Disadvantaged Economic Conditions and Stricter Border Rules Shape Afghan Refugees’ Ethnobotany: Insights from Kohat District, NW Pakistan

The study of migrants’ ethnobotany can help to address the diverse socio-ecological factors affecting temporal and spatial changes in local ecological knowledge (LEK). Through semi-structured and in-depth conversations with ninety interviewees among local Pathans and Afghan refugees in Kohat District, NW Pakistan, one hundred and forty-five wild plant and mushroom folk taxa were recorded. The plants quoted by Afghan refugees living inside and outside the camps tend to converge, while the Afghan data showed significant differences with those collected by local Pakistani Pathans. Interviewees mentioned two main driving factors potentially eroding folk plant knowledge: (a) recent stricter border policies have made it more difficult for refugees to visit their home regions in Afghanistan and therefore to also procure plants in their native country; (b) their disadvantaged economic conditions have forced them to engage more and more in urban activities in the host country, leaving little time for farming and foraging practices. Stakeholders should foster the exposure that refugee communities have to their plant resources, try to increase their socio-economic status, and facilitate both their settling outside the camps and their transnational movement for enhancing their use of wild plants, ultimately leading to improvements in their food security and health status.


Introduction
One of the most promising trajectories in current and future debates in ethnobotany concerns the temporal and spatial changes in local ecological knowledge (LEK) and the diverse socio-ecological factors that may affect these dynamics. For this reason, the study of migrants' ethnobotany can help to reveal some general human ecological concepts underpinning the universal phenomenon of human relocation. Worldwide, 89.3 million people are forcibly displaced, of which 27.1 million are refugees [1]. Whenever refugees leave their homeland and reach a new location they are challenged by the establishment of new relationships with the novel natural and socio-cultural environments, as well as with access to available food and health resources in the host country [2]. Often, migrant communities neither completely reject the majority culture's ethnobotanical knowledge nor totally give up their own traditional practices, but rather find a compromise in the continuum between strengthening their own identity and fully adapting to the host culture [3]. The majority of

•
To document the use of wild food and medicinal (human and veterinary) plants by local Pathans and Afghan refugees living inside and outside the camps in the Kohat District of NW Pakistan; • To identify commonalities and differences in the folk plant uses among the three groups and to possibly interpret these findings in cultural terms.

The Ethnobotany of Pathans and Afghan Refugees
We recorded the use of 145 wild plant taxa and four mushrooms belonging to 63 botanical families. Sixty-four percent of the recorded species were used as medicines, fifty-six percent as food, and twenty-one percent as veterinary applications (Table 1). In addition, a few plants were used as fuel wood (7%) and as cosmetics/fragrances (3%).
The highest number of taxa was reported by Afghan refugees living outside the refugee camps (AO; 73%), followed by Afghan refugees living inside the refugee camps (AI; 66%), while the lowest number of taxa was reported by local Pathans (48%) ( Table 2). FA AO FP AI, PO Fields and foothills M: leaves, powdered and then topically applied over skin and snake bite areas to remove thorns, and to treat seborrheic keratosis, as an anti-diabetic, and in antidotes, or sometimes leaves are punctured with a needle and placed in shoes, anti-diabetic; madar juice, topically applied to stop bleeding V: leaves as fodder, kills intestinal worms, sometimes the insects (Poekilocerus pictus), that feed on this plant are caught and powdered, which is then put in the nose of animals to kill brain worms; madar juice, topically applied over the skin and mammary glands of cows to remove piles and ticks O: madar juice, used in fermentation

Wild Food Plants
Our results show that most of the recorded wild food plants (WFPs) are used as raw snacks (39 taxa; 26%) or cooked as vegetables (38 taxa; 25%), while a few of them were reported as used both as raw snacks and cooked as vegetables, e.g., Allium carolinianum and Solanum nigrum. About 3% of the food uses were reported to be practiced in the past, while the remaining uses are still practiced by the studied communities. The highest number of plant taxa was mentioned by AO (n = 61), followed by AI (n = 46). Apteranthes tuberculata was the most common plant taxon among all the groups, although Allium rosenbachianum was very important among AI and Mentha longifolia among PO. Cooked preparations and raw snacks were equally important among refugees (30 and 29 among AO, and 25 and 26 among AI, respectively). Conversely, among PO, wild plant-based cooked dishes were far more common than raw snacks (28 vs. 14 taxa, respectively). All the other preparations were secondary. Most of the WFPs are cooked in a very simple way: washed, cut, boiled, and fried in oil with onion, tomatoes, and chilies. Some of the reported WFPs are boiled and cooked with a combination of other wild vegetables according to taste preferences, i.e., Atriplex laciniata and Chenopodium album are boiled, fried, and cooked together among PO. Among the reported taxa, five taxa were utilized raw as salad items, i.e., Apteranthes tuberculata, Nasturtium officinale, Thymus, and Malva neglecta (mentioned by all the groups), and Atriplex laciniata (mentioned only among PO). A few taxa, e.g., Achyranthes aspera (used only by PO), Thymus sp. (mentioned by AO), Myrtus communis (mentioned by refugees), and Olea europaea (mentioned by all the groups), are utilized in making tea which has some therapeutic value. Only three plants were reported to be used to make sauces: Allium carolinianum and Oxalis corniculata among refugees, and Mentha longifolia. Among the wild vegetables, the fruits of Carum carvi (all groups), Cuminum cyminum (AI), the aerial parts of Ferula foetida (AO), Mentha longifolia (all groups), and Trigonella foenum-graecum (AO and PO) among the wild vegetables are predominantly exploited for seasoning.
Knowledge related to wild food plant preparation is vertically transmitted mainly from mothers to daughters.

Wild Medicinal Plants
Ninety-six taxa were reported to be used against ninety ailments. In general, digestive disorders were reported to be treated by the highest diversity of plants (32 taxa), followed by diabetes with 24 plants, respiratory disorders with 20 plants, fever and cough with 16 plants, and skin disorders with 12 plant species, while a maximum of 40 diseases/ailments are documented as treated by a single plant taxon. Nearly 7% of the medicinal uses were reported to be practiced in the past, while the remaining uses are still practiced by the studied communities. The results show that the majority of wild medicinal plants (WMPs) were used to treat the most common or day-to-day problems, i.e., fever, cough, stomachrelated issues, itching, flu, chest pain, and internal and external wounds. It was reported that the communities also use the plant's ingredients for the treatment of some major and severe ailments such as arthritis, tuberculosis, heart-related problems, ulcers, malaria, typhoid, hepatitis B and C, paralysis, and sexual disorders.
The AI group used the highest number of WMPs (n = 67) followed by AO (n = 58). Artemisia scoparia and Berberis lycium were the most common WMPs among refugees (AI and AO). Withania coagulans was reported by all the groups, but especially by local Pathans. Fever and diabetes were common ailments treated with WMPs among all the groups. Stomach problems and cough were also commonly treated with wild plants among AI.
Zygophyllum indicum is used to treat the highest diversity of problems (12 ailments), followed by Berberis lycium and Nepeta laevigata which are used to treat 11 ailments each.
According to the study participants, Berberis lycium is thought to cure almost all types of ailments, such as digestive, respiratory, excretory, skeletal, and dermatological problems, as well as internal and external wounds. The majority of plants (41 species) were documented to have more than one part that is therapeutically important.
The majority of the medicinal formulations were administered both internally as an oral medication and externally as topical doses (skin, dental, and eye problems). Most of the remedies were prepared from a single species, while in a few cases, the preparations were a mixture of different plant species. For example, the root extract of Melia azedarach and the bark extract of Juglans regia were mixed and applied to the hair to permanently dye it black among refugees. The leaves of Nerium oleander and Fegonia indica are crushed and the extract is used against pimples and itching by AO and PO, while the leaves of Peganum harmala, Trigonella foenum-graecum, and Lepidium sativum are mixed and boiled, and the water is drunk to start menses in women among AI. In addition, AI boiled the seeds of Artemisia absinthium in water to soften them, after which they were given to children to cure the flu. Refugees mentioned that in the past adults chewed the seeds to treat the same problem. Likewise, the fruits of Juniperus communis were boiled in water, but only the water is given to children to treat fever, while elderly individuals consume both the fruits and water for the same problem. Overall, the common modes of preparation consisted of juice, paste, decoction, powder, infusion, and chewing raw plant parts.
Vertical transmission from mother to daughter was the most common way of sharing knowledge of wild medicinal plants. Nevertheless, it is men who harvest wild plants regardless of their final use.

Wild Veterinary Plants
We recorded 31 species concerned with the veterinary domain (plants used both for treating and feeding animals, see Table 1). More than half of the reported ethnoveterinary taxa (20 taxa) were used for feeding purposes. A maximum of 26 diseases or ailments were documented to be treated by the reported taxa, with the most common diseases being diarrhea, internal and external body wounds, constipation, intestinal worms, skeletal disorders, bloating, foot and mouth disease, fever, energy production, galactagogue, and urine blockage.
The top used wild veterinary plants (WVPs) were different among the three groups (Berberis lycium among AO, Nepeta laevigata among AI, Tamarix aphylla among PO), yet most of the WVPs were used for fodder and, to a lesser extent, wounds, intestinal worms, and diarrhea.
According to the study participants, most often ethnoveterinary knowledge is learned from elderly individuals and is passed to the next generation from father to child, but, in certain cases, it may be learned from friends or neighbors through general group discussions. The WVPs are often used in a very simple way, mostly through ingestion or application directly to the affected area. Decoctions are directly taken, or powders are mixed with milk or jaggery (a traditional non-centrifugal cane sugar) and given to the animal. Inside the plant body of Calotropis procera (mentioned by AI), a certain animal species (i.e., Poekilocerus pictus). These are dried, powdered, and put in the nose of animals (mostly cows and buffalos) to kill brain worms (any parasitic, worm-like species that inhabits the brain of another organism). In the past, the leaves of Juniperus communis were placed on embers by refugees and the smoke was inhaled by animals, shut in a closed room with the smoke for nearly half an hour, to remove intestinal worms. The decoction of Berberis lycium was reported by the majority of the informants in all the groups for healing internal and external wounds. One AI participant commented about the importance of Berberis lycium: "Once the foot of my goat was broken. I gave her the decoction of Ziar largai (Berberis lycium) for 15 days and she recovered. A year later when she reached the age of sacrifice [the practice of the slaughter of animals that Muslim do in the way of Allah to earn reward, when animals reach nearly one year of age] and we removed its skin, we found a hard shell of Ziar largai around the broken bone".
Nearly one-fourth of the ethnoveterinary uses were reported to be practiced in the past, while the remaining uses that are still practiced by the studied communities also have very low frequencies of quotation, which shows that the folk ethnoveterinary knowledge has eroded dramatically. The majority of the past ethnoveterinary uses were reported by AI, possibly due to the fact that inside the camps most families do not keep domestic animals on a regular basis. Only a few families were found to have animals (not more than two) in their homes due to poor economic conditions and limited availability of places to live. Both AI and AO mentioned that most of the time they were not allowed to freely herd their animals in the mountain regions, and therefore for this purpose they need to pay money to PO communities to rent a piece of land on an agreement for their herds, but most often it becomes difficult economically for both of the communities. The results show a dramatic loss of knowledge regarding ethnoveterinary practices due to the socio-economic shift occurring in the local communities as the majority of the inhabitants have decreased animal breeding and have abandoned pastoral activities.

Other Considerations on the Use of Wild Plants (Habitat, Gender, Plant Material Preservation)
More than half of the reported taxa mentioned by local Pathans are foraged in the fields near their houses as they are closely associated with agricultural practices and live mostly in plain areas, while the other two studied groups reported those taxa which are mostly available and foraged in mountains due to their geographical location and climate in their homeland in Afghanistan. The results show that 38% of wild taxa are foraged in Afghanistan, while only 19% of wild taxa are foraged in Pakistan; the remaining 42% of wild taxa can be foraged in both countries.
The wild plant taxa were collected from different habitats, e.g., fields, mountains, foothills, water banks, home gardens, graveyards, and sandy areas (Figure 1).
Mountain environments were the most common foraging habitat for refugees (62 taxa among AI and 69 among AO), followed by plain fields (50 taxa among AI and 61 taxa among AO). The proportion of common foraging habitats is very different among local Pathans, as fields were the most frequent (58 taxa), followed by mountains (25 taxa).
One-fifth of the wild taxa were reported by more than 40% of the informants. The majority of the informants stated that the growth and availability of wild plant taxa were heavily affected by environmental and anthropogenic factors. The availability of certain plants, i.e., Cichorium intybus, Tulipa sp., Apteranthes tuberculata, and Asparagus officinalis, were reported to have decreased as a result of the low availability of water and precipitation rate in the study area. Conversely, the availability of certain plants, i.e., Carum carvi, Punica granatum, Ephedra intermedia, and Nepeta laevigata, were reported to have increased as a result of government restrictions to high elevation areas for security reasons, as these areas are considered unsafe and vulnerable to acts of terrorism. Because of overexploitation (for both human and animal consumption), the availability of Zygophyllum indicum, Withania coagulans, and Prosopis juliflora was reported to have decreased.
Foraging is mostly performed by male family members in Pakistan, while in Afghanistan both men and women are involved in foraging, possibly due to the fact that the camp environment is not suitable for foraging as different families belonging to different tribes live alongside each other which prevents women from engaging in this activity as it is against Pathans culture for women to come out of their houses and freely move about in front of male community members. In rare cases, in Pakistan, women who live outside the camps also forage for those species which can be found near their houses. The younger generation does not show any interest in foraging as they are mostly involved in smallscale businesses or work as laborers because of their low economic position and give most of their attention to modern and market-based products. One of the AO study participants commented: "We don't find enough time for foraging after returning from our work, and also give attention to those products which are easily available at markets and have high taste appeal [meaning those market products which have a pleasant taste as compared to wild ones]." Afghanistan, while only 19% of wild taxa are foraged in Pakistan; the remaining 42% of wild taxa can be foraged in both countries.
Mountain environments were the most common foraging habitat for refugees (62 taxa among AI and 69 among AO), followed by plain fields (50 taxa among AI and 61 taxa among AO). The proportion of common foraging habitats is very different among local Pathans, as fields were the most frequent (58 taxa), followed by mountains (25 taxa).
One-fifth of the wild taxa were reported by more than 40% of the informants. The majority of the informants stated that the growth and availability of wild plant taxa were heavily affected by environmental and anthropogenic factors. The availability of certain plants, i.e., Cichorium intybus, Tulipa sp., Apteranthes tuberculata, and Asparagus officinalis, were reported to have decreased as a result of the low availability of water and precipitation rate in the study area. Conversely, the availability of certain plants, i.e., Carum carvi, Punica granatum, Ephedra intermedia, and Nepeta laevigata, were reported to have increased as a result of government restrictions to high elevation areas for security reasons, as these areas are considered unsafe and vulnerable to acts of terrorism. Because of overexploitation (for both human and animal consumption), the availability of Zygophyllum indicum, Withania coagulans, and Prosopis juliflora was reported to have decreased. We documented the trade of 40% (n = 59) of the reported taxa in local markets, in which the highest number of taxa (36 taxa) were bought from the markets by PO. The special foraging of Apteranthes tuberculata is performed by AI in the month of February when the availability of the species is considered high in the mountain pastures. Apteranthes tuberculata is an important economic taxon that is foraged by Afghan refugees and brought to local markets for sale. One of the AI participants commented about the importance of this species: "We don't have time in February because we have only one month for the foraging of Pamanai (Apteranthes tuberculata) and we hike 3 h every day to reach the particular patches of the mountains where it is available and collect 5 to 6 kg each day which pays good money compared to any other occupation".
We have observed that among the quoted taxa, only 15% (n = 22) of taxa were reported to be stored, the majority of which were wild food plants (i.e., Quercus incana, Pinus gerardiana, Mentha longifolia). The most used plant parts were seeds (36%) followed by fruits (27%).
We observed that both AI and AO reported taxa that could be imported from Afghanistan for personal use without the need for permission (i.e., Alkanna tinctoria, Berberis lycium, Buxus wallichiana). The interviews also highlighted that the new border policy recently devised by Pakistani authorities has drastically decreased the opportunity of acquiring plant materials from the homeland of Afghanistan as now only those Afghans holding a valid passport, visa, or rahdari (a card issued since 2015 to facilitate frequent cross-border movement) are allowed to enter Pakistan [10].

Commonalities and Differences among the Three Groups
There is a remarkable difference in the ethnobotany of PO compared with both AI and AO (Figure 2). The three groups share 41 taxa, while in total the refugee groups (AI and AO) share 75 taxa. The least similar groups in terms of shared plant taxa are AI and PO with only 44 taxa (Jaccard Similarity Index = 0.37). Both these groups also mentioned 16 taxa not shared with the other groups. Venn diagram with the total identified quoted taxa and relative Jaccard Similarity Indexes (JI). We recorded 16 unidentified taxa, but 1 unidentified species was mentioned by AI and AO.
Of the 133 identified taxa, 25 were mentioned by more than 40% of the interviewees in each group (Figure 3). Among those, only three were shared by the three groups (Zygophyllum, Withania, and Apteranthes). Fifteen taxa were found exclusively among local Pathans who shared only five top used species with the other groups. Indeed, local Pathans' Jaccard Similarity Indexes drop to 0.11 with AI and 0.19 with AO, while the two refugee groups maintain a high similarity index (0.50).
Local Pathans reported a maximum of 18 idiosyncratic plants, the majority of which were WFPs, while the two groups of refugees (AI and AO) reported a fewer number of wild taxa. Of the 133 identified taxa, 25 were mentioned by more than 40% of the interviewees in each group (Figure 3). Among those, only three were shared by the three groups (Zygophyllum, Withania, and Apteranthes). Fifteen taxa were found exclusively among local Pathans who shared only five top used species with the other groups. Indeed, local Pathans' Jaccard Similarity Indexes drop to 0.11 with AI and 0.19 with AO, while the two refugee groups maintain a high similarity index (0.50).
Local Pathans reported a maximum of 18 idiosyncratic plants, the majority of which were WFPs, while the two groups of refugees (AI and AO) reported a fewer number of wild taxa.
Nearly one-third of WFPs are commonly shared by the studied groups ( Figure 4). Afghan Pathans living outside of camps reported the largest number (20) of idiosyncratic WFPs used, while AI reported a smaller number (9) of idiosyncratic WFPs. Local Pathans also reported 10 idiosyncratic WFPs. The most similar groups are the two refugee communities, yet their divergence from PO is less evident than for the overall taxa. Nearly one-third of WFPs are commonly shared by the studied groups ( Figure 4). Afghan Pathans living outside of camps reported the largest number (20) of idiosyncratic WFPs used, while AI reported a smaller number (9) of idiosyncratic WFPs. Local Pathans also reported 10 idiosyncratic WFPs. The most similar groups are the two refugee communities, yet their divergence from PO is less evident than for the overall taxa.  Sixteen WMPs and WVPs were commonly shared by the studied groups ( Figure 5). The highest similarity index was found between the two refugee groups (JI = 0.44). Afghan Pathans living inside camps quoted the greatest number of medicinal and veterinary plants and reported the maximum of 23 idiosyncratic taxa among their overall use of 70 taxa. Sixteen WMPs and WVPs were commonly shared by the studied groups ( Figure 5). The highest similarity index was found between the two refugee groups (JI = 0.44). Afghan Pathans living inside camps quoted the greatest number of medicinal and veterinary plants and reported the maximum of 23 idiosyncratic taxa among their overall use of 70 taxa. In comparison with AI, and somehow paradoxically, AO reported fewer medicinal and veterinary plants despite the fact that this group also comprised Kochi Afghans (pastoral nomads) and practice pastoralism, and thus they are thought to have remained in close association with their natural environment and have extensive information about traditional remedies. Elderly community members still use traditional plants for the treatment of various diseases and the majority of the study participants from inside the camps kept a lot of medicinal plants in their houses. A 65-year-old elderly AO participant commented: "In the past, we used the majority of medicinal plants, and we did not go to the hospital until some big problem happened to us, because we had a strong belief in those plants and that was the reason that we recovered soon after taking those folk remedies. But today we do not use all those plants because of the availability and use of the Western medical system and drugs. Now if we get a chance to use the traditional remedies against any disorder, we are not mentally satisfied with the results of traditional remedies despite them working".
The above statement shows that AO had strong beliefs about traditional therapies in the past, but due to the interference from the Western medical system, they have nearly lost this connection.

Discussion
The results show two main findings. First, Afghan Pathans outside the camps tend to collect and use more wild food plants, as already shown in a previous study conducted by some of the current authors in the Mansehra area [8], most likely because their exposure to nature is higher than that of Afghan Pathans living in camps. However, refugees In comparison with AI, and somehow paradoxically, AO reported fewer medicinal and veterinary plants despite the fact that this group also comprised Kochi Afghans (pastoral nomads) and practice pastoralism, and thus they are thought to have remained in close association with their natural environment and have extensive information about traditional remedies. Elderly community members still use traditional plants for the treatment of various diseases and the majority of the study participants from inside the camps kept a lot of medicinal plants in their houses. A 65-year-old elderly AO participant commented: "In the past, we used the majority of medicinal plants, and we did not go to the hospital until some big problem happened to us, because we had a strong belief in those plants and that was the reason that we recovered soon after taking those folk remedies. But today we do not use all those plants because of the availability and use of the Western medical system and drugs. Now if we get a chance to use the traditional remedies against any disorder, we are not mentally satisfied with the results of traditional remedies despite them working".
The above statement shows that AO had strong beliefs about traditional therapies in the past, but due to the interference from the Western medical system, they have nearly lost this connection.

Discussion
The results show two main findings. First, Afghan Pathans outside the camps tend to collect and use more wild food plants, as already shown in a previous study conducted by some of the current authors in the Mansehra area [8], most likely because their exposure to nature is higher than that of Afghan Pathans living in camps. However, refugees inside the camps paradoxically use more medicinal plants than Afghan Pathans living outside the camps, possibly because of more disadvantaged economic conditions, making it more difficult for them to access Pakistani health services. These two divergent variables (decreased exposure to the natural environment and difficulties in accessing health care services due to social marginalization) have shaped different trajectories with respect to the wild food and medicinal plants that they gather and use.
However, while Afghan ethnobotanies show some convergences, significant differences do exist with the ethnobotany of local Pathans, who use far fewer plants. Although Afghan refugees in fact utilize wild plants from across different habitats in both Afghanistan and Pakistan (also via family networks), this does not occur for local Pathans, who use wild plants found only in their environment.

Afghan Refugees Share Similar Ethnobotanical Knowledge, with Some Differences
More than half of the identified taxa were common to AI and AO, possibly due to the fact that both these communities inhabited the same socio-ecological environment before their migration and had similar cultures and customs. Additionally, the two groups are exogamous, which allows intermarriages and consequently promotes exchanges of LEK, mainly through vertical transmission (transferred from grandmother to mother and from mother to daughter); a similar finding has been documented in other studies [11,12].
Moreover, we found a remarkable divergence in the ethnobotany of PO with both AI and AO, possibly because local Pathans have lived together with other Pakistani ethnic groups for centuries, sharing the same religious faith (Sunni Muslim) and social life, while Afghans relocated only a few decades ago. According to some authors [13], the nature and degree of contact and exchange with neighboring socio-culture groups may determine the level of homogenization of LEK. Nearly one-third of the taxa reported by AI and AO were not quoted by PO, possibly due to the fact that the majority of the reported taxa were harvested in Afghanistan where refugees regularly visit their relatives, friends, and family. This may reinforce the vulnerability of refugees' ethnobotanical knowledge as a result of a kind of "displaced" foraging (living in one country and foraging in another one). However, the strong kinship ties that Afghan refugees in Pakistan have always fostered with their relatives in their home country could also allow them to easily acquire dried medicinal plants without any actual gathering.
The comparatively high number of idiosyncratic plants reported by local Pathans is possibly due to their centuries-old familiarity with the natural environment. Additionally, they are highly dependent upon farming which could enhance the resilience of LEK related to plant resources. We observed that both refugee groups are economically disadvantaged and sometimes tend to work in cities, leaving little time for farming and foraging. One of the AI study participants commented: "The economic condition of the majority of the families is not good, so from a very small age, we try to engage our children in different kinds of work, e.g., working as a laborer or trying to run a small shop in the markets or in most cases we send them abroad to earn money. So after returning, we neither want to discuss wild plants nor have time for foraging".
The above statement suggests that the progressive weakening of the plant-people relationship between the older and younger generations could probably further lead to the erosion of LEK.
While the higher consumption of wild food plants among Afghan Pathans living outside the camps is possibly due to their greater opportunity of roaming in the environment as a result of their agricultural activities, the smaller number of plants mentioned by the refugees living within camps could be due to their more robust food security, as food is often provided to them by humanitarian organizations.
On the other hand, the higher number of wild medicinal plants (both in human and veterinary medicine) among AI (vs. AO) may be due to two factors: (a) refugees inside the camps have more limited access to proper health services, and therefore they have to rely more on their traditional plant medicine; (b) camps in the study area include many Kochi Afghans (traditional pastoral nomads), who still retain vast ethnoveterinary knowledge.

Refugees Utilize Wild Plants across Habitats and Borders
Nearly two out of five plants mentioned in this study are foraged in Afghanistan, only one out of five in Pakistan, and two out of five in both countries.
Local Pathans tend to forage plants in the fields near their houses, while refugees outside the camps foraged until recently in the mountains in Afghanistan. Indeed, the proximity of the plants to villages could possibly lead to an increase in the knowledge and use of synanthropic plants [14][15][16], while they may have less knowledge of those plants which grow far from their houses [17][18][19].
Until the recent past, Afghan refugees (outside the camps) used to continually visit relatives in their homeland, and, as described in other studies [20], these family networks allow diasporic groups to acquire plants transnationally: when people go to visit friends and families in their place of origin, dried plant materials are acquired before traveling back to the host country [21]. In recent years, however, border rules between Pakistan and Afghanistan have become stricter and our study participants agreed that these trans-border movements of people and plants are more difficult now.
Additionally, many informants feel uneasy about foraging in the mountains in their host country (Pakistan). One of the AI participants commented: "Without an expert or a local person, it is an unwise decision to go for research or the collection of wild plants because something bad could happen to us".
The above statement shows that the refugees do not feel comfortable in their new environment, whereas, despite the difficulties in crossing the border and the increasing security problems in Afghanistan, they still wish to go forage in their homeland. A 71-yearold AI mentioned: "In Afghanistan, when we go to visit the mountains, people show great interest in collecting wild cumin while returning to their own homes as it is used as a flavoring agent in our food".
The AO group reported a high number of idiosyncratic uses of WFPs, which might possibly be associated with the fact that the majority of the quoted taxa derives from their homeland, where they regularly visit their relatives and place of origin in Afghanistan. In addition, the community has daily exposure to nature and practical activities which may lead to enhancing the resilience of LEK. The divergences in LEK regarding WFPs of Afghan Pathans outside the refugee camps are probably linked with their original home places, which are sometimes geographically isolated, as other works have pointed out (for example [6]) isolation could enhance the resilience of robust LEK.
On the other hand, possible reasons for the abandonment of foraging practices by Afghan Pathans inside the camps could be that they mainly live on remittances sent from abroad and they are mostly dependent on the cultivated vegetables available in markets. One of the AI study participants commented: "The camp environment is not suitable for finding a large number of wild food plants, so for this purpose, we would need to go to the mountains, which I think is hard work and a time-consuming process-that is the reason we go for those products which are easily accessible and available in the markets".
The continuation of plant use may also be related to the legal aspects of cross-border plant imports. We observed that both AI and AO reported plant taxa (i.e., Alkanna tinctoria, Berberis lycium, Buxus wallichiana) widely imported from Afghanistan for personal use without the need for permission. Our findings are in line with the work of [22] in the Netherlands, where Surinamese migrants in Amsterdam heavily relied on importation because of the flexible Dutch entry laws. Conversely, the strict entry law for plant imports into the UK makes it difficult for Latino migrants to acquire some of their homeland plants [4,23].
Since 2001, however, due to the unrestricted movement of militants across the Pakistan-Afghanistan border, both countries have faced internal security threats [24], and so to improve its control over the border, Pakistan has developed certain strategies such as closing or fencing the border and tightening rules at the border crossing for Afghans [25]. These changes in border policies have ultimately weakened the bond tying Afghan refugees to their motherland and will heavily affect the import of plant materials from Afghanistan and therefore the erosion of LEK of the refugees; similar findings were documented among Afghan refugees living in Mansehra [8]. One AO study participant mentioned: "Due to the new border policies and strictness, we are not even able to meet with our family members who remain in Afghanistan nor can they come to meet us because financially it is very costly and difficult to continue visiting our motherland. We have never experienced such restrictions since the time of migration until now and could have visited whenever we wanted".
We also observed that due to new border policies Afghan pastoralist communities have abandoned pastoralism because in summer they cannot move their herds into the mountain pastures in Afghanistan, since border laws are no longer flexible. According to one AO participant: "It is quite hard for us to keep our animals in a very hot climate in summer instead of moving them into the mountain pastures when cross-border movement is stopped and we face some major problems like the scarcity of water and fodder for animals. That's why we have now sold our herds and are trying to invest the money in different businesses".
We also observed in our study that Kochis Afghan inside the refugee camps sometimes rent a piece of land in the Pakistani mountains from local PO upon agreement; a few family members are therefore sent in the summer to the pastures with their herds while the rest of the family remain in the camp. The majority of the ethnoveterinary plant uses are not actively practiced by the study participants anymore but are only remembered. One of the AI study participants (a 53-year-old man) commented: "We don't want to keep animals in our houses because of certain reasons: we are given a very limited place to live where the whole family can hardly survive, the structure and texture of the soil inside the camp do not look suitable for growing fodder for animals, and the last is the poor economic position of the local communities".
Indeed, due to poor economic status and government restrictions, the majority of the population inside the camps do not keep any domestic animals in their houses, and only a few families have reported keeping one or two domestic animals in their houses. Moreover, the high number of plants mentioned by Afghan refugees may be misleading because the plants are mentioned by very few interviewees.
Recent developments in the political landscape of Afghanistan were not specifically considered in our field study. However, the overthrow of the Ashraf Ghani government on 15th August 2021 has further drastically affected the lives of Afghans in Pakistan as well. Many of the diasporas do not want to visit Afghanistan in the current situation, as its economy is collapsing. As the bank accounts of millions of citizens in Afghanistan have been frozen and most people have become jobless, many Afghan citizens are selling their possessions to buy some food items and face food insecurity even in urban areas. All these factors have led the diaspora in Pakistan to discontinue regular visits to their homeland, families, and friends, which ultimately has made the borders even tougher than our study could highlight.

Study Area and Communities
The field survey was conducted in Kohat District (1,112,452 inhabitants: [26], NW Pakistan. Most of the population lives in rural areas (75.71%). The average elevation of Kohat District is 489 m.a.s.l., and it is located between 70 • 34 and 72 • 17 E and 32 • 47 and 33 • 53 N ( Figure 6).

Study Area and Communities
The field survey was conducted in Kohat District (1,112,452 inhabitants: [26], NW Pakistan. Most of the population lives in rural areas (75.71%). The average elevation of Kohat District is 489 m.a.s.l., and it is located between 70°34′ and 72°17′ E and 32°47′ and 33°53′ N ( Figure 6). The area comprises a succession of irregular mountains, ranging from 610 to 1526 m which are separated by open valleys. In high mountainous areas winter is cold (−5 °C to 3 °C) while summer remains fresh (10 °C to 15 °C), in plain areas the summer is extremely hot with temperatures reaching 50 °C, while the winter remains mild. Due to the low precipitation rate, there are no running streams in the hills and water for agricultural purposes is scarce, as the soil is mostly sandy and stony (Figure 7). The vegetation consists of several xerophyte plants, such as Monotheca buxifolia, Withania coagulans, Senegalia modesta, Peganum harmala, and Calotropis procera. In plain areas, people sometimes are able to grow guava in gardens for economic purposes. Kohat has also a well-developed hydrocarbon extraction and mining. The area is home to various groups of people associated with different spoken languages and various religious faiths. Indeed, in the early sixteenth century three main Pathan tribes, namely Afridi, Banoori, and Bangash, settled in Kohat District, which remained a part of the Mughal Empire until the eighteenth century [27]. The area comprises a succession of irregular mountains, ranging from 610 to 1526 m which are separated by open valleys. In high mountainous areas winter is cold (−5 • C to 3 • C) while summer remains fresh (10 • C to 15 • C), in plain areas the summer is extremely hot with temperatures reaching 50 • C, while the winter remains mild. Due to the low precipitation rate, there are no running streams in the hills and water for agricultural purposes is scarce, as the soil is mostly sandy and stony (Figure 7). The vegetation consists of several xerophyte plants, such as Monotheca buxifolia, Withania coagulans, Senegalia modesta, Peganum harmala, and Calotropis procera. In plain areas, people sometimes are able to grow guava in gardens for economic purposes. Kohat has also a well-developed hydrocarbon extraction and mining. The area is home to various groups of people associated with different spoken languages and various religious faiths. Indeed, in the early sixteenth century three main Pathan tribes, namely Afridi, Banoori, and Bangash, settled in Kohat District, which remained a part of the Mughal Empire until the eighteenth century [27]. In 1978, after the takeover by the communist party in Afghanistan, Afghan citizens started migrating to neighboring countries such as Pakistan and Iran. As many as 193,000 Afghan refugees received asylum in Pakistan by the end of 1979 [28]. A significant acceleration in the influx of refugees occurred after the invasion of Afghanistan by Soviet forces when 80,000 to 90,000 Afghan Pathans crossed the Pakistani border every month between January and December 1980 [29]. In 1989, the number of Afghan refugees (3,270,000) reached its peak, which represented 3% of the total population of Pakistan at that time [30,31].
In 1992, over a period of six months, approximately 1.2 million Afghan refugees moved back to their own country from Pakistan after the fall of Kabul to Mujahidin, and by the start of 1994 the population had fallen from 3.2 million to 1.47 million. However, the migration of refugee communities back to Afghanistan occurred for only a short time because of the continuation of the civil war [32]. In late 2001, when the Taliban regime came to an end, the government of Pakistan devised a plan to send all Afghans back to Afghanistan. In 2002, an agreement was signed between united nations high commissioner for refugees (the Government of Afghanistan, and the Government of Pakistan, according to which all returns had to be performed willingly. However, in the summer of 2005, the Government of Pakistan started to close different refugee camps in different tribal agencies, such as North and South Waziristan, Bajaur, and Kurram, which resulted in the displacement of approximately 200,000 Afghan refugees, who then returned to Afghanistan. The Government of Pakistan constructed several dozen camps in the North-West Frontier Province that remained safe from the most critical problems (such as malnutrition and epidemics) that a largely displaced population face. Since the Soviets invaded Afghanistan, the Government of Pakistan has given basic rights to and has adopted a liberal policy toward the Afghan refugees. According to the 1973 constitution, only Pakistani citizens have a right to stay in their choice of residence and freedom of movement, but the Government of Pakistan grants the same privileges (i.e., to reside where one chooses, travel throughout the country, do business) to all registered Afghans. In 1978, after the takeover by the communist party in Afghanistan, Afghan citizens started migrating to neighboring countries such as Pakistan and Iran. As many as 193,000 Afghan refugees received asylum in Pakistan by the end of 1979 [28]. A significant acceleration in the influx of refugees occurred after the invasion of Afghanistan by Soviet forces when 80,000 to 90,000 Afghan Pathans crossed the Pakistani border every month between January and December 1980 [29]. In 1989, the number of Afghan refugees (3,270,000) reached its peak, which represented 3% of the total population of Pakistan at that time [30,31].
In 1992, over a period of six months, approximately 1.2 million Afghan refugees moved back to their own country from Pakistan after the fall of Kabul to Mujahidin, and by the start of 1994 the population had fallen from 3.2 million to 1.47 million. However, the migration of refugee communities back to Afghanistan occurred for only a short time because of the continuation of the civil war [32]. In late 2001, when the Taliban regime came to an end, the government of Pakistan devised a plan to send all Afghans back to Afghanistan. In 2002, an agreement was signed between united nations high commissioner for refugees (the Government of Afghanistan, and the Government of Pakistan, according to which all returns had to be performed willingly. However, in the summer of 2005, the Government of Pakistan started to close different refugee camps in different tribal agencies, such as North and South Waziristan, Bajaur, and Kurram, which resulted in the displacement of approximately 200,000 Afghan refugees, who then returned to Afghanistan. The Government of Pakistan constructed several dozen camps in the North-West Frontier Province that remained safe from the most critical problems (such as malnutrition and epidemics) that a largely displaced population face. Since the Soviets invaded Afghanistan, the Government of Pakistan has given basic rights to and has adopted a liberal policy toward the Afghan refugees. According to the 1973 constitution, only Pakistani citizens have a right to stay in their choice of residence and freedom of movement, but the Government of Pakistan grants the same privileges (i.e., to reside where one chooses, travel throughout the country, do business) to all registered Afghans.
In 2011, only 33% of the population was reported as still residing inside the camps, while the remaining 67% had moved into various urban and rural areas [33]. NW Pakistan hosts the majority of the Afghan diaspora (58.1%) in the country [34].

Data Collection
The field ethnobotanical study was conducted over a period of five months from February to June 2022. Ninety study participants were conveniently selected among middleaged and elderly community members, later adopting a snowball sampling technique. Convenience selection is a type of nonprobability or nonrandom sampling "where members of the target population that meet certain practical criteria, such as easy accessibility, geographical proximity, availability at a given time, or the willingness to participate are included for the purpose of the study" [35]. The interviewees, 30 individuals from each studied group, were mainly elderly adults ranging from 50 to 85 years of age because they were thought to be more knowledgeable in terms of LEK ( Table 3). The study participants belonged to three groups, i.e., Afghan Pathans inside the refugee camps (AI), Afghan Pathans outside the refugee camps (AO), and local Pathans, naturally living outside the refugee camps (PO). Afghan Pathans living outside the camps perform two types of pastoralism: one in which they herd their animals near their houses, and a second in which they rent pastures from local Pathans and send 2-3 family members with their herds into the mountain pastures, while the rest of the family remains in their homes. Afghan Pathans have little opportunity of sharing knowledge with local Pathans as they are not in close contact, despite the fact that they live in the same environment. Refugee groups have developed a certain social network to acquire some medicinal plant remedies in Kohat city, Kurram Agency (an area sharing a border with Kohat District), and from their homeland whenever they travel there to visit their friends and families.
All the interviews were conducted by the first author in the local language, Pashto, which was the first language of all the study participants.
Before commencing each interview, prior informed consent (PIC) was obtained verbally, and the Code of Ethics of the International Society of Ethnobiology (ISE) [36] was strictly followed. Information about folk wild plant use was obtained through free listing first, and semi-structured interviews afterward. The questions mostly focused on the medicinal, veterinary, and food uses of wild plants, in the present and in the past (e.g., childhood); additionally, some culturally relevant uses of wild plants as fuel and cosmetics/fragrances were recorded. For each botanical taxon, the researcher asked about the exact homemade preparations and uses. All available plants were collected while in the field, photographed, dried, and identified with the help of a plant taxonomist at the Department of Botany, University of Peshawar. Voucher specimens of all documented plants were prepared and subsequently deposited in the Herbarium of the Department of Botany at the University of Peshawar. Those taxa for which no specimen could be collected were identified via folk plant names, photographs, and detailed ecological and plant descriptions. Moreover, study participants gave their consent to publish the photos that they had taken on their own. The nomenclature followed World Flora Online [37] for plant taxa and Index Fungorum [38] for mushroom taxa, with family assignments aligned with the current Angiosperm Phylogeny Group IV recommendations [39].
This study has as a main limitation an unavoidable sampling bias with regard to gender. Despite the fact that women retain a large part of the folk plant knowledge, in this study, only a few women were interviewed. This is because in Pathan/Afghan cultures it is often inappropriate or undesirable for men to talk to women due to the extremely conservative attitude regarding social relations in the area from both a cultural (observing the strict Pashtunwali or Pakhtunwali code of conduct) and religious (within a rigid Sunni Muslim sphere) point of view. Unfortunately, it was impossible for our research group to involve Pathan female scientists, due to the fact that their families did not allow them to spend days and nights with male colleagues in the field. Nevertheless, we were able to interview a few elderly women outside of their houses when they were willing to talk.

Data Analysis
The documented wild plant ingredients were organized in MS Excel. We utilized emic food categories as much as possible. Raw snacks referred to (parts of) plants eaten on the spot without any processing. According to this definition, salads are not considered raw snacks, as they are eaten after some preparation and during meals. The data were categorized into four datasets: one containing all identified taxa, the second one comprised of the identified most frequently quoted taxa (reported by more than 40% of the informants), the third one containing wild food plant taxa, and the fourth one comprised medicinal and veterinary taxa. These four datasets, which were generated for each of the three considered groups, were compared through proportional Venn diagrams and the Jaccard Similarity Index (for each pair of datasets) following the application designed for the use of this ecological index in the ethnobotanical domain [40].

Jaccard Index =
Number o f plant taxa reported by both group A and group B Total number o f plant taxa reported by group A

Conclusions
The current research yielded two main findings. Firstly, there is a general convergence of the ethnobotanies of Afghan refugees living inside and outside the camps. This is possibly due to the fact that both refugee communities used to live in the same environment before relocating and have the same culture and customs.
However, the study shows some subtle differences between the refugees living outside and inside the camps: the former use more wild food plants, while the latter uses more medicinal plants.
This can be explained by two concomitant phenomena: (a) Afghan Pathans outside the camps still have extensive exposure to the natural environment, which allows them to forage more frequently and consistently; (b) Afghan Pathans inside the camps have limited access to health services-because of their disadvantaged economic conditionsand therefore rely more on traditional plant remedies for managing their health.
Secondly, the data show significant differences between Afghan and Pathan ethnobotanies, with the refugees knowing more about wild plant uses than local Pathans. In fact, until the recent past, refugees used to collect wild plants across different habitats in both Afghanistan and Pakistan, while this has never been the case for local Pathans. However, these Afghan transnational plants and human mobilities are fading, due to stricter border rules.
The main findings of the study suggest therefore that stakeholders should pay careful attention to facilitating (i) the proper settling of refugees, their engagement in farming, and thus increased exposure to the natural environment and plant foraging practices; and (ii) the transnational movement of refugees since this could facilitate the further resilience of their traditional medical plant systems; both of these measures could ultimately enhance refugees' food security, health status, and well-being.