Dataset Information

Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on agricultural land.

ABSTRACT: In conservation, the use of unmanned aerial vehicles (drones) carrying various sensors and the use of deep learning are increasing, but they are typically used independently of each other. Untapping their large potential requires integrating these tools. We combine drone-borne thermal imaging with artificial intelligence to locate ground-nests of birds on agricultural land. We show, for the first time, that this semi-automated system can identify nests with a high performance. However, local weather, type of arable field and height of the drone can affect performance. The results' implications are particularly relevant to conservation practitioners working across sectors, such as biodiversity conservation and food production in farmland. Under a rapidly changing world, studies like this can help uncover the potential of technology for conservation and embrace cross-sectoral transformations from the onset; for example, by integrating nest detection within the precision agriculture system that heavily relies on drone-borne sensors.

SUBMITTER: Santangeli A

PROVIDER: S-EPMC7360548 | biostudies-literature | 2020 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on agricultural land.

Santangeli Andrea A Chen Yuxuan Y Kluen Edward E Chirumamilla Raviteja R Tiainen Juha J Loehr John J

Scientific reports 20200714 1

In conservation, the use of unmanned aerial vehicles (drones) carrying various sensors and the use of deep learning are increasing, but they are typically used independently of each other. Untapping their large potential requires integrating these tools. We combine drone-borne thermal imaging with artificial intelligence to locate ground-nests of birds on agricultural land. We show, for the first time, that this semi-automated system can identify nests with a high performance. However, local wea ...[more]

PMID: 32665596

Similar Datasets

Project description:Temporal variation in the composition of species assemblages could be the result of deterministic processes driven by environmental change and/or stochastic processes of colonization and local extinction. Here, we analyzed the relative roles of deterministic and stochastic processes on bird assemblages in an agricultural landscape of southwestern France. We first assessed the impact of land cover change that occurred between 1982 and 2007 on (i) the species composition (presence/absence) of bird assemblages and (ii) the spatial pattern of taxonomic beta diversity. We also compared the observed temporal change of bird assemblages with a null model accounting for the effect of stochastic dynamics on temporal beta diversity. Temporal assemblage dissimilarity was partitioned into two separate components, accounting for the replacement of species (i.e. turnover) and for the nested species losses (or gains) from one time to the other (i.e. nestedness-resultant dissimilarity), respectively. Neither the turnover nor the nestedness-resultant components of temporal variation were accurately explained by any of the measured variables accounting for land cover change (r(2)<0.06 in all cases). Additionally, the amount of spatial assemblage heterogeneity in the region did not significantly change between 1982 and 2007, and site-specific observed temporal dissimilarities were larger than null expectations in only 1% of sites for temporal turnover and 13% of sites for nestedness-resultant dissimilarity. Taken together, our results suggest that land cover change in this agricultural landscape had little impact on temporal beta diversity of bird assemblages. Although other unmeasured deterministic process could be driving the observed patterns, it is also possible that the observed changes in presence/absence species composition of local bird assemblages might be the consequence of stochastic processes in which species populations appeared and disappeared from specific localities in a random-like way. Our results might be case-specific, but if stochastic dynamics are generally dominant, the ability of correlative and mechanistic models to predict land cover change effects on species composition would be compromised.

Project description:Protected areas are one of the primary conservation tools used worldwide. However, they are often embedded in a landscape that is intensely used by people, such as for agriculture or urban development. The proximity of these land-use types to protected areas can potentially affect the ecological effectiveness (or conservation effectiveness) of protected areas. In this article, we examine to what degree adjacent agricultural and urban land uses affect the ecological effectiveness of protected areas over the greater Gauteng region of South Africa. We selected 198 common, resident bird species, and analysed detection/non-detection data for these species collected over regular grid cells (approximately 61 km2 in area). For each species, we estimated abundance per grid cell with the Royle-Nichols model in relation to the proportion of protected area as a covariate. Our study focused on how this relationship between proportion of protected area and abundance (which we term the 'protection-abundance relationship') changed as a function of other land-use types in the grid cell. Specifically, we examined the interaction effects between protected area and both urban and agricultural land-use type per grid cell on bird abundance. We assigned each species to one of seven guilds, namely: frugivores, gleaners, granivores, ground-feeders, hawkers, predators and vegivores, and examined how the protection-abundance relationship varied across guilds in relation to agriculture and urban area. As urban area within a grid cell increased, the protection-abundance relationship became more positive for 58% of all species. At the level of guilds, the protection-abundance relationship became more positive for two guilds (granivores and ground-feeders), more negative for frugivores, and remained unchanged for the other four guilds (gleaners, hawkers, predators and vegivores). As agricultural area within a grid cell increased, the protection-abundance relationship became more positive for 49% of all species. At the guild level, the protection-abundance relationship became more positive for six guilds (frugivores, gleaners, ground-feeders, hawkers, predators and vegivores) and remained unchanged for the granivores. Our results show land-use type near protected areas modified the effect protected areas had on bird abundances, and hence the ecological effectiveness of protected areas. Our results suggest that protected areas should be viewed as constituents within the landscape, rather than islands of protection.

Dataset Information

Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on agricultural land.

Publications

Integrating drone-borne thermal imaging with artificial intelligence to locate bird nests on agricultural land.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets