Alex M. Lechner

The sensitivity of population viability analysis to uncertainty about habitat requirements: implications for the management of the endangered southern brown bandicoot

Whenever population viability analysis (PVA) models are built to help guide decisions about the management of rare and threatened species, an important component of model building is the specification of a habitat model describing how a species is related to landscape or bioclimatic variables. Model-selection uncertainty may arise because there is often a great deal of ambiguity about which habitat model structure best approximates the true underlying biological processes. The standard approach to incorporate habitat models into PVA is to assume the best habitat model is correct, ignoring habitat-model uncertainty and alternative model structures that may lead to quantitatively different conclusions and management recommendations. Here we provide the first detailed examination of the influence of habitat-model uncertainty on the ranking of management scenarios from a PVA model. We evaluated and ranked 6 management scenarios for the endangered southern brown bandicoot (Isoodon obesulus) with PVA models, each derived from plausible competing habitat models developed with logistic regression. The ranking of management scenarios was sensitive to the choice of the habitat model used in PVA predictions. Our results demonstrate the need to incorporate methods into PVA that better account for model uncertainty and highlight the sensitivity of PVA to decisions made during model building. We recommend that researchers search for and consider a range of habitat models when undertaking model-based decision making and suggest that routine sensitivity analyses should be expanded to include an analysis of the impact of habitat-model uncertainty and assumptions.

A tool for simulating and communicating uncertainty when modelling species distributions under future climates

Tools for exploring and communicating the impact of uncertainty on spatial prediction are urgently needed, particularly when projecting species distributions to future conditions. We provide a tool for simulating uncertainty, focusing on uncertainty due to data quality. We illustrate the use of the tool using a Tasmanian endemic species as a case study. Our simulations provide probabilistic, spatially explicit illustrations of the impact of uncertainty on model projections. We also illustrate differences in model projections using six different global climate models and two contrasting emissions scenarios. Our case study results illustrate how different sources of uncertainty have different impacts on model output and how the geographic distribution of uncertainty can vary. Synthesis and applications: We provide a conceptual framework for understanding sources of uncertainty based on a review of potential sources of uncertainty in species distribution modelling; a tool for simulating uncertainty in species distribution models; and protocols for dealing with uncertainty due to climate models and emissions scenarios. Our tool provides a step forward in understanding and communicating the impacts of uncertainty on species distribution models under future climates which will be particularly helpful for informing discussions between researchers, policy makers, and conservation practitioners.

Characterizing spatial uncertainty when integrating social data in conservation planning

Recent conservation planning studies have presented approaches for integrating spatially referenced social (SRS) data with a view to improving the feasibility of conservation action. We reviewed the growing conservation literature on SRS data, focusing on elicited or stated preferences derived through social survey methods such as choice experiments and public participation geographic information systems. Elicited SRS data includes the spatial distribution of willingness to sell, willingness to pay, willingness to act, and assessments of social and cultural values. We developed a typology for assessing elicited SRS data uncertainty which describes how social survey uncertainty propagates when projected spatially and the importance of accounting for spatial uncertainty such as scale effects and data quality. These uncertainties will propagate when elicited SRS data is integrated with biophysical data for conservation planning and may have important consequences for assessing the feasibility of conservation actions. To explore this issue further, we conducted a systematic review of the elicited SRS data literature. We found that social survey uncertainty was commonly tested for, but that these uncertainties were ignored when projected spatially. Based on these results we developed a framework which will help researchers and practitioners estimate social survey uncertainty and use these quantitative estimates to systematically address uncertainty within an analysis. This is important when using SRS data in conservation applications because decisions need to be made irrespective of data quality and well characterized uncertainty can be incorporated into decision theoretic approaches.

Comparison of relative radiometric normalization methods using pseudo-invariant features for change detection studies in rural and urban landscapes

Abstract. Relative radiometric normalization (RRN) to remove sensor effects, solar and atmospheric variation from at-sensor radiance values is often necessary for effective detection of temporal change. Traditionally, pseudo-invariant features (PIFs) are chosen subjectively, where as an analyst manually chooses known objects, often man-made, that should not change over time. An alternative method of selecting PIFs uses a principal component analysis (PCA) to select the PIFs. We compare the two RRN methods using PIFs in multiple Landsat images of urban and rural areas in Australia. An assessment of RRN quality was conducted including measurements of slope, root mean square error, and normalized difference vegetation index. We found that in urban areas both methods performed similarly well. However, in the rural area the automated PIF selection method using a PCA performed better due to the rarity of built features that are required for the manual PIF selection. We also found that differences in performance of the manual and automated methods were dependent on the accuracy assessment method tested. We conclude with a discussion on the relative merits of different RRN methods and practical advice on how to apply the automated PIF selection method.

Characterising landscape connectivity for conservation planning using a dispersal guild approach

ContextLand use changes have modified the extent and structure of native vegetation, resulting in fragmentation of native species habitat. Connectivity is increasingly seen as a requirement for effective conservation in these landscapes, but the question remains: ‘connectivity for which species?’.ObjectiveThe aim of this study was to develop and then apply a rapid, expert-based, dispersal guild approach where species are grouped on similar fine-scale dispersal behaviour (such as between scattered trees) and habitat characteristics.MethodsDispersal guilds were identified using clustering techniques to compare dispersal and habitat parameters elicited from experts. We modelled least-cost paths and corridors between patches and individual movement probabilities within these corridors for each of the dispersal guilds using Circuitscape. We demonstrate our approach with a case study in the Tasmanian Northern Midlands, Australia.ResultsThe dispersal guild approach grouped the 12 species into five dispersal guilds. The connectivity modelling of those five guilds found that broadly dispersing species in this landscape, such as medium-sized carnivorous mammals, were unaffected by fragmentation while from the perspective of the three dispersal guilds made up of smaller mammals, the landscape appeared highly fragmented.ConclusionsOur approach yields biologically defensible outputs that are broadly applicable, particularly for conservation planning where data and resources are limited. It is a useful first step in multi-species conservation planning which aims to identify those species most in need of conservation efforts.

Measuring fire severity using UAV imagery in semi-arid central Queensland, Australia

ABSTRACT Remote-sensing methods for fire severity mapping have traditionally relied on multispectral imagery captured by satellite platforms carrying passive sensors such as Landsat Thematic Mapper /Enhanced Thematic Mapper Plus or Moderate Resolution Imaging Spectroradiometer. This article describes the analysis of high spatial resolution Unmanned Aerial Vehicle (UAV) imagery to assess fire severity on a 117 ha experimental fire conducted on coal mine rehabilitation in an open woodland environment in semi-arid Central Queensland, Australia. Three band indices, Excess Green Index, Excess Green Index Ratio, and Modified Excess Green Index, were used to derive differenced (d) fire severity maps from UAV data. Fire severity data sets derived from aerial photograph interpretation were used to assess the utility of employing UAV technology to determine fire severity impacts. The dEGI was able to separate high severity, low severity, and unburnt areas with an overall classification accuracy of 58% and Kappa statistic of 0.37; outperforming the dEGIR (overall accuracy 55%, Kappa 0.31) and the dMEGI (overall accuracy 38%, Kappa 0.06). Classification accuracy increased for all indices when canopy shadows were masked, with dEGI improving to an overall accuracy of 68% and 0.48 Kappa. The McNemar’s test indicated that there was no significant difference between the classification accuracies for dEGI and dEGIR (p < 0.05). The test also demonstrated that dMEGI was significantly lower in accuracy compared to dEGI and dEGIR (p < 0.05). We quantified the proportion of burnt area within each severity class and calculated that 32% of the site was burnt at high severity, 34% was burnt at low severity, and 34% of the block was unburnt due to the patchy nature of the fire. We discuss the UAV-specific errors associated with fire severity mapping, and the potential for UAVs to assist land managers to assess the extent and severity of fire and subsequent recovery of burnt ecosystems at local scales (104m2–1 km2).

Challenges of integrated modelling in mining regions to address social, environmental and economic impacts

Planning in mining regions needs to accommodate the extraction of minerals/energy resources in co-existence with established land uses, such as agriculture and ecological conservation. Here, we first identify six critical aspects of planning in mining regions: i) the temporal nature of mining operations; ii) spatial dimensions of mining operations; iii) irreversible changes that create post-mining landscapes; iv) social dimensions of mining impacts and corporate responsibility; v) cumulative dimensions of impacts; and vi) a need to integrate methods from a range of disciplines. We then illustrate the potential to address these challenges using integrative modelling nested within a participatory approach to allow for clear, transparent, and stakeholder-inclusive decision-making. We describe a 5-step framework that supports a broadening of strategic assessments and offers mining companies forewarning about potential environmental and social conflicts. Case studies are needed to assess and refine the proposed framework and develop guidance for its use. Display Omitted Planning for mining regions needs to accommodate multiple established land uses.Five aspects of mining affect the application of integrated regional modelling.We describe an integrated, multi-disciplinary modelling approach for mining regions.Such modelling and decision making need to be participatory in order to reduce conflict.This framework supports a broadening of strategic assessments for mining.

SPOTing long-term changes in vegetation over short-term variability

This study investigates change detection in the vegetation cover of a closed gold mine to assess whether the rehabilitated vegetation responds in a similar manner to the surrounding environment. Rehabilitation took place in seven rehabilitation areas within the mine site. SPOT scenes covering the site were acquired at six time periods from September 2004 to September 2005, and annually in the dry season from 2004 to 2010, except for 2008. Normalized difference vegetation index, soil-adjusted vegetation index and transformed soil-adjusted vegetation index were tested to estimate the percentage vegetation cover (PVC) using a linear regression model. The results showed higher PVC during wet season and lower PVC during the dry season in the native vegetation surrounding the mine site. However, temporal and spatial patterns of PVC in rehabilitated and native areas were similar only in the TD40ha rehabilitation area, one of the seven rehabilitation areas. This area was the first to be rehabilitated and had the most intensive rehabilitation effort using tube stock planting in March 1998. The seasonal variability showed a high correlation with an r2 value of 0.77 in TD40ha rehabilitation area in tailings dam, which was similar to the native area with an r2 value of 0.82. The findings of this study suggest that it is important for monitoring programs to take into account seasonal variation and environmental covariates, such as rainfall in order to successfully assess patterns in vegetation condition over time.

Object-based classification of semi-arid vegetation to support mine rehabilitation and monitoring

Abstract Mining activities result in significantly modified landscapes that require rehabilitation to mitigate the negative environmental impacts and restore ecological function. The aim of this study was to develop a remote sensing method suitable for monitoring the vegetation cover at mine rehabilitation sites. We used object-based image analysis (OBIA) methods and high-spatial resolution SPOT-5 imagery to identify discrete land-cover patterns that occur at fine spatial scales. These patterns relate to spatial processes that are important drivers of successful restoration of mine sites. SPOT-5 imagery of the Kidston Gold mine tailing dam in semi-arid tropical north Queensland was acquired in July 2005, comprising four 10-m spectral bands and a 2.5-m panchromatic (PAN) band. The classification scheme used in this study was adapted to the spatial scale of SPOT-5 imagery from mine closure criteria cover requirements, according to a mine rehabilitation plan. Four land-cover classes were identified: tree cover, dense grass, sparse grass, and bare ground. First, textural layers (contrast, dissimilarity, and homogeneity) were derived for each vegetation class except for bare ground from the PAN and multispectral bands. Of all textural layer combinations, homogeneity and contrast in the PAN band were identified using a Z -test as the most useful for differentiating between multiple land-cover classes. Next, an optimal segmentation scale parameter of 15 was identified using an analysis of spatial autocorrelation. Finally, the SPOT-5 image bands, derived textural layers, and normalized difference vegetation index (NDVI) were used in an OBIA fuzzy membership classification approach to map vegetation land-cover classes. The classification results were assessed with the traditional error matrix approach and the object-based accuracy assessment method. The overall classification accuracy using the error matrix was 92.5% and 81% using the object-based method. The relatively high-classification accuracy demonstrates the potential of SPOT-5 imagery for monitoring mine rehabilitation. The complete spatial coverage associated with remote sensing data at fine spatial scales has the potential to complement field-based approaches commonly used in rehabilitation monitoring. Furthermore, SPOT-5 data along with OBIA can characterize vegetation spatial patterns at spatial scales appropriate for monitoring rehabilitated landscapes, providing an important tool for landscape function analysis.

Biodiversity conservation should be a core value of China’s Belt and Road Initiative

To the Editor — China’s Belt and Road Initiative (BRI; also known as ‘One Belt One Road’) is potentially the largest infrastructure development in our lifetime. In 2013, President Xi Jinping revealed his vision for BRI, which is expected to be core to China’s development strategy for at least the next decade1. With an estimated cost of over 4 trillion US dollars, BRI will connect roughly half of the world’s population, across more than 65 countries, with land and marine routes2,3. Although much has been discussed about its economic and geopolitical implications, the implications of BRI for biodiversity must also be considered, especially in Asia. Infrastructure and its impacts are key drivers of biodiversity loss. BRI will cross several terrestrial and marine biodiversity hotspots4,5, wilderness areas6 and other key conservation areas, such as southeast Asia’s Coral Triangle6 (Fig. 1). These disruptions will create obvious threats to biodiversity. Roads, for example, open a Pandora’s box of environmental impacts, such as habitat loss, fragmentation, invasive species, and illegal activities such as poaching and logging7. In the marine environment, increased sea traffic exacerbates the movement of invasive species and pollution8,9. Poorly planned infrastructure has the risk of locking in undesirable environmental practices for decades to come. BRI could have disastrous consequences for biodiversity. We challenge decision-makers, infrastructure planners and conservationists to work together not only to mitigate BRI’s negative impacts, but also to think how to transform this juggernaut into an opportunity for biodiversity. If BRI adopts biodiversity conservation as one of its core values, it could, for example, plan and implement a network of protected areas and wildlife corridors across Eurasia. In much of BRI’s region, especially in southeast, central and western Asia, there is a clear