Exploiting virtual globes for ecology and conservation in the Digital Earth era

Olea, Pedro P.; Mateo‐Tomás, Patricia

doi:10.1002/feeolealetter.1

Cited by 3 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…). Olea and Mateo‐Tomás () complain that virtual globes are currently underused by ecologists. The insights of our nest detection analyses and our workflow are broadly applicable, and we hope they will stimulate researchers and practitioners to get more out of non‐traditional image data, for wildlife surveys and beyond.…”

Section: Discussionmentioning

confidence: 99%

“…However, while absolute pixel values may not be reliable, the relationships between neighboring pixels are, and image texture and object-based analyses, such as those described here, of the imagery within virtual globes can support rigorous research and diverse applications (Barbier et al 2010;Mering et al 2010;Ploton et al 2012;Luo et al 2014). Olea and Mateo-Tom as (2016) complain that virtual globes are currently underused by ecologists. The insights of our nest detection analyses and our workflow are broadly applicable, and we hope they will stimulate researchers and practitioners to get more out of non-traditional image data, for wildlife surveys and beyond.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Semi‐automated detection of eagle nests: an application of very high‐resolution image data and advanced image analyses to wildlife surveys

Andrew

Shephard

2017

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

Very high-resolution (VHR) image data, including from unmanned aerial vehicle (UAV) platforms, are increasingly acquired for wildlife surveys. Animals or structures they build (e.g. nests) can be photointerpreted from these images, however, automated detection is required for more efficient surveys. We developed semi-automated analyses to map white-bellied sea eagle (Haliaeetus leucogaster) nests in VHR aerial photographs of the Houtman Abrolhos Islands, Western Australia, an important breeding site for many seabird species. Nest detection is complicated by high environmental heterogeneity at the scale of nests (~1-2 m), the presence of many features that resemble nests and the variability of nest size, shape and context. Finally, the rarity of nests limits the availability of training data. These challenges are not unique to wildlife surveys and we show how they can be overcome by an innovative integration of object-based image analyses (OBIA) and the powerful machine learning one-class classifier Maxent. Maxent classifications using features characterizing object texture, geometry and neighborhood, along with limited object color information, successfully identified over 90% of high quality nests (most weathered and unusually shaped nests were also detected, but at a slightly lower rate) and labeled <2% of objects as candidate nests. Although this overestimates the occurrence of nests, the results can be visually screened to rule out all but the most likely nests in a process that is simpler and more efficient than manual photointerpretation of the full image. Our study shows that semi-automated image analyses for wildlife surveys are achievable. Furthermore, the developed strategies have broad relevance to image processing applications that seek to detect rare features differing only subtly from a heterogeneous background, including remote sensing of archeological remains. We also highlight solutions to maximize the use of imperfect or uncalibrated image data, such as some UAV-based imagery and the growing body of VHR imagery available in Google Earth and other virtual globes.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Semi‐automated detection of eagle nests: an application of very high‐resolution image data and advanced image analyses to wildlife surveys

Andrew

Shephard

2017

Remote Sens Ecol Conserv

View full text Add to dashboard Cite

show abstract

“…Individual riparian vegetation models and the final co‐occurrence models were all locally validated by visual interpretation with coetaneous aerial photographs and Google Street View (Olea & Mateo‐Tomás, ) and field campaigns in selected locations. We made a particular effort in areas of high uncertainty in which low probability values co‐occur for different habitats.…”

Section: Methodsmentioning

confidence: 99%

“…iognomic and phytosociological) described above. The former map identified the distribution of the riparian forest classes, while the latter defined the composition of these physiognomic units at the habitat level.Individual riparian vegetation models and the final co-occurrence models were all locally validated by visual interpretation with coetaneous aerial photographs and Google Street View(Olea & Mateo- Tomás, 2016) and field campaigns in selected locations. We made a particular effort in areas of high uncertainty in which low probability values co-occur for different habitats.…”

mentioning

confidence: 99%

Modelling riparian forest distribution and composition to entire river networks

Pérez‐Silos

Álvarez‐Martínez

Barquín

2019

Applied Vegetation Science

View full text Add to dashboard Cite

Aim: Developing a methodology to map the distribution of riparian forests to entire river networks and determining the main environmental factors controlling their spatial patterns. Location: Cantabrian region, northern Spain. Methods: We mapped the riparian forests at a physiognomic and phytosociological level by delimiting riparian zones and generating vegetation distribution models based on remote sensing data (Landsat 8 OLI and LiDAR PNOA). We built virtual watersheds to define a spatial framework where the catchment environmental information can be specified for each river reach, in combination with the vegetation map. In order to determine the drivers that play a significant role in the observed spatial patterns in riparian forests, based on our data sets we modelled interactions between environmental information and riparian vegetation by using the Random Forest algorithm. Results: The modelling results obtained reliably reproduced the variation of riparian forest structure and composition across Cantabrian watersheds. The produced maps were highly accurate, with a more than 70% overall accuracy for forest occurrence. A clear differentiation between Eurosiberian (habitats 91E0 and 9160) and Mediterranean (92E0) riparian forests was shown on both sides of the mountain range. Topography and land use were the main drivers defining the distribution of riparian forest as a physiognomic unit. In turn, altitude, climate and percentage of pasture were the most relevant factors determining their composition (phytosociological approach). Conclusions: Our study confirms that anthropic control ultimately defines the distribution of vegetation in the riparian area at a regional to local scale. Human disturbances constrain the extension of forest patches across their potential distribution defined by topoclimatic boundaries, which establish a clear limit between Mediterranean and Eurosiberian biogeographical regions.

show abstract

Google Street View as an alternative method to car surveys in large-scale vegetation assessments

Deus

Silva

Catry

et al. 2016

Environ Monit Assess

View full text Add to dashboard Cite

Car surveys (CS) are a common method for assessing the distribution of alien invasive plants. Google Street View (GSV), a free-access web technology where users may experience a virtual travel along roads, has been suggested as a cost-effective alternative to car surveys. We tested if we could replicate the results from a countrywide survey conducted by car in Portugal using GSV as a remote sensing tool, aiming at assessing the distribution of Eucalyptus globulus Labill. wildlings on roadsides adjacent to eucalypt stands. Georeferenced points gathered along CS were used to create road transects visible as lines overlapping the road in GSV environment, allowing surveying the same sampling areas using both methods. This paper presents the results of the comparison between the two methods. Both methods produced similar models of plant abundance, selecting the same explanatory variables, in the same hierarchical order of importance and depicting a similar influence on plant abundance. Even though the GSV model had a lower performance and the GSV survey detected fewer plants, additional variables collected exclusively with GSV improved model performance and provided a new insight into additional factors influencing plant abundance. The survey using GSV required ca. 9 % of the funds and 62 % of the time needed to accomplish the CS. We conclude that GSV may be a cost-effective alternative to CS. We discuss some advantages and limitations of GSV as a survey method. We forecast that GSV may become a widespread tool in road ecology, particularly in large-scale vegetation assessments.

show abstract

Exploiting virtual globes for ecology and conservation in the Digital Earth era

Cited by 3 publications

References 14 publications

Semi‐automated detection of eagle nests: an application of very high‐resolution image data and advanced image analyses to wildlife surveys

Semi‐automated detection of eagle nests: an application of very high‐resolution image data and advanced image analyses to wildlife surveys

Modelling riparian forest distribution and composition to entire river networks

Google Street View as an alternative method to car surveys in large-scale vegetation assessments

Contact Info

Product

Resources

About