Integrating environmental variables by multivariate ordination enables the reliable estimation of mineland rehabilitation status

Abstract

Highlights➢ A multivariate estimation of the environmental status of rehabilitating areas is presented ➢ The method integrates processes, diversity and structure in a single estimation ➢ Rehabilitation status is successfully estimated after iron ore mining in Corumbá ➢ The approach enables tests for biases regarding single variables or variable groups ➢ Defining indicators reduces the number of variables needed for further assessments AbstractDespite the wide variety of variables commonly applied to measure different aspects of rehabilitation, the assessment and subsequent definition of indicators of environmental rehabilitation status are not simple tasks. The main challenges are comparing rehabilitated sites with target ecosystems as well as integrating individual environmental and eventually collinear variables into a single tractable measure of the state of a system before effective indicators that track rehabilitation may be modeled. For that, a consensus is lacking regarding which and how many variables need to be surveyed. Our approach considered ecological processes, vegetation structure, and community diversity from nonrehabilitated, rehabilitating and reference sites. We applied this approach to a curated set of 32 environmental variables retrieved from nonrevegetated, rehabilitating and reference sites associated with iron ore mines from the Urucum Massif, Mato Grosso do Sul, Brazil. By integrating variables from a single attribute or the entire set of variables into a single estimation of rehabilitation status, the proposed multivariate approach is straightforward and able to adequately address collinearity among variables. The proposed approach allows for the identification of biases towards single variables, surveys or analyses, which is necessary to rank environmental variables regarding their importance to the assessment. Furthermore, we show that bootstrapping permitted the detection of the minimum number of environmental variables necessary to achieve reliable estimations of the rehabilitation status. Finally, we show that the proposed variable integration enables the definition of environmental indicators for more comprehensive monitoring of mineland rehabilitation. Thus, the

Land Cover Changes in Open-Cast Mining Complexes Based on High-Resolution Remote Sensing Data

“…Thus, these spectral differences were sufficient to track the land use changes and accurately quantify the effectively revegetated areas. However, the assessment of revegetation quality still requires ground monitoring [13]. Specifically, the two-step segmentation of the high-resolution images enabled the identification of small revegetated structures, such as strait slopes, indicating the suitability of the proposed methodology for this study [22].…”

DNA-Based Arthropod Diversity Assessment in Amazonian Iron Mine Lands Show Ecological Succession Towards Undisturbed Reference Sites

“…As mentioned previously, primer bias can cause DNA amplification of certain arthropod taxa that may be more or less correlated to these vegetation parameters and therefore result in differences between primers used. Apart from the presence or absence of certain taxa, for a complete assessment of environmental rehabilitation status, functional diversity should also be taken into account (Gastauer et al, 2020a). We found the compositional differences in arthropod communities in the waste piles to be driven by the presence or absence of certain groups of arthropods.…”

“…Vegetation structure such as basal area, tree density, tree species richness and leaf area index have previously been shown to be important when assessing environmental rehabilitation status (Ruiz-Jaen and Mitchell Aide, 2005;Gastauer et al, 2020a). As arthropod diversity and composition are known to be correlated with vegetation structure (Basset et al, 2012;Zhang et al, 2016), we fitted vegetation data with arthropod community and found that the basal area and tree species richness were correlated with the detected arthropod community.…”

“…Despite the need for monitoring of these areas, no consensus has been reached about which environmental variables are the best indicators for measuring ecosystem state and change (Gastauer et al, 2018(Gastauer et al, , 2020a. However, measurements of vegetation structure such as canopy openness, tree density, vegetation cover and soil organic carbon are commonly used (Wortley et al, 2013;Lorenz et al, 2019).…”

“…However, measurements of vegetation structure such as canopy openness, tree density, vegetation cover and soil organic carbon are commonly used (Wortley et al, 2013;Lorenz et al, 2019). Specifically, basal area, tree density, tree species richness and leaf area index have previously been shown to be important when assessing environmental rehabilitation status (Ruiz-Jaen and Mitchell Aide, 2005;Gastauer et al, 2020a). Alternatively, arthropods can be used, as they directly make up a large proportion of terrestrial biodiversity and because arthropod species diversity and composition closely follow the diversity and composition of plant species (Basset et al, 2012;Zhang et al, 2016), providing a convenient way to measure both sets of taxa.…”