Skip to main content Accessibility help
Hostname: page-component-564cf476b6-dr4jh Total loading time: 0.485 Render date: 2021-06-19T20:08:47.157Z Has data issue: true Feature Flags: { "shouldUseShareProductTool": true, "shouldUseHypothesis": true, "isUnsiloEnabled": true, "metricsAbstractViews": false, "figures": true, "newCiteModal": false, "newCitedByModal": true, "newEcommerce": true }

The first rapid forest inventory and resource use assessment of Dashtijum Nature Reserve, Tajikistan: a mixed methods approach

Published online by Cambridge University Press:  01 June 2020

Fred Pilkington
Department of Life Sciences, Imperial College London, Silwood Campus, Ascot, Berkshire, UK
Minerva Singh
Centre of Environmental Policy, Imperial College London, London, UK
Vicky Wilkins
Fauna & Flora International, Cambridge, UK
Colin Clubbe
Royal Botanic Gardens, Kew, Richmond, UK
E-mail address:
Rights & Permissions[Opens in a new window]


There is critical need for more information on the status of forests in Central Asia, to inform conservation management. Here we assess the status and use of the globally important, threatened walnut–fruit forests of Dashtijum Nature Reserve, Tajikistan. We use a mixed methods approach combining a semi-structured interview based socio-economic survey with statistical analysis. Thirty-three tree species were recorded, with Rosaceae the most common family. The mean basal area of the forest was 20.8 m2 per ha and most tree species exhibited stable population structures. Resource use was prevalent throughout the community; most households harvested at least one fruit or nut species (78%), firewood (88%) and grazed livestock (85%). Most respondents noted declines in availability of fruits, nuts and firewood and reported a decrease in the overall area and quality of grazing. Statistical analysis and machine learning identified that distance to nearest settlement significantly affected forest total basal area, tree species richness and the Shannon index, a metric of diversity sensitive to rare species. Our results suggest that conservation management by the government and NGOs is partly effective but there is a need for increased enforcement of grazing restrictions, to prevent further forest degradation.

Copyright © Fauna & Flora International 2020


Tajikistan is the poorest country in Central Asia (Zhupankhan et al., Reference Zhupankhan, Tussupova and Berndtsson2017), with one of the highest levels of livelihood and food insecurity risks (Sutton et al., Reference Sutton, Block and Srivastava2009). Almost half of Tajikistan's population live rurally and there is intense pressure on natural resources (Ajwad et al., Reference Ajwad, Hut, Abdulloev, Audy, Kataoka, Larrison and Torracchi2014). During the 20th Century, Tajikistan's population increased six-fold (Safarov, Reference Safarov2003) resulting in prolonged unsustainable exploitation of its limited natural resources (Wolfgramm et al., Reference Wolfgramm, Stevenson, Lerman, Zähringer and Liniger2011) and widespread land degradation (Lerman, Reference Lerman2011). To prevent further environmental degradation and help alleviate poverty, management options, including expansion of land holdings and the reforestation of key areas, have been proposed (Lerman, Reference Lerman and Squires2012). With effective management, ecosystem services provided by forests, such as carbon sequestration and biodiversity conservation, can be substantially enhanced (Safarov et al., Reference Safarov, Novikova and Shermatov2014).

Sustainable forest management encompasses the provision of all forest ecosystem goods and services in perpetuity, including the supply of non-timber forest products (NTFPs; Lindenmayer et al., Reference Lindenmayer, Margules and Botkin2000). Tajikistan's forests have experienced almost a century of overexploitation because of poor management (Maselli & Rahim, Reference Maselli, Rahim and Squires2012). Forest cover declined from 25% in the 1920s to < 3% in 2007, making Tajikistan the most deforested country in Central Asia (FAO, 2007). Efforts to encourage land restoration have been limited by poor decision making, combined with chronic overgrazing and land tenure issues (FAO, 2010; Halimova, Reference Halimova and Squires2012). A key challenge facing the government's Forestry Department is the lack of reliable data on forest distribution and composition (Mislimshoeva et al., Reference Mislimshoeva, Herbst and Koellner2016).

Five mountain ranges converge in Tajikistan, creating a global biodiversity hotspot of which the forest community and its associated tree species are essential elements (Squires & Safarov, Reference Squires, Safarov, Ning, Rawat, Joshi, Ismail and Sharma2013; Orozumbekov et al., Reference Orozumbekov, Cantarello and Newton2014). The walnut–fruit forests of Central Asia are of special scientific interest as they have been identified as the source of numerous domesticated tree species (Beer et al., Reference Beer, Kaiser, Schmidt, Ammann, Carraro, Grisa and Tinner2008). Tajikistan's forest community contains 10 fruit and nut species, five of which are categorized as threatened on the IUCN Red List (Squires & Safarov, Reference Squires, Safarov, Ning, Rawat, Joshi, Ismail and Sharma2013; Supplementary Table 1). These ancestral species are harvested for fruits, nuts and firewood during May–November and play a vital role in the region's culture and livelihood security (FAO, 2010). Tajikistan's forests, and the communities that rely on them, are the focus of monitoring research, protection efforts to facilitate natural regeneration, reforestation of native species, and capacity building combined with outreach with local Forestry Management Units (Darwin Initiative, 2017; FFI, 2018).

This study is the first forest inventory conducted in Dashtijum Nature Reserve and only the second conducted in Tajikistan since 1943 (Nowak et al., Reference Nowak, Nowak and Nobis2011). Our aims were to evaluate the magnitude and spatio-temporal pattern of anthropogenic disturbance and forest resource use within the Reserve and to identify drivers of biotic diversity using regression and machine learning methods.

Study area

We conducted our research during May–July 2017 in the 350 km2 Dashtijum Nature Reserve, Khatlon province, Tajikistan (Fig. 1). The region is characterized by a temperate climate with a mean annual temperature of 11 °C, total annual precipitation of 500–1,500 mm and relic broadleaf montane deciduous forests (Safarov, Reference Safarov2003). The c. 150 km2 study area has a mean altitude of 1,445 ± SD 265 m (range 1,014–2,307 m). There is a mixed economy in the Reserve, with NTFPs harvested both for subsistence and for cash income (World Bank, 2004). Animal husbandry is equally important, with most of the population owning a combination of cows, equids and sheep (Squires & Safarov, Reference Squires, Safarov, Ning, Rawat, Joshi, Ismail and Sharma2013).

Fig. 1 Location of the 40 survey plots, and the four villages where people were interviewed, in Dashtijum Nature Reserve, Tajikistan.



We identified areas for forest surveys through discussions with Forestry Department employees, with a bias towards larger forested areas (Gordon et al., Reference Gordon, Hawthorne, Reyes-García, Sandoval and Barrance2004). We used 0.25 ha plots to ensure adequate representativeness as the heterogeneity and mean patch size of the forest was unknown (Kenkel & Podani, Reference Kenkel and Podani1991). To maximize proportionality between the number of samples and the forest size (Southwood & Henderson, Reference Southwood and Henderson2015), up to three plots were surveyed within each forest area, with a minimum of 300 m between plots. Within each survey area, a random walk technique was employed to select the north-west corner of each plot (Dytham, Reference Dytham2015).

A tree was defined as having ≥ 30 cm circumference at breast height and ≥ 1.5 m height (breast height = 1.3 m from ground level; Newton, Reference Newton2007). We measured the circumference at breast height of every tree within each of 40 plots and converted to diameter (d), which was used to calculate basal area (g) using

(1)$$g\lpar {\rm m}^2\rpar = \displaystyle{{\pi d^2} \over {\lpar 4 \times 10\comma 000\rpar }}$$

from Husch et al. (Reference Husch, Beers and Kershaw2003).

We recorded presence of hay harvesting along with a visual estimate of canopy cover using a 5-point scale (1: very open; 5: very dense; Korhonen et al., Reference Korhonen, Korhonen, Rautiainen and Stenberg2006). Additionally, we recorded the livestock disturbance indicators proposed by Reimoser et al. (Reference Reimoser, Armstrong and Suchant1999) of browse line and trampling incidence using a 4-point scale (high, medium, low, absent). Livestock dung, sapling and seedling density were measured within a 5 × 5 m subplot located randomly within each plot. Species-specific dung counts were combined into a measure of potential threat using the grazing conversion metrics 0.6 for cattle, 0.58 for equids and 0.1 for sheep (Chilonda & Otte, Reference Chilonda and Otte2006). We recorded plot location and altitude using a GPS and measured slope angle using a clinometer. Distance to nearest settlement, known to influence anthropogenic disturbance positively and biodiversity negatively (Ahrends et al., Reference Ahrends, Burgess, Milledge, Bulling, Fisher and Smart2010), was measured post-hoc using the Open Layers plugin on QGIS 2.18.2 (QGIS Development Team, 2020).

Stakeholder interviews

We conducted semi-structured interviews to gain insight into the spatio-temporal patterns of forest resource use, and the economic and cultural importance of these forests. Forest resource use encompassed NTFP harvesting, timber extraction and forest grazing practices. The interview (Supplementary Material 1) was based primarily on a set of questions following Cantarello et al. (Reference Cantarello, Lovegrove, Orozumbekov, Birch, Brouwers and Newton2014), refined through discussion with Forestry Department employees and Fauna & Flora International staff. The interview was subsequently translated into Tajik, piloted with three members of the local populace, and edited to ensure its contextual accuracy and retention of intended meaning. Resource use was shared between men and women, and therefore we aimed for a 50:50 split between male and female respondents. For comprehensive coverage of the four largest villages within the Reserve, we divided the total number of households (n = 280) by the desired number of interviews (n = 40) and therefore visited every 7th household. Interviews were with one representative from each household and only took place with free, prior and informed consent.

Data analysis and modelling

The Shannon (Shannon & Weaver, Reference Shannon and Weaver1949) and Simpson (Simpson, Reference Simpson1949) indices both combine two properties of diversity: richness (the number of species), and evenness (the proportion of each species). We used Shannon's index as it is more sensitive to changes in the abundance of rare species (Peet, Reference Peet1974), of which the survey area has a considerable number, whereas Simpson's index is more responsive to dominant species (Nagendra, Reference Nagendra2002). To complement Shannon's index, species richness (McIntosh, Reference McIntosh1967) was also used in the modelling. A species accumulation curve was analysed visually to judge sampling adequacy. Stem diameters, grouped into 5 cm classes, were analysed visually to infer species-specific age distributions. The Importance Value Index of each tree species was calculated as the mean of relative frequency, relative density and relative dominance, all expressed as percentages (Curtis & McIntosh, Reference Curtis and McIntosh1950). Relative frequency was calculated by dividing the frequency (the number of quadrats with the species present) by the sum of all species frequencies. Relative density calculations followed the same process, using the number of trees per ha. Relative dominance was calculated using total basal area divided by the sum of the basal area of all species.

We used a stepwise multiple linear regression to identify the most significant variables (Supplementary Table 2) influencing the variation in total basal area and Shannon index (Badreldin & Sanchez-Azofeifa, Reference Badreldin and Sanchez-Azofeifa2015). A generalized linear model was used for modelling species richness because the count data exhibited a Poisson distribution (Lovett et al., Reference Lovett, Marshall and Carr2006; Fletcher & Fortin, Reference Fletcher and Fortin2018).

Random Forest (Breiman, Reference Breiman2001), a machine learning algorithm, was used to quantify the influence of each predictor variable on the three response variables (species richness, Shannon index and total basal area). Marginal response curves (derived using the Random Forest algorithm) were used to identify the impact of individual predictor variables (such as those relating to disturbance) on the four response variables. The influence of qualitative predictor variables on the response variables was identified and quantified using Kruskal-Wallis tests (Singh et al., Reference Singh, Evans, Friess, Tan and Nin2015).


Survey data

In total, we recorded 1,380 trees representing 33 species (Table 1). The species with the highest Importance Value Index was Prunus sogdiana (21.6%) followed by Morus alba (17%) and Juglans regia (10.1%; Table 1); 48% of relative dominance was explained by P. sogdiana and M. alba (Supplementary Table 3). The mean basal area of the forest was 20.8 ± SD 1.3 m2 per ha and the mean density of saplings and seedlings were 1,320 ± SD 101 per ha and 4,397 ± SD 248 per ha, respectively. Of the 17 most numerous species, accounting for 93% of individuals, 13 species had type I and four species had type II stem diameter distributions (see Supplementary Fig. 1 for the 10 most utilized species). The species accumulation curve (Supplementary Fig. 2) almost reached an asymptote, indicating that most species were found with the sampling effort employed.

Table 1 Stand structure of the forests of Dashtijum Nature Reserve, with mean tree basal area, tree density and Importance Value Index (IVI; see main text for details of calculation) of all tree species found.

We interviewed an equal number of men and women. All interviewees were engaged in at least one type of resource extraction. The mean number of species gathered was 4 ± SD 2.68 (range 0–8) and 78% of households collected at least one species of fruit or nut from the forests; P. sogdiana (73%), Malus sieversii (68%) and Crataegus pontica (65%) were the most commonly harvested species (Fig. 2).

Fig. 2 Per cent of 40 households that harvested nine fruit or nut species from the forests in Dashtijum Nature Reserve (Fig. 1).

Most households reported an increase in the travel time to the fruit and nut collection sites (55%) and a decrease in the total amount of fruit and nuts collected (65%) compared with 5 years previously (Fig. 3). All participants noted that most tree species were either very important or important for the health of the forest ecosystem.

Fig. 3 Responses from 40 household interviews regarding compositional and resource changes within the forests of Dashtijum Nature Reserve (Fig. 1) over the previous 5 years.

For firewood, 88% of households harvested it and 12% purchased it locally. Of those harvesting firewood, all commented that it was economically important or very important and 78% stated it was for subsistence. The majority of households collected firewood every day from any tree species (48%) or from two or more species (43%); 92% of sample plots had branches cut and 47% of all trees surveyed had branches cut for firewood harvesting. Households reported a decrease in the total amount (83%; Fig. 3) and an increase in the travel time to arrive at harvesting sites (83%; Fig. 3) over the previous 5 years. All respondents noted an increase in the effectiveness of management by the Forestry Department and widespread community involvement in enforcing the ban on harvesting of live branches. Most households (98%) reported no collection of timber and those that did stated they did so rarely and for subsistence purposes only. Most households (80%) noted an increase in the number of young trees (Fig. 3).

Most households (85%) owned livestock and 53% admitted to grazing their livestock in the forest, despite it being illegal. Evidence of disturbance by livestock was found in 90% of plots and few plots contained seedlings (28%) and saplings (15%). The total area and quality of grazing in the forest was reported to have decreased by 88% and 70%, respectively (Fig. 3), with a similar level of unanimity (75%) on the decreased quality of pastureland grazing.

Disturbance modelling

Forward selection with ordinary least squares regression identified the variables distance to nearest settlement and slope as the most parsimonious model that best explained the variation in total basal area. The overall model was significant (P < 0.05) and explained a moderate amount of variance in the response variable (R 2 = 0.20, F = 5.28; Supplementary Table 4). A strong positive association was found between distance to settlements and total basal area (r = 0.42).

Random Forest identified and ranked the variables in terms of their importance in influencing the response variables. The variables low trampling and distance to the nearest settlement were key influencers of both species richness and Shannon's index (Table 2). An increase in distance to the nearest settlement resulted in an increase of all three response variables (Shannon's index, species richness and total basal area; Supplementary Fig. 3). For both measures of species diversity, the relationship with dung count was strongly negative, whereas for total basal area it was positive.

Table 2 The importance (i.e. how important the predictor variables in a given model are in explaining the response variable), of various predictor variables in influencing species richness, the Shannon index and total basal area, computed using Random Forest models.


This study provides a rapid assessment of forest and resource use in Dashtijum Nature Reserve. The forest community comprised 33 species, similar to a comparable study in Kyrgyzstan (Cantarello et al., Reference Cantarello, Lovegrove, Orozumbekov, Birch, Brouwers and Newton2014), and the asymptotic nature of the species accumulation curve indicates an adequate survey effort, a satisfactory completeness of the inventory (Moreno & Halffter, Reference Moreno and Halffter2000), and therefore a reliable estimation of species number.

Of the 17 species diameter distributions analysed, 11 had reverse J-curves which is indicative of balanced uneven age structures (Keyser & Loftis, Reference Keyser and Loftis2013) and stable, self-maintaining populations (Peters, Reference Peters1994) indicative of a healthy community. Two species, C. pontica and P. sogdiana, had steep reverse J-curves (i.e. type I) of diameter distribution, indicating an absence of large individuals and implying intense exploitative cutting in the 20th century (Wolfgramm et al., Reference Wolfgramm, Stevenson, Lerman, Zähringer and Liniger2011). The stem diameter distribution curves of M. alba and J. regia, and to a lesser extent M. sieversii and Amygdalus bucharica, exhibited Type II size-class structures symptomatic of irregular seedling establishment (Peters, Reference Peters1994). These four species are economically important and therefore overharvesting appears to be the most likely cause of these distribution shapes as they are specific to four species, two of which we found to be heavily harvested. However, other factors such as direct physical damage to seedlings, a lack of pollinators or a lack of dispersal agents could also have influenced the shape of these size-class structures (Bond, Reference Bond1994).

We observed a heavy reliance on firewood, consistent with previous research in the area (FAO, 2007). Respondents noted a decrease in the availability of firewood over the previous 5 years and an increase in travel time to harvesting locations, indicating this extraction may be unsustainable. However, the Forestry Department is implementing an intensive reforestation programme to combat 20th century exploitation, planting c. 15,000 native trees per year (M. Yorahmadov, pers. comm., 2017). The impact of this was noted by most households (Fig. 3). The widespread knowledge of fines, adherence to the law and enforcement as a social norm indicate the cutting of green branches is taboo within the community and comes with the threat of social ostracization if broken (Lade et al., Reference Lade, Tavoni, Levin and Schlüter2013).The power of taboos to promote the harvest of wild species has been noted in traditional societies worldwide (Colding & Folke, Reference Colding and Folke2001), especially when coupled with small scale comanagement (Jones et al., Reference Jones, Andriamarovololona and Hockley2008).

Illegal timber harvesting was noted by only 2% of the households surveyed, indicating that laws banning timber harvest may be widely respected. Large fines for illegal timber harvesting were identified as motivation for compliance by a high proportion of respondents. Together with this effective disincentive, we found the demand for timber to be sated by high domestic production of predominately Populus spp., as in temperate regions globally (Ball et al., Reference Ball, Carle and Del Lungo2005).

There were high levels of livestock ownership within the Reserve, consistent with previous reports (Squires & Safarov, Reference Squires, Safarov, Ning, Rawat, Joshi, Ismail and Sharma2013). The total area available for grazing in the forest has declined as a result of land privatization (U. Gulamadshoev, pers. comm., 2017), and the quality of grazing in both forest and pasturelands has declined, indicating unsustainable use. This is reinforced by two findings: most households admitted to grazing livestock illegally in the forest, and machine learning identified low trampling as a strong influence on both species richness and Shannon's index. Illegal grazing demonstrates a disregard for grazing laws, which we believe to be systemic throughout the community, resulting from family ties between those protecting the forest and those grazing the livestock. This is compounded by the cooperative grazing management system in which a guard could have to impose a fine on somebody grazing a herd that includes the guard's livestock.

The greater species diversity and total basal area at increased distances from human settlements indicate that anthropogenic disturbance has a negative impact on this ecosystem, suggesting that continued unsustainable resource use will lead to further environmental degradation. The strong positive relationship of total basal area and the negative relationships of both measures of diversity with dung count may indicate that dung provides nutrients that enhance the growth of disturbance tolerant species, and that grazing, for which dung count is a proxy, causes a decline in species diversity. The dominance of P. sogdiana suggests it may be a pioneer species, as reported by Vera (Reference Vera2000), thriving in the disturbed landscape despite being preferentially harvested for firewood. With its rapid growth rates in high light conditions, the dominance of P. sogdiana may be maintained by high levels of disturbance supressing less tolerant species (sensu Bongers et al., Reference Bongers, Poorter, Hawthorne and Sheil2009). The high recruitment rate of P. sogdiana despite significant disturbance may confer resilience to the functioning of this ecosystem (Grime, Reference Grime1998).

Our data provides a baseline in a chronosequence for the analysis of temporal trends in disturbance and biodiversity (Newton & Echeverría, Reference Newton, Echeverría, Coomes, Burslem and Simonson2014). Future work could employ Rapid Rural Assessment (Chambers, Reference Chambers1992) to cover a greater percentage of households than in our study, to provide deeper insight into the sustainability and social heterogeneity (Waylen et al., Reference Waylen, Fischer, McGowan and Milner-Gulland2013) of NTFP use and livestock management.

Tajikistan suffers from lack of an effective legal framework, with unregulated transhumance and high levels of forest grazing being particularly damaging to forest health (Mislimshoeva et al., Reference Mislimshoeva, Herbst and Koellner2016). These factors, combined with unsustainable NTFP extraction, present significant challenges for the Forestry Department. Certain positive aspects of current government-led managerial practices could be applied to improve the regulation of local forest use, to increase the sustainability of the system. For example, a fine coupled with an enforcement scheme reinforced by social norms, as is used to control the harvesting of green branches and timber, could be initiated to decrease the prevalence of livestock grazing in the forest. The management successes to date have been, in part, a result of well-defined laws reinforced by cultural taboos with respect to resource extraction and significant autonomy to enact and enforce these laws. Other factors positively influencing management, as detailed by Ostrom (Reference Ostrom1999), are the relatively small sizes and homogeneity of communities, who trust the local authority to deliver appropriate management and restoration. Our findings will augment stakeholder engagement strategies in Dashtijum Nature Reserve and other protected areas within Tajikistan and, specifically, will continue to inform reforestation site selection for 400,000 native trees, to enhance forest diversity and resilience to climate change (D. Gill, Fauna & Flora International, pers. comm., 2020; Darwin Initiative, 2017).


This work was supported by the Global Trees Campaign, the Tropical Agricultural Award Fund and the Darwin Initiative. We thank Andrew Knight and Margaret Pasquini for help with survey design, Dilovar Sharipov and Bakhtovar Nurov for assisting with logistics, translation and community liaison, Ubayd Gulamadshoev and Muqaddas Milikbekova of Fauna & Flora International for support and advice throughout the project, Sam Gregory and Harriet Branson for digital mapping support, and Martin Fisher and two anonymous reviewers for comments and suggestions.

Author contributions

Study conception, design: FP, CC, VW; fieldwork: FP; data analysis: FP, MS; writing: all authors; revisions: FP, MS, CC.

Conflicts of interest


Ethical standards

Prior to fieldwork, an ethical assessment was conducted and clearance was granted by Imperial College London (2017-01288869-PILKINGTON-FABP), and this research otherwise abided by the Oryx guidelines on ethical standards.



Present address: Fauna & Flora International, The David Attenborough Building, Pembroke Street, Cambridge, CB2 3QZ, UK

Supplementary material for this article is available at


Ahrends, A., Burgess, N.D., Milledge, A.H., Bulling, M.T., Fisher, B., Smart, J.C.R. et al. (2010) Predictable waves of sequential forest degradation and biodiversity loss spreading from an African city. Proceedings of the National Academy of Sciences of the United States of America, 107, 1455614561.CrossRefGoogle ScholarPubMed
Ajwad, M.I., Hut, S., Abdulloev, I., Audy, R., Kataoka, S., Larrison, J. & Torracchi, F. (2014) The Skills Road: Skills for Employability in Tajikistan. World Bank, Washington, DC, USA.Google Scholar
Badreldin, N. & Sanchez-Azofeifa, A. (2015) Estimating forest biomass dynamics by integrating multi-temporal landsat satellite images with ground and airborne LiDAR Data in the Coal Valley Mine, Alberta, Canada. Remote Sensing, 7, 28322839.CrossRefGoogle Scholar
Ball, J., Carle, J. & Del Lungo, A. (2005) Contribution of poplars and willows to sustainable forestry and development. Unasylva, 56, 39.Google Scholar
Beer, R., Kaiser, F., Schmidt, K., Ammann, B., Carraro, G., Grisa, E. & Tinner, W. (2008) Vegetation history of the walnut forests in Kyrgyzstan. Quaternary Science Reviews, 27, 621632.CrossRefGoogle Scholar
Bond, W.J. (1994) Do mutualisms matter? Assessing the impact of pollinator and disperser disruption on plant extinction. Philosophical Transactions of the Royal Society B: Biological Sciences, 344, 8390.Google Scholar
Bongers, F., Poorter, L., Hawthorne, W. & Sheil, D. (2009) The intermediate disturbance hypothesis applies to tropical forests, but disturbance contributes little to tree diversity. Ecology Letters, 12, 18.CrossRefGoogle ScholarPubMed
Breiman, L. (2001) Random forests. Machine Learning, 45, 532.CrossRefGoogle Scholar
Cantarello, E., Lovegrove, A., Orozumbekov, A., Birch, J., Brouwers, N. & Newton, A.C. (2014) Human impacts on forest biodiversity in protected walnut–fruit forests in Kyrgyzstan. Journal of Sustainable Forestry, 33, 454481.CrossRefGoogle Scholar
Chambers, R. (1992) Rural Appraisal: Rapid, Relaxed and Participatory. Institute for Development Studies, Brighton, UK.Google Scholar
Chilonda, P. & Otte, J. (2006) Indicators to monitor trends in livestock production at national, regional and international levels. Livestock Research for Rural Development, 18, 117.Google Scholar
Colding, J. & Folke, C. (2001) Social taboos: ‘invisible’ systems of local resource management and biological conservation. Ecological Applications, 11, 584600.Google Scholar
Curtis, J. & McIntosh, R. (1950) The interrelations of certain analytic and synthetic phytosociological characters. Ecology, 3, 434455.CrossRefGoogle Scholar
Darwin Initiative (2017) Enhancing Forest Biodiversity and Community Resilience to Tajikistan's Changing Climate. Darwin Initiative, Penicuik, UK. [accessed 11 February 2020].Google Scholar
Dytham, C. (2015) Choosing and Using Statistics: A Biologist's Guide, 3rd edition. Wiley-Blackwell, Chichester, UK.Google Scholar
FAO (Food and Agriculture Organization of the United Nations) (2007) People, Forests and Trees in West and Central Asia: Outlook for 2020. FAO Forestry Paper 152, Rome, Italy.Google Scholar
FAO (Food and Agriculture Organization of the United Nations) (2010) Global Forest Resources Assessment Country Reports: Tajikistan. FAO, Rome, Italy.Google Scholar
FFI (Fauna & Flora International) (2018) Conserving Threatened Fruit-and-Nut Forests in Tajikistan. Fauna & Flora International, Cambridge, UK. [accessed 11 February 2020].Google Scholar
Fletcher, R. & Fortin, M.-J. (2018) Spatial Ecology and Conservation Modeling: Applications with R. Springer International Publishing, Springer Nature, Switzerland.CrossRefGoogle Scholar
Gordon, J.E., Hawthorne, W.D., Reyes-García, A., Sandoval, G. & Barrance, A.J. (2004) Assessing landscapes: a case study of tree and shrub diversity in the seasonally dry tropical forests of Oaxaca, Mexico and southern Honduras. Biological Conservation, 117, 429442.CrossRefGoogle Scholar
Grime, J.P. (1998) Benefits of plant diversity to ecosystems. Journal of Ecology, 86, 902910.CrossRefGoogle Scholar
Halimova, N. (2012) Land Tenure Reform in Tajikistan: Implications for Land Stewardship and Social Sustainability: A Case Study. In Rangeland Stewardship in Central Asia: Balancing Improved Livelihoods, Biodiversity Conservation and Land Protection (ed. Squires, V.), pp. 305329. Springer, Dordrecht, The Netherlands.CrossRefGoogle Scholar
Husch, B., Beers, T.W. & Kershaw, J.A. (2003) Forest Mensuration. 4th edition.Wiley, Hoboken, USA.Google Scholar
Jones, J.P.G., Andriamarovololona, M.M. & Hockley, N. (2008) The importance of taboos and social norms to conservation in Madagascar. Conservation Biology, 22, 976986.CrossRefGoogle ScholarPubMed
Kenkel, N. & Podani, J. (1991) Plot size and estimation efficiency in plant community studies. Journal of Vegetation Science, 2, 539544.CrossRefGoogle Scholar
Keyser, T.L. & Loftis, D.L. (2013) Long-term effects of single-tree selection cutting on structure and composition in upland mixed-hardwood forests of the southern Appalachian Mountains. Forestry, 86, 255265.CrossRefGoogle Scholar
Korhonen, L., Korhonen, K.T., Rautiainen, M. & Stenberg, P. (2006) Estimation of forest canopy cover. Silva Fennica, 40, 577588.CrossRefGoogle Scholar
Lade, S.J., Tavoni, A., Levin, S.A. & Schlüter, M. (2013) Regime shifts in a social-ecological system. Theoretical Ecology, 6, 359372.CrossRefGoogle Scholar
Lerman, Z. (2011) Tajikistan's Vulnerability to Climate Change. Discussion Paper No. 7.11. The Hebrew University of Jerusalem, Jerusalem, Israel.Google Scholar
Lerman, Z. (2012) Rangeland stewardship in Central Asia. In Rangeland Stewardship in Central Asia (ed. Squires, V.), pp. 165187. Springer, Dordrecht, The Netherlands.CrossRefGoogle Scholar
Lindenmayer, D.B., Margules, C.R. & Botkin, D.B. (2000) Indicators of biodiversity for ecologically sustainable forest management. Conservation Biology, 14, 941950.CrossRefGoogle Scholar
Lovett, J.C., Marshall, A.R. & Carr, J. (2006) Changes in tropical forest vegetation along an altitudinal gradient in the Udzungwa Mountains National Park, Tanzania. African Journal of Ecology, 44, 478490.CrossRefGoogle Scholar
Maselli, D. & Rahim, I. (2012) Setting the stage. In Rangeland Stewardship in Central Asia (ed. Squires, V.), pp. 331. Springer, New York, USA.CrossRefGoogle Scholar
McIntosh, R.P. (1967) An index of diversity and the relation of certain concepts to diversity. Ecology, 48, 392404.CrossRefGoogle Scholar
Mislimshoeva, B., Herbst, P. & Koellner, T. (2016) Current pathways towards good forest governance for ecosystem services in the former soviet republic Tajikistan. Forest Policy and Economics, 63, 1119.CrossRefGoogle Scholar
Moreno, C.E. & Halffter, G. (2000) Assessing the completeness of bat biodiversity inventories using species accumulation curves. Journal of Applied Ecology, 37, 149158.CrossRefGoogle Scholar
Nagendra, H. (2002) Opposite trends in response for the Shannon and Simpson indices of landscape diversity. Applied Geography, 22, 175186.CrossRefGoogle Scholar
Newton, A.C. (2007) Forest Ecology and Conservation. Oxford University Press, New York, USA.CrossRefGoogle Scholar
Newton, A.C. & Echeverría, C. (2014) Analysis of anthropogenic impacts on forest biodiversity as a contribution to empirical theory. In Forests and Global Change (eds Coomes, D., Burslem, D. & Simonson, W.D.), pp. 417446. Cambridge University Press, Cambridge, UK.CrossRefGoogle Scholar
Nowak, A., Nowak, S. & Nobis, M. (2011) Distribution patterns, ecological characteristic and conservation status of endemic plants of Tadzhikistan. Journal for Nature Conservation, 19, 296305.CrossRefGoogle Scholar
Orozumbekov, A., Cantarello, E. & Newton, A.C. (2014) Status, distribution and use of threatened tree species in the walnut–fruit forests of Kyrgyzstan. Forests, Trees and Livelihoods, 24, 117.CrossRefGoogle Scholar
Ostrom, E. (1999) Sel f-Governance and Forest Resources. Occasional Paper No. 20: Center For International Forestry Research, Bogor, Indonesia.Google Scholar
Peet, R.K. (1974) The measurement of species diversity. Annual Review of Ecology and Systematics, 5, 285307.CrossRefGoogle Scholar
Peters, C.M. (1994) Sustainable Harvest of Non-Timber Plant Resources in Tropical Moist Forest: an Ecological Primer. Biodiversity Support Program, Washington, DC, USA.Google Scholar
QGIS Development Team (2020) QGIS Geographic Information System. Open Source Geospatial Foundation Project. [accessed 20 March 2020].Google Scholar
Reimoser, F., Armstrong, H. & Suchant, R. (1999) Measuring forest damage of ungulates. Forest Ecology and Management, 120, 4758.CrossRefGoogle Scholar
Safarov, N. (2003) Republic of Tajikistan: Natural Strategy and Action Plan on Conservation and Sustainable Use of Biodiversity. Government of the Republic of Tajikistan, Dushanbe, Tajikistan.Google Scholar
Safarov, N., Novikova, T. & Shermatov, K. (2014) Fifth National Report on Preservation of Biodiversity of the Republic of Tajikistan. Republic of Tajikistan, Dushanbe, Tajikistan.Google Scholar
Squires, V. & Safarov, N. (2013) High altitude ecosystems and biodiversity of Tajikistan: conservation and management. In High-Altitude Rangelands and their Interfaces in the Hindu Kush Himalayas (eds Ning, W., Rawat, G.S., Joshi, S., Ismail, M. & Sharma, E.), pp. 7888. International Centre for Integrated Mountain Development, Kathmandu, Nepal.Google Scholar
Shannon, C.E. & Weaver, W. (1949) The Mathematical Theory of Communication. The University of Illinois Press, Urbana, USA.Google Scholar
Simpson, E.H. (1949) Measurement of diversity. Nature, 163, 688.CrossRefGoogle Scholar
Singh, M., Evans, D., Friess, D.A., Tan, B.S. & Nin, C.S. (2015) Mapping above-ground biomass in a tropical forest in Cambodia using canopy textures derived from Google Earth. Remote Sensing, 7, 50575076.CrossRefGoogle Scholar
Southwood, T.R.E. & Henderson, P.A. (2015) Ecological Methods. 3rd edition. Wiley-Blackwell, Hoboken, USA.Google Scholar
Sutton, W., Block, R. & Srivastava, J. (2009) Adaptation to Climate Change in Europe and Central Asia Agriculture, pp. 161. World Bank Group, Washington, DC, USA.Google Scholar
Vera, F. (2000) Grazing Ecology and Forest History. CABI, Wallingford, UK.CrossRefGoogle Scholar
Waylen, K.A., Fischer, A., McGowan, P.J.K. & Milner-Gulland, E.J. (2013) Deconstructing community for conservation. Human Ecology, 41, 575585.CrossRefGoogle ScholarPubMed
Wolfgramm, B., Stevenson, S., Lerman, Z.Zähringer, J., Liniger, H. (2011) Tajikistan – Pilot Programme for Climate Resilience, Sustainable Land Management and Climate Change Adaptation, PPCR Component A5. Centre for Development and Environment, University of Bern, Bern, Switzerland.Google Scholar
World Bank (2004) Tajikistan: Dashtidzhum Biodiversity Conservation Project. Vol. 1. World Bank, Washington, DC, USA.Google Scholar
Zhupankhan, A., Tussupova, K. & Berndtsson, R. (2017) Could changing power relationships lead to better water sharing in Central Asia? Water, 9, 139.CrossRefGoogle Scholar
Figure 0

Fig. 1 Location of the 40 survey plots, and the four villages where people were interviewed, in Dashtijum Nature Reserve, Tajikistan.

Figure 1

Table 1 Stand structure of the forests of Dashtijum Nature Reserve, with mean tree basal area, tree density and Importance Value Index (IVI; see main text for details of calculation) of all tree species found.

Figure 2

Fig. 2 Per cent of 40 households that harvested nine fruit or nut species from the forests in Dashtijum Nature Reserve (Fig. 1).

Figure 3

Fig. 3 Responses from 40 household interviews regarding compositional and resource changes within the forests of Dashtijum Nature Reserve (Fig. 1) over the previous 5 years.

Figure 4

Table 2 The importance (i.e. how important the predictor variables in a given model are in explaining the response variable), of various predictor variables in influencing species richness, the Shannon index and total basal area, computed using Random Forest models.

Supplementary material: PDF

Pilkington et al. Supplementary Materials

Pilkington et al. Supplementary Materials

Download Pilkington et al. Supplementary Materials(PDF)
PDF 698 KB
You have Access

Send article to Kindle

To send this article to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

Note you can select to send to either the or variations. ‘’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

The first rapid forest inventory and resource use assessment of Dashtijum Nature Reserve, Tajikistan: a mixed methods approach
Available formats

Send article to Dropbox

To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Dropbox.

The first rapid forest inventory and resource use assessment of Dashtijum Nature Reserve, Tajikistan: a mixed methods approach
Available formats

Send article to Google Drive

To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your <service> account. Find out more about sending content to Google Drive.

The first rapid forest inventory and resource use assessment of Dashtijum Nature Reserve, Tajikistan: a mixed methods approach
Available formats

Reply to: Submit a response

Please enter your response.

Your details

Please enter a valid email address.

Conflicting interests

Do you have any conflicting interests? *