Author: isabel

Science is suffering a reproducibility crisis that is associated with the impacts of many biases on the outcomes of scientific research. Among these biases, those caused by the properties of the human mind are seldom appreciated by scientists working in the fields of ecology and environmental science. In the present paper, we asked to what extent studies of spatial patterns in insect herbivory suffer from various biases. We answered this question by comparing the outcomes of studies that have used methods of data collection which were potentially prone or not prone to particular biases. This comparison was conducted by meta-analysis of 125 scientific paper containing data on plant losses to insects in different environments. We found that biases emerged when scientists non-randomly selected one or a few plant species when assessing community-wide herbivory or when  scientists did not apply randomisation procedures when selecting leaves for measurements of herbivory. The reported effects were greater when the persons that collected samples and measured leaf damage were aware of the research hypothesis or sample origin than when the collectors/measurers were blinded to these factors. We conclude that our present-day knowledge regarding differences in insect herbivory between environments is considerably biased due to an unconscious tendency of researchers to find support for their hypotheses and expectations, as this generally leads to overestimation of the effects under study. These cognitive biases can be avoided by (i) applying randomisation procedures in sampling and (ii) blinding of the research hypotheses and sample origins. These measures should be seen as obligatory; otherwise, accumulation of biased results in primary studies can ultimately lead to false general conclusions in subsequent research syntheses.

Reference: Zvereva, E., Kozlov, M.V. (2019) Biases in studies of spatial patterns in insect herbivory. Ecological Monographs.

https://doi.org/10.1002/ecm.1361

You can read the full article here.

Picture: A tropical herbivore (bush cricket) mimicking leaf with insect damage (photo: V. Zverev, 2019)

In this study, Lily Zhao and colleagues use carbon and nitrogen stable isotope analysis to infer summer dietary differences of Svalbard reindeer found in different valleys. Through regression models, they were able to predict how reindeer dietary intake depend on climatic variables, with July temperature being the strongest predictor.  Using stable isotope analysis, they found robust dietary differences between and within reindeer populations separated by less than 50 km in the proportions of non-mycorrhizal vascular plants and mosses. These localized differences in forage availability related to the onset of the growing season as shown through the onset of the growing season in each valley through time. Overall, this publication provides evidence that carbon and nitrogen stable isotope analysis can detect dietary shifts over short time periods and at fine spatial scales, which means that isotopic studies can help monitor herbivory in the changing Arctic.

Reference: Zhao, L.Z., Colman, A.S., Irvine, R.J. et al. (2019) Isotope ecology detects fine-scale variation in Svalbard reindeer diet: implications for monitoring herbivory in the changing Arctic. Polar Biology.

https://doi.org/10.1007/s00300-019-02474-8

You can read the full article here.

Picture: Couple of Svalbard reindeer females and calf in April
(photo: Justin Irvine)

This paper highlights the strong relationship between forage phenology in alpine areas and the dynamics of plant consumption by alpine herbivores.

In this study, Johan Espunyes and colleagues assessed the seasonal changes in diet composition of Pyrenean chamois (Rupicapra p. pyrenaica) and its relationship to primary production trends in a Mediterranean alpine environment. Moreover, the diet composition in two populations with contrasting livestock pressure was compared in order to study the effect of sheep flocks on the feeding behaviour of chamois. Using monthly sampling during three consecutive years, they showed an intra-annual variation in the diet composition of Pyrenean chamois and demonstrated a strong relationship between plant consumption dynamics and phenology in alpine areas. This study could not detect differences between both chamois populations despite the presence of sheep flocks in only one area. However, the detection of a shift in the diet of chamois in both areas after the arrival of high densities of multi-specific livestock suggest a general livestock effect. In conclusion, Pyrenean chamois are well adapted to the variations in the seasonal availability of plants in alpine habitats but could be disturbed by the seasonal presence of livestock. Due to the key plants in their diet, it is suggested that population management programmes should focus on the preservation of mixed grasslands composed of patches of shrubs and herbs. The effects of climate change and shrub expansion should be studied as they may potentially affect chamois population dynamics through changes in habitat composition and temporal shifts in forage availability.

Reference: Espunyes, J., Bartolomé, J., Garel, M., Gálvez-Cerón, A., Fernández Aguilar, X., Colom-Cadena, A., et al. (2019) Seasonal diet composition of Pyrenean chamois is mainly shaped by primary production waves. PLoS ONE 14(1): e0210819

https://doi.org/10.1371/journal.pone.0210819

You can read the full article here.

Picture: Pyrenean chamois (Rupicapra p. pyrenaica) (photo: Johan Espunyes)

This paper highlights the impacts that hyperabundant herbivores can have on tundra habitats and sympatric-nesting species.

In this study Scott Flemming and colleagues examined how grazing and grubbing by a lesser snow (Chen caerulescens) and Ross’ goose (Chen rossii) colony influenced the availability and quality of habitats selected by eight different shorebird species in the eastern Canadian Arctic.

During randomly selected habitat surveys conducted at three study sites that varied in goose presence and influence (High, Moderate, Low), they estimated the proportional contribution of difference habitat types, amount of lateral concealment provided, and measured the height of various cover types (e.g. grasses, rock, willow). They also searched for shorebird nests at plots situated within each study site and conducted habitat surveys at all shorebird nests. At random sites, the availability of sedge meadow was negatively related to increasing goose influence while the availability of exposed sediment was positively related. The amount of lateral concealment provided by cover types was also negatively related to increasing goose influence and this trend was largely driven by changes in sedges and grasses. The densities of cover-nesting shorebirds (species that prefer nesting in sedge meadow habitat that provides high concealment) were lowest at the High goose influence site and highest at the Low goose influence site. Furthermore, cover-nesting shorebirds at the Moderate goose influence site selected nest sites with less sedge meadow and concealment than those at the Low goose influence site. Many shorebirds select nesting habitats that provide them with high concealment from predators; the authors therefore suggest that goose-induced habitat alteration may not only be influencing shorebird nest site selection, but also nest survival probability and recommend more research in this field.

This study demonstrates how habitat alteration by hyperabundant populations of geese can influence nest site selection of Arctic shorebirds at multiple scales and highlights the importance of goose population management.

Reference: Flemming, S.A., Nol, E., Kennedy, L.V., Smith, P.A. (2019) Hyperabundant herbivores limit habitat availability and influence nest site selection of Arctic‐breeding birds. Journal of Applied Ecology
https://doi.org/10.1111/1365-2664.13336

You can read the full article here.

Picture: Foraging Lesser snow goose (Chen caerulescens) in Nunavut (photo credit: Scott Flemming, Trent University)