Search

Accurate estimates of pasture biomass in grazing lands are currently a time-consuming and resource-intensive task. The process generally includes physically cutting, bagging, labelling, drying, and weighing grass samples using multiple “quadrats” placed on the ground. Quadrats vary in size but are typically in the order of 0.25 m2 (i.e., 0.5 m × 0.5 m) up to 1.0 m2. Measurements from a number of harvested quadrats are then averaged to get a site estimate. This study investigated the use of photographs and ‘machine learning’ to reduce the time factor and difficulty in taking pasture biomass measurements to potentially make the estimations more accessible through the use of mobile phone cameras. A dataset was created from a pre-existing archive of quadrat photos and corresponding hand-cut pasture biomass measurements taken from a diverse range of field monitoring sites. Sites were clustered and one was held back per model for testing. The models were based on DenseNet121. Individual quadrat errors were large but more promising results were achieved when estimating the site mean pasture biomass. Another two smaller additional datasets were created post-training which were used to further assess the ensemble; they provided similar absolute errors to the original dataset, but significantly larger relative errors. The first was made from harvested quadrats, and the second was made using a pasture height meter in conjunction with a mobile phone camera. The models performed well across a variety of situations and locations but underperformed when assessed on some sites with very different vegetation. More data and refinement of the approach outlined in the paper will reduce the number of models needed and help to correct errors. These models provide a promising start, but further investigation, refinement, and data are needed before becoming a usable application.

Accurate estimates of pasture biomass in grazing lands are currently a time-consuming and resource-intensive task. The process generally includes physically cutting, bagging, labelling, drying, and weighing grass samples using multiple "quadrats" placed on the ground. Quadrats vary in size but are typically in the order of 0.25 m2 (i.e., 0.5 m × 0.5 m) up to 1.0 m2. Measurements from a number of harvested quadrats are then averaged to get a site estimate. This study investigated the use of photographs and 'machine learning' to reduce the time factor and difficulty in taking pasture biomass measurements to potentially make the estimations more accessible through the use of mobile phone cameras. A dataset was created from a pre-existing archive of quadrat photos and corresponding hand-cut pasture biomass measurements taken from a diverse range of field monitoring sites. Sites were clustered and one was held back per model for testing. The models were based on DenseNet121. Individual quadrat errors were large but more promising results were achieved when estimating the site mean pasture biomass. Another two smaller additional datasets were created post-training which were used to further assess the ensemble; they provided similar absolute errors to the original dataset, but significantly larger relative errors. The first was made from harvested quadrats, and the second was made using a pasture height meter in conjunction with a mobile phone camera. The models performed well across a variety of situations and locations but underperformed when assessed on some sites with very different vegetation. More data and refinement of the approach outlined in the paper will reduce the number of models needed and help to correct errors. These models provide a promising start, but further investigation, refinement, and data are needed before becoming a usable application. [ABSTRACT FROM AUTHOR]

Seasonal variation in seagrass growth and senescence affects the provision of ecosystem services and restoration efforts, requiring seasonal monitoring. Remotely piloted aircraft systems (RPAS) enable frequent high-resolution surveys at full-meadow scales. However, the reproducibility of RPAS surveys is challenged by varying environmental conditions, which are common in temperate estuarine systems. We surveyed three eelgrass (Zostera marina) meadows in Newfoundland, Canada, using an RPAS equipped with a three-color band (red, green, blue [RGB]) camera, to evaluate the seasonal reproducibility of RPAS surveys and assess the effects of flight altitude (30–115 m) on classification accuracy. Habitat percent cover was estimated using supervised image classification and compared to corresponding estimates from snorkel quadrat surveys. Our results revealed inconsistent misclassification due to environmental variability and low spectral separability between habitats. This rendered differentiating between model misclassification versus actual changes in seagrass cover infeasible. Conflicting estimates in seagrass and macroalgae percent cover compared to snorkel estimates could not be corrected by decreasing the RPAS altitude. Instead, higher altitude surveys may be worth the trade-off of lower image resolution to avoid environmental conditions shifting mid-survey. We conclude that RPAS surveys using RGB imagery alone may be insufficient to discriminate seasonal changes in estuarine subtidal vegetated habitats.

Many of the methods used for estimating population size from ecological surveys have limitations on precision, cost, and/or applicability. The CountEm method was proposed recently for estimating the number of individuals in large groups from single images. It is simple and efficient, and can be applied to any species. Here we present a case study by applying CountEm to a real ecological survey with 278 images of Greater Snow Geese (Anser caerulescens atlanticus) and Common Eiders (Somateria mollissima) flocks taken from fixed‐wing aircraft in Eastern Canada. First, we evaluated the precision and counting time of CountEm on single images. Second, we developed and tested a new multi‐image version of the CountEm software. We show that flock sizes of N > 35,000 can be estimated on single images in ∼5 min, from counting a sample of ∼200 birds, yielding relative SEs in the 5%−10% range. Processing times increased to 10–20 min when simultaneously processing large numbers of images that contained over half a million birds with only modest increases in relative SE (range: 10%−15%). Our results suggest that CountEm may be used to save time and resources if incorporated into monitoring programs that utilize imagery in the abundance estimates. [ABSTRACT FROM AUTHOR]

Abstract Many of the methods used for estimating population size from ecological surveys have limitations on precision, cost, and/or applicability. The CountEm method was proposed recently for estimating the number of individuals in large groups from single images. It is simple and efficient, and can be applied to any species. Here we present a case study by applying CountEm to a real ecological survey with 278 images of Greater Snow Geese (Anser caerulescens atlanticus) and Common Eiders (Somateria mollissima) flocks taken from fixed‐wing aircraft in Eastern Canada. First, we evaluated the precision and counting time of CountEm on single images. Second, we developed and tested a new multi‐image version of the CountEm software. We show that flock sizes of N > 35,000 can be estimated on single images in ∼5 min, from counting a sample of ∼200 birds, yielding relative SEs in the 5%−10% range. Processing times increased to 10–20 min when simultaneously processing large numbers of images that contained over half a million birds with only modest increases in relative SE (range: 10%−15%). Our results suggest that CountEm may be used to save time and resources if incorporated into monitoring programs that utilize imagery in the abundance estimates.

Marinas are high-priority targets for marine non-indigenous species (NIS), where they compose a large portion of the biofouling communities. The practicality of water samples collection makes environmental DNA (eDNA) metabarcoding an interesting tool for routine NIS surveys. Here the effectiveness of water-eDNA-metabarcoding to identify biofouling NIS, in 10 marinas from western France, was examined. Morphological identification of specimens collected in quadrats brought out 18 sessile benthic NIS beneath floating pontoons. Water-eDNA-metabarcoding detected two thirds of them, failing to detect important NIS. However, sampling and bioinformatics filtering steps can be optimized to identify more species. In addition, this method allowed the detection of additional NIS from neighboring micro-habitats. Caution should, however, be taken when reporting putative novel NIS, because of errors in species assignment. This work highlights that water-eDNA-metabarcoding is effective for active (targeted) NIS surveys and could be significantly improved for its further use in marine NIS passive surveys. [ABSTRACT FROM AUTHOR]

A survey to assess the effects and sustainability of charcoal production on the local vegetation cover was conducted in the Sekyere Central District of Ashanti region from December 2017 to April 2018. The study involved the use of structured questionnaires, focus group discussions (FGDs), key informants, secondary data and field observation by reconnaissance survey. The socio-economic survey was to identify who in the district is undertaking charcoal production as an economic venture, how widespread the activity is, and the preferred tree species being harvested to produce charcoal. In the field survey, the vegetation types in the study area were identified in a reconnaissance survey as secondary forests with increasing grassland. The relative abundance of tree species which could be used for charcoal production was also identified and quantified. A 640-meter square plot was delimited and divided into 64 square plots within each of which five-meter quadrats were marked at random and presence or absence was recorded for the individual tree species. The results of the study indicated that charcoal production is widely carried out throughout the district especially in the Afram Plains portion. It is practiced by the Akan and the Sissala ethnic groups who have very little education. Female producers (8%) were less than their male counterparts. Majority (73%) of the producers was 31 and above years whiles those of 30 years and below was 27%. The study revealed that most tree species are becoming rare and are only common in the protected area (Kogyae strict natures reserve) in the district.

Wild Water Buffalo (WWB) Bubalus arnee is an endangered species and a protected animal in Nepal. The remaining WWB population is located in Koshi Tappu Wildlife Reserve (KTWR), and it appears to have low viability under prevailing conditions. We assessed the habitat use and conservation threats to wild water buffalo in KTWR. For habitat analysis the quadrant method was used. Eighty-four quadrants of 25 m2 for trees, 168 quadrants of 10m2 for shrubs and 336 quadrants of 1 m2 for herbs were laid out in the study area. Ivlev's electivity index (IV) was calculated to assess the use of different habitat components. The important Value Index (IVI) was used for vegetation assessment. A relative threat ranking method was used to assess conservation threats for wild buffalo and their habitats. Wild buffalo mostly preferred habitats with distance to water resources less than 500 m (IV= 0.4), less than 25 % crown coverage (IV= 0.39) and more than 75 % ground coverage (IV= 0.42). The trees species Phyllanthus emblica, Acacia catechu, shrub species Mimosa pudica and the herb species Brachiaria distachya, Vetiveria zizanioides, Imperata cylindrica, and Saccharum spontaneum were preferred by WWB in the study area. Among the different plant categories, we found that Acacia catechu was the most preferred tree species (IVI= 156.95), Mimosa pudica the most preferred shrub species (IVI= 58.68), and Imperata cylindrica the most preferred herb species (IVI= 64.73). Major conservation threats perceived by local stakeholders for wild buffaloes were overgrazing by cattle and genetic swamping through crossbreeding with domestic buffalo. Therefore, conservation of grass species through control of grazing, and prevention of cross breeding are measures supported by this study. Additionally, site-specific conservation strategies should be adopted, based on identified threats in the study area. [ABSTRACT FROM AUTHOR]

Aims: Vegetation sampling typically involves the use of quadrats to estimate species cover abundance. Such surveys do not generally record small‐scale vegetation patch structure at sub‐quadrat scales. Here we test a simple method to quantify patch structure that complements conventional techniques. We compare the two methods, and analyse metrics derived from small‐scale patch surveys with environment/management data. Location: Northumberland, United Kingdom. Methods: We recorded cover abundance of all species in an upland moor with 1‐m2 quadrats. These were divided into 100 'sub‐quadrats', 10 cm × 10 cm, and the dominant and subdominant species identified. Patch metrics (number, area and shape) for individual species recorded as dominant or subdominant in the subquadrat survey were analysed using multi‐variate generalised linear models with environmental and management data. Sub‐quadrat data were also aggregated for each quadrat, to create species composition data. The two sets of compositional data, from whole‐quadrat and sub‐quadrat aggregations, were compared via Procrustes rotation of ordination scores. Results: Patch number, area and shape for dominant and subdominant species were all significantly affected by soil pH, soil water content, slope and elevation. Effects of proximity to sheep tracks and drainage ditches were less consistent amongst species. Ordinations of vegetation data from conventional and sub‐quadrats were similar, with significant Procrustes R2 of 67% and 70% for dominant and subdominant species respectively. Conclusions: Sub‐quadrat surveys can easily be used to complement existing whole‐quadrat surveys at little cost in time or resources. Their patch metrics can provide additional insights into the environmental and management drivers that may affect the growth of individual plants or clumps, potentially in relation to plant traits, and thus alter the overall community composition. The methods we describe can readily be adapted to other sizes of quadrats and sub‐quadrats in a wide range of vegetation communities. [ABSTRACT FROM AUTHOR]

Sea pens are ecologically important habitats for associated marine organisms, serving as ecosystem engineers in sandy or muddy seafloor environments. In such areas, sea pens can form habitats with high population densities known as "sea pen fields". However, the presence and importance of sea pen fields have not been well studied in shallow waters in East Asia. Here, we report a sea pen field of Virgularia sp. aff. gustaviana in the shallow waters of Ushibuka Marine Park, in the Amakusa Islands of southern Japan. The average colony numbers of the field across all depths (7-20 m) was 10.3 colonies/m2 (live colonies) to 13.6 colonies/m2 (all: live+dead colonies+holes), and the area of the sea pen field was at least 50,000 m2. At a depth of 15 m, the substratum consisted of sand and fallen leaves of terrestrial origin, and the highest sea pen density was observed (averages=17.2 live colonies/m2, =25.8 total (live+dead+holes) colonies/m2). At a depth of 20 m, the substratum consisted of broken shells and rocks and had the lowest density (live colonies: average=0.8 colonies/m2, all: average=1.0 colonies/m2). There were significant differences in colony number of Virgularia sp. aff. gustaviana between the "sand", "sand+leaves", and "broken shells/rocks" substrates. We hypothesize that the strength of the water currents caused by local geographic features and tidal movements produce suitable sedimentation and habitat for this species of sea pen. Therefore, we suggest that preserving the natural coastline is crucial to protect this and other sea pen fields in shallow waters and their benthic marine communities. [ABSTRACT FROM AUTHOR]

Presence–absence sampling is an important method for monitoring state and change of both individual plant species and communities. With this method, only the presence or absence of the target species is recorded on plots and thus the method is straightforward to apply and less prone to surveyor judgement compared to other vegetation monitoring methods. However, in the basic setting, all plots must be equally large or otherwise it is unclear how data should be analysed. In this study, we propose and evaluate five different methods for estimating plant density based on presence–absence registrations from surveys with variable plot sizes.Using artificial plant population data as well as empirical data from the Swedish National Forest Inventory, we evaluated the performance of the proposed methods. The main analysis was conducted through sampling simulation in artificial populations, whereby bias and variance of density estimators for the different methods were quantified and compared.Both for state and change estimation of plant density, we found that the best method to handle variable plot size was to perform generalized least squares regression, using plot size as an independent variable. Methods where plots smaller than a certain threshold were excluded or their registrations recalculated were, however, almost as good. Using all registrations as if they were obtained from plots with the nominal plot size resulted in substantial bias.Our findings are important for plant population studies in a wide range of environmental monitoring programmes. In these programmes, plots are typically randomly laid out and may be located across boundaries between different land‐use or land‐cover classes, resulting in subplots of variable size. Such splitting of plots is common when large plots are used, for example, with the 100 m2 plots used in the Swedish National Forest Inventory. Our methods overcome problems to estimate plant density from presence–absence data observed in plots that vary in size. [ABSTRACT FROM AUTHOR]

Population size estimation is essential in ecology and conservation studies. Aerial photography can facilitate this laborious task with high resolution images. However, in images with thousands of individuals exhaustive manual counting is tedious, slow and difficult to verify. Computer vision software may work under some particular conditions but they are generally biased and known to fail in several situations. The CountEm software is a simple alternative based on geometric sampling. It provides a fast and unbiased size estimation for all sorts of populations. The only requirement is that the discrete objects (e.g. animals) in the target population are unambiguously distinguishable for counting in a still image. Typical relative standard errors in the 5–10% range are obtained after counting ~200 properly sampled animals in about 5 min irrespective of population size. The CountEm ver. 1.4.1 is presented here, which includes a guided mode with a simple software interface. [ABSTRACT FROM AUTHOR]

During long-term monitoring, protocols suitable in the initial context may have to change afterward because of unforeseen events. The outcome for management can be important if the consequences of changing protocols are not understood. In Tuamotu Archipelago atolls, French Polynesia, the density of giant clams (Tridacna maxima) has been monitored for 12 years, but massive mortalities and collapsing densities forced to shift from a line-intercept transects and quadrats (LIT-Q) method to a belt-transect (BT) method. We investigated with a simulation approach the conditions (density, size structure, aggregation of giant clam populations) under which the two methods provided different results. A statistical model relating the BT density to the LIT-Q density was validated using new field data acquired on the same sites with both protocols, on two atolls. The BT method usually provided higher estimates of density than the LIT-Q method, but the opposite was found for very high densities. The shape of the relationship between measurements depended on population size structure and on aggregation. Revisiting with the model the historical LIT-Q densities suggested that densities have been underestimated in the past but previously detected trends in population trajectories remained valid. The implication of these findings for management are discussed. [ABSTRACT FROM AUTHOR]

The structure and composition of forest-floor anuran diversity in the biologically rich rain forests of southern Western Ghats were studied. The study was carried out in the relatively undisturbed rainforest tract of Kalakad-Mundanthurai Tiger Reserve. A quadrat search method was used to estimate species richness and density of the forest floor anuran community. Sample locations were distributed across three distinct watersheds in the reserve. In total, we captured 509 individuals belonging to 20 species. The density of anurans was 149 individuals ha-1 on the rain-forest floor, and along streams their densities exceeded 300 individuals ha-1. Total diversity across all watershed locations was partitioned into within and among watershed components, corresponding to alpha and beta diversity. We found high levels of species turnover (high b diversity) among watersheds, a process that may partly explain the high species richness and large number of endemic anuran species known to occur in the rain forests of the Western Ghats. Anurans density was low and comparable to other studies conducted in South-East Asian forests. A re-examination of the reports on anuran densities revealed that Central American anuran densities were an order of magnitude higher than South Asian and South American anuran densities. Existing hypotheses do not adequately explain the differences in anuran densities in different parts of the world. [ABSTRACT FROM AUTHOR]

We present a collection of variance models for estimators obtained by geometric systematic sampling with test points, quadrats, and n-boxes in general, on a bounded domain in n-dimensional Euclidean space n, n = 1, 2, ... , and for systematic rays and sectors on the circle. The approach adopted − termed the filtering approach − is new and different from the current transitive approach. This report is only preliminary, however, because it includes only variance models in terms of the covariogram of the measurement function. The estimation step is in preparation. [ABSTRACT FROM AUTHOR]

When field plots are prevented from overlapping the boundary of the tract being sampled, the selection probabilities of population elements near the edge differ from what they are when plots are permitted to overlap the boundary. In both situations, selection probabilities of edge elements differ from those of elements located far to the interior of the tract. An analytical expression and graphical depiction of the altered selection probabilities are presented, accompanied by an empirical study of the consequences on estimation for the situation where plots are pulled back from the edge until just tangent with the boundary. [ABSTRACT FROM AUTHOR]