal in Desert Grassland

    J. Arid Environ.

    (2006)
  • M.P. McClaran et al.

    Recent drought phase in a 73-year record at two spatial scales: implications for livestock production on rangelands in the Southwestern United States

    Agric. For. Meteorol.

    (2014)
  • O.W. Van Auken

    Causes and consequences of woody plant encroachment into western North American grasslands

    J. Environ. Manage.

    (2009)
  • M.J. Anderson

    A new method for non-parametric multivariate analysis of variance

    Austral Ecol.

    (2001)
  • M.J. Anderson

    Distance-based tests for homogeneity of multivariate dispersions

    Biometrics

    (2006)
  • S. Archer

    Tree-grass dynamics in a Prosopis-thornscrub savanna parkland: reconstructing the past and predicting the future

    Ecoscience

    (1995)
  • [BLM] Bureau of Land Management,

    Rangeland Interagency Ecological Site Manual. US Department of the Interior, Bureau of Land Management, Manual 1734-1, Washington, D.C

    (2010)
  • B.T. Bestelmeyer et al.

    Soil-geomorphic heterogeneity governs patchy vegetation dynamics at an arid ecotone

    Ecology

    (2006)
  • B.T. Bestelmeyer et al.

    Spatial perspectives in state-and-transition models: a missing link to land management?

    J. Appl. Ecol.

    (2011)
  • D.M. Browning et al.

    Woody plants in grasslands: post-encroachment stand dynamics

    Ecol. Appl.

    (2008)
  • D.M. Browning et al.

    Hierarchical analysis of vegetation dynamics over 71 years: soil—rainfall interactions in a Chihuahuan Desert ecosystem

    Ecol. Appl.

    (2012)
  • J.R. Cox et al.

    The influence of climate and soils on the distribution of four African grasses

    J. Range Manag.

    (1988)
  • K.D. Eggemeyer et al.

    Biogeography of woody encroachment: why is mesquite excluded from shallow soils?

    Ecohydrology

    (2009)
  • Cited by (9)

    • Restoration of urban waterbird diversity: A case study of the construction of a waterbird ecological corridor in the Guangdong-Hong Kong-Macao Greater Bay Area, Southern China

      2022, Global Ecology and Conservation
      Citation Excerpt :

      Secondly, the current study provides an ecological corridor node selection method. Correct selection of ecological corridor nodes can increase the effectiveness of ecological corridor networks for achieving conservation outcomes (Williamson et al., 2016; Yu et al., 2018). Previous studies have combined the source-sink theory with habitat suitability evaluation for the identification of ecological corridor nodes(Li et al., 2010; Teng et al., 2011; Yu et al., 2018).

    • Optimization of ecological node layout and stability analysis of ecological network in desert oasis:a typical case study of ecological fragile zone located at Deng Kou County(Inner Mongolia)

      2018, Ecological Indicators
      Citation Excerpt :

      The ecological source node, which plays a role of engine in the overall ecological network, is energy output. The correct classification of the ecological source can make the extracted ecological network more accurate (Williamson et al., 2016). Many scholars have studied the classification of ecological source by area, ecosystem service value and so on (Chuai et al., 2016), but the grading factors are single.

    • State-and-transition models in geomorphology

      2017, Catena
      Citation Excerpt :

      State-and-transition models have gained widespread currency among ecologists, particularly those with research foci on arid and semi-arid environments in which state transitions are common. Indeed, state-transition frameworks have been used to inform ecological restoration projects, guide long-term monitoring experiments, assist with routine management activities, and simulate ecosystem trajectories under a range of climate change scenarios (e.g., Standish et al., 2009; Chambers et al., 2014; Provencher et al., 2016; Williamson et al., 2016). Many ecologists now leverage the vocabulary of STMs to frame their analyses and describe landscape dynamics.

    • Using Resilience and Resistance Concepts to Manage Persistent Threats to Sagebrush Ecosystems and Greater Sage-grouse

      2017, Rangeland Ecology and Management
      Citation Excerpt :

      Developing an understanding of the spatial linkages among processes and feedbacks for adjoining ESs, such as changes in vegetation cover and hydrologic and geomorphic processes, can provide early warning indicators of the potential for state transitions (Bestelmeyer et al., 2011). Comparisons of the rate and magnitude of change among ESs in response to different triggers can provide an assessment of the relative risk of state transitions among areas and time periods (Williamson et al., 2016). State-and-transition models that incorporate resilience and resistance concepts can be used to better evaluate change in landscape components and attributes in response to disturbance and management actions (Briske et al., 2008; Chambers et al., 2014a, 2014b).

    View all citing articles on Scopus
    View full text