Note OP refers to original breeding population. Of those scores that were most noteworthy, the highest scoring breed for Ne was the Red Ruby Devon while the Luing scored most for percentage change to Ne over the last 5 years. Multiple breeds scored the same for several criteria e. Breeds scoring highly for adaptability and hardiness tended to score highly for conservation grazing demand. The relationship between the different criteria used for scoring breeds is explored using PCA and is plotted using a bi-plot in Figure 5.
Plot A depicts the variable loadings and principal components for all the criteria while Plot B depicts the variable loadings and principal components for the criteria nodes. The variables are plotted as vectors and the observations i. The closer the points are to one another the more similar they are in terms of variable scores. The relative importance of the variables in explaining variation in breed scores is shown based on their distance to the origin, the point where the two axes intersect at zero.
The cosine of the arrows is directly proportional to the correlation between the variables and their length corresponds to the strength of that effect. Figure 5. A bi-plot showing principle components one and two for unweighted breed scores based on the 17 different criteria Plot A and the 6 different criteria nodes Plot B. This suggests breeds scores situated in the positive spectrum of PC1 tend to be those most endangered in situ.
Additionally, a number of variables are strongly positively correlated e. The loadings for PC1 see Supplementary Information S8 show three variables are positively correlated with PC1; diversity, ex situ and traits. Breed scores in the positive spectrum of PC1 therefore scored higher for these factors and were generally less endangered. For PC2, utility and traits had the highest loadings, suggesting they were positively correlated with PC2.
The variable vectors show in situ and diversity are fairly strongly negatively correlated. This suggests breeds with a high diversity score were generally less threatened in situ —a logical finding given that a reduction in population size can cause genetic erosion. Additionally, in situ and ex situ are also negatively correlated, suggesting breeds most at risk ex situ i. None of the criteria nodes were positively correlated, suggesting these five factors are important determinants of breed status. Hypothetical funds to support the initiatives of breed societies were allocated based on a basic formula that considers the breed indicator scores and constraints relating to breed population size.
The hypothetical BIF was allocated according the four weighting scenarios outlined in section Weighting the Criteria. A summary of results from all scenarios are presented in Supplementary Information S8. Figure 6. A,B show the total budget allocation while plots C,D show the pro rata budget allocation based on a per animal constraint.
A summary of results from all scenarios are presented in Supplementary Information S9. The results demonstrate the different weighting scenarios have subtle differences on budget allocation, suggesting structural stability in the model. Moreover, the BIF allocation demonstrates the importance of applying a pro rata budget constraint; highlighting the sensitivity of the model to breed population estimates.
Criteria to monitor breed status can be extensive Eaton et al. Yet, data underpinning such criteria tends to be highly limited for rare breeds reflecting a lack of characterisation and documentation. This study selected case study breeds based partly on the availability of information that could inform the MCDA model and while this may be seen as a limitation of the approach, is an example of the trade-offs apparent between scientific rigour and practicality.
Indeed, during expert workshops a clear trade-off emerged between the desired robustness of approaches to monitor breed status and the need for more realistic, often proximate measures.
GENETIC EROSION OF CROP POPULATIONS IN CENTERS OF DIVERSITY: A REVISION
Perhaps most complex to measure is diversity. The variation that exists within and between breeds can be captured through different metrics including measures of inbreeding usually to monitor genetic drift , introgression to monitor genetic purity , and genetic difference through phylogeny or genomics. Simplistic measures of diversity assessment have been employed by Defra a , b to calculate Ne based on Wright's formula using numerical population data. However, the Wright equation assumes random selection and Poisson distributed progeny sizes, which are unlikely assumptions for most livestock populations Gandini et al.
This means it can produce an estimate of Ne that is higher than would be produced by a calculation using pedigree data Hall, However, such data are not readily available for many native breeds and Verrier et al. Thus, Defra data using Wright's formula a , b were employed for this application but this is a limitation of the analysis. Moving to genetic difference, work by Blott et al.
No studies have yet employed genetic techniques e. Consequently, we used origin as a proxy to estimate the genetic difference across breeds but we acknowledge the limitations inherent in such an approach. The two nodes of marketability utility and traits received lowest weight by experts, partly reflecting limitations of the input data. Additional criteria contributing to utility could include a variable denoting the presence of a rare breed in farm parks to capture cultural value as proximate to public demand for seeing a rare breed.
Such cultural and heritage attributes may be at odds with maximising diversity Lenstra et al. Aside non-use values, criteria to measure direct use-value associated with marketability and consumption can be approximated through product branding and designations, including geographical indicators—e.
The latter may act in perverse ways to concentrate breeding stock if a geographical production constraint is imposed e. There is therefore an explicit need to consider such criteria in a broader framework, as demonstrated here. While option value is promoted for conservation Drucker, ; Hoffmann, the characterisation of productive and adaptive traits in native breeds is often poorly documented Bowles, The limited number of breed traits reported in this work reflects this knowledge gap, suggesting more work is needed to characterise rare breeds through the application of genotyping technologies, including whole genome sequencing Tixier-Boichard et al.
Turning to endangerment, although ex situ conservation was considered least important by stakeholders, it nonetheless serves as an important risk reduction strategy Hiemstra, Yet, our understanding of the legitimacy of current accessions is poor and the criteria we employ merely quantify the material stored, rather than providing broader analytics concerning quality attributes. Developing proxies pertaining to the efficacy of material stored in genebanks would provide more accurate assessment of germplasm safeguards.
For in situ populations, this assessment was limited to breed data reported to Defra and stored in their breed inventory 2. But a range of additional parameters may also reveal risk, including global breed population estimates and demographic trends concerning breeders—e.
Further exploration of these factors is needed in future prioritisation models. Multiple indicators have been constructed for diversity European Environment Agency, ; Villanueva et al. This is perhaps related to the incommensurability of many biological criteria which makes them difficult to compare on common scales without the use of analytical frameworks like MCDA. Although endangerment received the highest overall weight in this indicator, our results demonstrate the inclusion of other criteria nodes is equally important for decision making. This is highlighted where the Vaynol received the highest endangerment score but was ranked 3rd overall because the total benefit of conservation was considered less.
Alternative weighting scenarios reveal how these conservation priorities may change through a focus on different value attributes. This raises broader questions concerning who should assign criteria weights and how periodically they should be reviewed for composite indicators to be robust. To develop this indicator, multiple expert discussions were needed to systematically construct a list of criteria that could be used to measure and report breed status.
However, the PCA suggests some criteria could be omitted in future iterations due to correlation e. Yet, data concerning these criteria are readily available in the UK suggesting little benefit in dropping them from the indicator. The PCA also shows that each criteria node contributes differently to explaining the variance in overall breed scores, suggesting these criteria nodes are actively important for determining breed status. Construction of composite indicators can also reveal relationships between criteria that can be used to test the validity of the results.
For instance, the PCA shows conservation grazing demand for cattle is indeed linked to the traits of grazing animals i. This demonstrates the value of a rare breed is indeed partially linked to their adaptability and hardiness characteristics, a finding often promoted in the literature e. Additionally, we show the collection and storage of germplasm in genebanks is rationalised by in situ endangerment risk both via population metrics and geographical concentration suggesting recommendations outlined by the FAO concerning rationalisation of ex situ collections are indeed being implemented by conservationists.
The preservation of biodiversity, including breed diversity, is hindered by the absence of a workable, cost effective model for determining preservation priorities Metrick and Weitzman, The defining limitation is the lack of an overarching objective to guide investments that has led to untargeted policy interventions seeking to preserve diversity indiscriminately.
Conceptually, this MCDA model suggests differentiated breed support could improve the cost effectiveness of conservation strategies through the distribution of a BIF prioritised by breed indicator scores. Approaches in the UK currently preserve rare breeds through conservation grazing subsidies Natural England, but breeds or species that are not employed for conservation grazing are ultimately under-supplied through such initiatives. More direct support measures, such as those employed in Europe, are therefore necessary to supply more diversity. Differentiated support can also facilitate more targeted interventions, which our results suggest are necessary given that the contribution of each breed to the criteria nodes is heterogeneous.
For instance, the Red Ruby Devon scored high in diversity but relatively low in marketability, suggesting investments in breed promotion may be more effective at improving breed status rather than collecting germplasm, for instance. While unconventional, some countries in Europe e. Piloting such an approach in the UK context may provide valuable insights for constructing a new policy framework.
Although the important role of breed societies has been acknowledged in previous work e. Importantly, the BIF ensures all breeds receive some proportion of funding.
While the proposed BIF can discriminate well between breeds with similar population sizes, it is less effective for breeds with particularly small populations because the pro rata allocation constraint means final funding allocations are very low Alternative approaches to the prioritisation of breeds for conservation e. We argue such approaches are perhaps defeatist and is an ethically pernicious approach to decision making Noss, ; Vucetich et al.
Beyond these factors, this MCDA model does not imply abolishing support for redundant or overlapping diversity; it simply suggests a step change on the supply-side that prioritises preferences exhibited on the demand side. The GPA for FAnGR has stressed the need to construct indicators to better monitor breed attributes and, crucially, develop more systematic conservation responses.
The conservation community currently lacks a harmonised approach to document changes in the status of breed diversity that can be used to inform conservation policies.
BE THE FIRST TO KNOW
We present a framework that at the core considers diversity, marketability and endangerment attributes as a function of conservation need and utilitarian benefit. While the specific sub-criteria to document breed status may require adapting to country-specific circumstances and data provision Porter et al. We argue a committee of national or regional experts could play a pivotal role in adopting such a framework and, if necessary, commissioning surveys to acquire relevant information if not readily available, while noting that many developing countries may only have limited data capture resources.
This highlights the importance of developing proxies where applicable. Indeed, a National Committee could determine the weights of the different criteria as a function of country-specific conditions i. The panel of experts must be diverse enough to represent different backgrounds, breeding associations and professional activities related to the livestock sector Duruz et al. The high-level criteria used to score breeds should remain consistent across countries and we suggest similar to those documented here and in other, earlier works e.
Such a web-based platform could follow a similar approach to that documented by Duruz et al. This study aimed to outline a new methodological framework for monitoring and allocating conservation investments across rare breeds, through a multi criteria approach. This is important because financial resources for species and livestock conservation remain significantly below that required to meet the Aichi biodiversity targets McCarthy et al. The opportunity cost of conservation i. However, Brexit also creates an opportunity to adjust how the UK Government supports the public good properties of rare breeds that span the TEV framework.
Better informed decision-making should consider information on the values of breeds held by stakeholders, the expected benefit to diversity from investments, and the cost of action. Considering these factors through the prioritised allocation of a BIF could better guide investments in FAnGR that promote the longer-term sustainability of breeds. The former relies on empowering breed societies to selectively fund initiatives aimed at improving breed status relative to the multiple values that rare breeds encode.
Prioritising conservation activities is important because extinction risk may take a number of different forms, including introgression, inbreeding depression and genetic drift Berthouly-Salazar et al. A key feature of MCDA is its emphasis on the judgement of the decision-making team and the subjectivity that pervades this can be a matter of concern. While we have attempted to account for such limitations through multiple stakeholder workshops and sensitivity analysis, it should be appreciated that different stakeholder views may produce conflicting results.
Indeed we acknowledge this limitation but stress that participants contributing to this analytical exercise represented a broad spectrum of interests across the FAnGR community. This range of interests, coupled with the expertise of stakeholders, is a key strength of MCDA approaches Dodgson et al.
- Myths of the Oil Boom American National Security in a Global Energy Market.
- Merav Ziv Symposium - Book of Abstracts.
- Understanding crop genetic diversity under modern plant breeding.
- Brothers of Valor (Legends of Valor Book 2)?
- About the Editors | Heredity.
- 1. Introduction.
The way in which questions are posed to elicit criteria weights may also affect outcomes Choo et al. A logical extension to this study would be to validate the approach across different species, while also noting that species specific adjustments to some criteria may be needed. Moreover, the criteria used here reflect circumstances in the UK and alternative criteria may be needed for application in other regions and especially developing countries, where the available information and primary causes of genetic erosion vary Verrier et al. Lastly, this work shows large gaps in information persist for rare and native breeds which impedes characterisation of FAnGR.
These gaps are most prevalent in developing countries where the characterisation of genetic resources is severely limited due to financial and technological constraints. There is a need to define key phenotypic traits and characteristics so FAnGR can be evaluated through comparable data sets, which is important for climate change adaptation Hoffmann, ; Bruford et al. A growing arsenal of increasingly sophisticated genetic technologies are now falling in price e. The research for this paper was funded as part of The Scottish Government Strategic Research Programme —, work package 2.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Efficient Conservation Of Crop Genetic Diversity
Thanks to Ruth Dalton, Libby Henson and Julian Hosking for additional support during the project and RBST for assisting in organisation of workshops and supplying a venue for them to be held. Finally, we thank workshop participants for their assistance during initial scoping, scoring, and weighting exercises.
Abdi, H. Principal component analysis. Wiley Interdiscip. Adem Esmail, B.
- Spatial Planning and Climate Change (Natural and Built Environment Series).
- The Kindness of Strangers (Skip Langdon, Book 6).
- Reviews Editor.
- Researching & Writing a Dissertation: An Essential Guide for Business Students, 3rd Edition?
- Complete Works of Ambrose Bierce.
Multi-criteria decision analysis for nature conservation: a review of 20 years of applications. Methods Ecol. Alderson, L. Breeds at risk: definition and measurement of the factors which determine endangerment. Breeds at Risk: Criteria and Classification. Google Scholar. Belton, V. Springer Science and Business Media. Berthouly-Salazar, C. Uncontrolled admixture and loss of genetic diversity in a local Vietnamese pig breed. Blott, S. Genetic relationships among European cattle breeds. Bojkovski, D.
Supports for local breeds in the European region—an overview. PoljoPrivreda 21, 7— Bowles, D. Recent advances in understanding the genetic resources of sheep breeds locally-adapted to the UK uplands: opportunities they offer for sustainable productivity. Bruford, M. Prospects and challenges for the conservation of farm animal genomic resources, Catalyze Hiview 3: A Starter Guide. Choo, E. Interpretation of criteria weights in multicriteria decision making.
Defra Authors were also approached to gain missing information on the sources of heterogeneity. When relevant data were presented as a time series of repeated measures, I used data from the final time point to calculate effect sizes. I used this measure of effect size for experimental studies because many of these studies presented effects for only two levels of genetic diversity; it would not have been possible to fit a correlation to these data.
These were then used to calculate the standardized mean difference for each unique pairwise combination of genetic diversity levels. For example, a study that created experimental communities with one, three and six clones of the focal species would yield three effect sizes contrasting ecological responses for three vs.
Standardized mean difference effect sizes d and effect sizes based on correlation coefficients z r were analysed separately. To take account of effect size precision, I fitted measurement error variance values mev about the effect sizes as a set of variance components. The sources of heterogeneity explanatory variables were fitted as fixed effects. I fitted both categorical and continuous sources of heterogeneity individually i. This was because imbalance meant that some pairs of these variables had combinations of factor levels containing no effect sizes.
For each of these models, I excluded effect sizes where the corresponding source of heterogeneity predictor contained unknown or missing information. Heterogeneity between putatively adaptive and neutral effect sizes in this data set was assessed, and then, the adaptive effect sizes were removed to allow analyses that focussed on neutral genetic diversity. Default settings were used for priors for the explanatory variables.
The extent of chain mixing was assessed using autocorrelation analysis, effective size statistics and by plotting and inspecting MCMC samples. I checked for publication bias within the effect size data sets using enhanced funnel plots implemented in the R package Metafor; Viechtbauer Effect sizes for species diversity, stocks and fluxes of elements and nutrients and ecosystem stability were not consistently different from zero effect sizes for fluxes and stability were positive, but were not significantly so; Fig.
In studies that manipulated genetic diversity, effect sizes were not significantly predicted by the mismatch in sampling scale between the focal plant collection sites and the sizes of experimental plots used Fig. However, there was an apparent trend in the variance of effect sizes; studies with more closely matched sampling and experimental spatial scales had more variable effect sizes Fig.
The studies with greatest disparity in sample scales and least variable effect sizes used Arabidopsis thaliana as the focal plant species. However, studies that used agricultural cultivars to form focal plant populations showed lower effect sizes that did not differ significantly from zero Fig. Neutral genetic diversity showed no overall consistent association with ecological structure at either the community or ecosystem levels, or for any component type of ecological response Fig.
The largest effect sizes were observed in situations where populations of the focal species both varied in spatial extent and were isolated from each other as effective demographic islands. Variation in effect sizes among studies that manipulated adaptive variation accounted for between If there had been publication bias towards studies showing a positive relationship between genetic diversity and measures of ecological structure, then we would expect to see a deficit of statistically insignificant negative effect sizes i.
Forest plots of effect size against effect size standard error showed little evidence for asymmetry in the distribution of effect size estimates that would indicate the presence of publication bias Fig. These analyses were novel in allowing an assessment of the strength and direction of the relationship between genetic diversity and ecological structure.
They also enabled a first comparison of ecological effects associated with adaptive and neutral genetic diversity. My results show that ecological responses stemming from adaptive genetic diversity are significantly positive at both community and ecosystem levels of organization. In contrast, the pooled effect size for neutral genetic diversity did not differ consistently from zero. However, these latter effect sizes were also consistently positive under certain demographic conditions.
Neither species diversity e. Shannon diversity nor evenness reacted in a similar way Fig. Collectively, these results suggest that genetically controlled phenotypic changes within focal plant populations drive differences in the richness, but not the diversity or evenness of the communities that depend on them. This relationship may be explained by a positive effect of plant genetic diversity on plant population productivity Fig.
However, specialization of the dependent species on different plant genotypes could also be important in enhancing their species richness on genetically diverse plant populations Hutchinson Thus, the ecological effects of genetic diversity were weaker within communities occupying a single trophic level than they were for systems where dependent communities respond to a dominant foundation species. The component of ecosystem functioning most strongly associated with adaptive genetic diversity was biomass productivity Fig.
Other elements of ecosystem functioning, such as stocks and fluxes of nutrients and elements, were not significantly predicted by genetic diversity. My results demonstrated an ecological effect of adaptive diversity per se Fig. The results presented in this review do support the idea that management to avoid low levels of genetic diversity in foundation species could bring benefits to their productivity and to the species richness of dependent communities e.
However, avoidance of the very low levels of genotypic diversity typically investigated in the community genetics literature e. In addition, our understanding of these effects is generally limited to fine spatial scales e. Furthermore, any benefits to species richness could be accompanied by an equivalent cost to species evenness Fig.
The greater richness of arthropod species was attributed to a greater total resource for arthropods living on genotypic mixtures, while the decrease in diversity and evenness was ascribed to the disproportionate increase in abundance of a dominant arthropod species Plagiognathus politus on genotypic mixtures, relative to monocultures. Therefore, modifications to arthropod community structure evenness could result in a selective feedback to the focal plant population that subsequently destabilizes initial benefits to arthropod species richness.
The message for conservation is that we still know too little regarding the impacts of genetic diversity on ecological structure to recommend management for high genetic diversity as a means to enhance the diversity or functioning of ecosystems. These effect sizes are also small in comparison with average effect sizes observed more widely in the evolutionary ecology literature range in average effect sizes d , 0. In summary, my results show that ecological responses to adaptive genetic diversity are smaller and less predictable than the previous analysis had suggested.
This heterogeneity may also arise because it is extremely difficult to predict how the effects of intraspecific genetic diversity will influence associated species that are either highly dominant within their trophic level or are ecosystem engineers and that modify the environment for a dependent community.
In short, part of the unpredictability of the responses of dependent communities to genetic diversity within a foundation species arises because it is difficult to anticipate precisely how ecological interactions within associated or dependent communities will play out. A key objective of this review was to determine whether neutral and adaptive measures of genetic diversity have comparable relationships with ecological structure. In contrast to adaptive genetic diversity, neutral genetic diversity showed no overall significant association with ecological structure Fig.
Unlike adaptive genetic diversity, neutral genetic variation cannot drive changes in community structure and functioning via the phenotypes of constituent individuals. In spite of this, however, neutral genetic diversity and ecological structure were correlated under certain circumstances. Taken together, my results indicate that the measures of adaptive and neutral genetic variation currently used in community genetics studies should not be viewed as ecologically interchangeable.
Hence, studies that utilize neutral genetic diversity should not attempt to make a general test of relationships between genetic diversity and ecological structure cf. These studies are only likely to demonstrate positive effects when the demographic conditions stated above are met. Similarly, positive ecological responses to adaptive genetic diversity are not a foregone conclusion; responses will vary depending of the type of ecological response in focus Fig.
At this point, it is important to note that studies that consider neutral genetic diversity often use a different design to those that employ adaptive genetic diversity. It is possible that surveys of adaptive genetic diversity that followed an observational design would detect relationships with ecological structure that are comparable with those reported here for neutral genetic diversity.
However, these studies have yet to document the extent of genetically controlled phenotypic differentiation between clonal individuals observed in the field. To close this knowledge gap, additional studies are needed that survey levels of genotypic variation or genetic variance across natural populations in the field e.
The results presented in this review enhance our understanding of when and how genetic diversity and measures of ecological structure are likely to be associated with each other. However, the review also suggests several opportunities for further empirical studies targeted at areas where the evidence base is currently more limited. The results illustrate how this relationship varies between the two types of genetic diversity commonly used in community genetics studies among different ecological response measures and in different ecological and spatial contexts.
Free Preview. Buy eBook.
Related Efficient Conservation Of Crop Genetic Diversity: Theoretical Approaches And Empirical Studies
Copyright 2019 - All Right Reserved