Introduction

The sustainability of boreal ecosystems under various types of environmental management in the face of global climate change is currently the subject of wide discussion, and traditional land use is one of the key factors affecting natural ecosystems throughout human history. Varying from partly modified ecosystems to secondary communities, contemporary forests are the legacy of previous impacts (Bürgi et al., 2017; Johnson & Miyanishi, 2012; Josefsson et al., 2009; Leuschner & Ellenberg, 2017; Mausolf et al., 2018; Seidl et al., 2020). Fire was the major means employed by humans in the clearing of forests for settlements, enhancement of hunting conditions, and for cultivation, notably for slash-and-burn (SAB) agriculture (Coughlan et al., 2018; Hörnberg et al., 2018; Mellars, 1976; Pausas & Keeley, 2009). SAB has a long history in boreal forests throughout the world (Hörnberg et al., 2015; Sazonov, 1993; Smirnova et al., 2017; Sovetov, 1867; Tretyakov, 1932).

Slash-and-burn agriculture (alternately referred to as forest fallow cultivation, shifting cultivation, swidden agriculture) in the past accounts for the fire dynamics of the last 400–500 years, shifts in fire regimes, as well as the structure of and succession in existing boreal forests (Dietze et al., 2018; Lehtonen et al., 1996; Lehtonen & Huttunen, 1997; Parviainen, 1996). The north-western provinces of European Russia resembled the Nordic countries in terms of human colonization and transformation of forests, as well as fire dynamics (Lehtonen & Kolström, 2000; Ryzhkova et al., 2020; Smirnova et al., 2017; Tikkanen et al., 2014; Wallenius et al., 2004) until the nineteenth century, when wood became a valuable commercial resource in Nordic countries, leading to the cessation of SAB agriculture, and a consequent decrease in the number of fires (Lehtonen, 1998; Lehtonen & Huttunen, 1997; Rolstad et al., 2017; Storaunet et al., 2013). Nevertheless, at the end of the nineteenth century, northern areas of the Russian plain and the western piedmont of the Ural Mountains that had been inhabited for centuries remained under various agricultural regimes, although extensive forested areas were still preserved in the northern provinces (Vologda, Olonec, Arkhangelsk province, and northern uyezds of the Perm province). Colonization continued into the twentieth century, and SAB agriculture remained in use until the 1930s (Aleinikov et al., 2018; Chagin, 2017a, b; Igoshina, 1930; PRM, 1957; Sarapulov, 2014, 2015; Shildaev, 1917).

The impact and the legacy of SAB agriculture on the ecosystem remain ambiguous. Data on the number and size of cultivated plots, distance to settlements, and seasonal occurrence are needed to distinguish man-made from climate-related fires (Aleinikov, 2019a, b; Drobyshev et al., 2004; Lehtonen & Huttunen, 1997; Rolstad et al., 2017; Ryzhkova et al., 2022). At the local level, SAB practices impacted the structure and composition of the vegetation and the physical and chemical properties of the soils (Certini, 2005; Paal et al., 2011). Fields cleared with fire, cultivated, and subsequently gradually overgrown by trees and shrubs differ from the natural forest communities in composition, age, and natural disturbances (Uotila et al., 2002), and retain traces of previous impacts even after a hundred years (Aleinikov et al., 2018; Aleynikov et al., 2015; Čugunovs et al., 2017; Karpin et al., 2017). However, a comparison of vegetation on former SAB fields and that of intact forests shows no fundamental dissimilarity (Tomson, 2018), which may most likely be due to insufficient study.

The felling and subsequent burning of trees slightly increased the fragmentation of large forests at the landscape level and supported greater structural biodiversity (Hokkanen, 2006; Karpin et al., 2017; Tomson et al., 2015). SAB cultivation might also have limited the increase of the extent spruce-dominated forests over time by facilitating their wider distribution (Lindbladh et al., 2014). The frequent shift of cultivated plots increased the degree and area of impact, while at the same time accidental burning, often due to careless handling of fire, could be very destructive, shifting the natural fire regime of forests (Groven & Niklasson, 2005; Lehtonen & Huttunen, 1997; Pitkänen et al., 2003; Terrail et al., 2019).

We investigate the certain characteristics of SAB farming at the end of the nineteenth and beginning of the twentieth century in the western piedmont of the North Urals Mountains based on our examination of remaining archival documents. The SAB ban produced documentation of violations including information on the forest that stands around settlements, SAB techniques, and details of land use. We (i) discuss the environment, population, and land use of the area during this period, (ii) review the historical and ethnographic literature on the Northern Ural region, (iii) introduce data from unpublished archives, and (iv) evaluate the environmental information contained in the archives.

Contemporary forests in the study area consist of mosaics of various forest types with differing dynamics (Syrjänen et al., 1994). Despite the use of SAB by the local population (Chagin, 2017a), a large area of forest remained intact. We estimate the historical impact of traditional land use practices on the forests of the region. We analysed the timing and the features of SAB practices, the size of the plots used, and their distance from the nearest settlement. We also examined the relationship between SAB practices and uncontrolled forest fires. We assume that despite being illegal, SAB agriculture continued to be practiced on small household plots located in the vicinity of the existing settlements.

Materials and Methods

Study Area

The study area is located on the eastern edge of the Russian plain within the Kama-Pechora watershed (Fig. 1). Historically a part of the Perm province, the piedmonts of Northern Ural at the turn of the nineteenth-twentieth century were a unique mosaic of various longstanding farming systems despite the statutory ban on SAB and strict enforcement by the forest service.

Fig. 1
figure 1

The location of the Northern part of the Perm Province (A, B), and an orographic map of the area with settlements in 1889 (C)

The climate is distinctly continental with long and severe winters and short and cool summers, reflecting the mountainous terrain with an altitudinal sequence of climatic zones and significant variability of temperatures, precipitation, and wind speeds at short distances. Snow cover forms in mid-October and melts in early May, although frosts might occur until early July and return as early as August. In the mountains, the amount of precipitation is 800–1000 mm per year, decreasing to 500 mm per year on the flatlands. The average monthly temperature in January is -16.8 ℃, while in July it reaches + 15.8 ℃; the average annual temperature is -0.6 °C. According to the Koppen-Geiger classification scheme, the climate of the area is Dfb (Kottek et al., 2006).

Piedmont ridges along the main Ural ridge are 500–700 m high, with several reaching 1000 m. The complex orography of the territory leads to diversity in vegetation communities, and the altitudinal sequence of vegetation communities is clearly seen on the high ridges, where the piedmonts are covered with forests dominated by dark coniferous spruce (Picea obovata Ledeb.), along with firs (Abies sibirica Ledeb.), Siberian pines (Pinus sibirica Du Tour), and a small admixture of birch (spp. Betula). Smooth hillsides are covered by green moss, large ferns, and tall herb spruce and fir forests with P. sibirica. The lowland between the ridges is overgrown by Sphagnum-type spruce forests with B. pubescens and P. sibirica (Smirnova et al., 2017). At higher levels, taiga forests are gradually replaced by mountain taigas, and even higher by subalpine meadows. Besides the orography, fires and felling drive vegetation formation, leading to the replacement of dark coniferous forests by birch and aspen dominated forests in piedmonts, and by pine forests in sandy river valleys.

Severe climatic and orographic conditions inhibited colonization of the piedmonts in the Northern Ural Mountains, which were populated much later than the southern parts of the Ural Mountains and the Russian plain. In the Middle Ages, the area was inhabited by tribes of Voguls (Mansi), nomads dependent on hunting, fishing, and gathering (Aleinikov, 2017).

The history of the forest use in the area reflects the development of forestry in Russia. Until the eighteenth century, forestry in Russia was regulated by separate laws. Some types of trees like oak, maple, elm, larch, and pine of large diameter were declared protected in 1703 and were available for state use only (particularly for shipbuilding). The felling of these trees less than 53 km from large rivers and less than 21 km along small ones was banned. The Forest Department was created in 1789, and in 1802 the Forest Regulation was established. Categories of forests, such as forests for shipbuilding, mountain forests, forests of state factories, forests of military establishments, as well as state forests, were assigned to different government departments. The provinces were divided into districts, and the districts into forests. In 1828 the positions of forester and of forest guard were introduced in state forests in 1869 (Stoletie, 1898).

The forests of the piedmonts in Northern Ural currently have different degrees of preservation. The most valuable and best-preserved forests are in the Vishera Nature Reserve and the world heritage site “Virgin Komi forests” Pechora-Ilych State Nature Reserve (UNESCO, 1995). Both nature reserves have biosphere status. Smaller parts of the forests were felled or burned during the twentieth century and are now occupied by secondary-growth small-leaved trees.

Settlement History

Until fifteenth century, when they were annexed by the Russian state, the western part of the study area was inhabited by Komi-Permyak hunters and fishers, while Mansi reindeer herders occupied the mountain area. Some of the subsequent Russian settlers formed communes with the existing Komi-Permyak population. Mass immigration of Russian peasants began in the sixteenth-seventeenth centuries (Oborin, 1990), and the first Russian settlements appeared in the second half of the seventeenth century. Due to the absence of roads, all settlements were located close to navigable rivers: upstream of the Kolva River (Korepino) and its tributaries, on the river Beryozovaya (Beryozovo), the river Visherka (Fadino), and the river Pechora (Ust’-Unja), all registered in the 1700s (Aleinikov, 2021; Chagin, 2017b). In the nineteenth century, the population in the Kolva and Vishera basins grew rapidly. By 1900, 43 settlements with a total of 2977 inhabitants were registered (Table 1). The Komi-Permyak population continued to live in permanent settlements in neighboring territories of the Upper Kama basin and Komi-Zyryans, in the Upper Vychegda basin (Rittih, 1875).

Table 1 The population and size of the farmer lands along rivers in 1900

At the end of the nineteenth century, 99% of the study area was covered by forests (Sokolov, 1909), and the lands in the Kolva and Pechora River basins were legally divided between state forests, under the authority of a sparsely staffed Forest Service (99.4%), and farming land (0.6%) (Fig. 2) (Kokovtsov, 1876). The Kolvinskoe forest district controlled a total area of 2,038,168 hectares. The status and use of state forests depended on the availability of markets and differed between neighbouring river basins. In the Kolva River basin, state forests were depleted by active logging for salt production, fuel for metallurgical factories located downstream, as well as for saleable timber. There were no factories in the Pechora River basin and therefore the state forests were much better preserved (Aleinikov, 2017; Sokolov, 1909). Leases for arable lands and hayfields cultivated on state forests before 1880 were allocated to farmers by the local forestry department. After 1885 development of new agricultural lands was strictly prohibited.

Fig. 2
figure 2

The location of farmer lands and settlements in between the Kolva and Pechora rivers at the end of the nineteenth century (right) and the detailed land-use plan of an individual farmer (left)

Farming lands were allocated from state forests around officially recognized and demarcated new settlements including building sites, hayfields, pastures, arable lands, and farmers’ forests, clearly distinguished within the state forests (Fig. 2, left). In 1900, the total area of farmers’ land within the Kolvinskoe forest district was 15,500 hectares, of which about one third was farmers’ forests (Table 1).

Data Sources: Written Archive Documents, Spatial Data, and Bibliographical Sources

We analysed archive inventories of the cases (AIC) and individual cases of forest regulation violations (ICV) between the years 1880 and 1894 from the Kolvinsky forest district, and the years 1894 and 1910 from the Chusovskoj, Kolvinsky, and Verhne-Pechorsky forest districts of Cherdyn municipality in Perm province. The documents contain information on new settlements, road construction, hayfields, arable lands, and fishing places, which were leased to the residents by the authorities. These data provide an understanding of traditional land use practices and the extent of exploitation of the area by local inhabitants.

Since many records from the archives have been lost over the years, our of inventory of cases (AIC) from the Kolvinskoe forest district from 1842 to 1894 (ChRM, 1894) is an important source of data on land-use and forest management, especially as regards records of the protocols for cases of forest regulation violations, such as illegal cutting, fires, cattle grazing, logging, hunting, fishing, haying, theft, illegal slash-and-burn cases, as well as data about on forest ownership (i.e., state or private) as well as the violators’ personal data and place of residence. Due to the absence of roads and navigable rivers, wood for building in the region had to be transported between the middle of October and the end of April by sledding on snow cover and frozen rivers (Materials, 1901; Spravochnik, 1990). Therefore, we include all records of unapproved felling and arson that occurred between May and September. Our theory is also supported by ICV data where all summer slashes correspond to the first stage of SAB. However, contemporary regulations stipulated that forest regulation violations had to be destroyed after 15 years. Around 100 ICVs were fortuitously retained in the archive, of which 32 cases report SAB events. Each case comprises a protocol on the specific forest regulation violation, an internal note, the questioning of suspected farmers, and the court’s decision. Each case comprised between five and 35 pages, depending on the complexity of the violation, the identification of a malefactor, and the extent of damage. The protocols of SAB cases detail the nature of the violation, noting extent of the cutting or fire-setting, who identified the case and when, the location of the plot relative to the nearest village, as well as dimensions of the plot. There is information on species composition of the forest, diameter of the trees on the site, what was cut down and burned, and what remained standing. The distance to the settlement nearest the SAB plot was recorded in 24 cases. The exact month of the forest burning is mentioned in 22 records, of which 11 provide exact dates.

Detailed mapping of the area began only in the 1930s. We used local maps of three forest districts made at the beginning of the century, which show settlements and distinguish between state forests and farmers lands, together with the modern topographic maps, to locate settlements. We used data from the first census survey of this region (Materials, 1901; Sbornik, 1889) and the ethnographic studies of the middle of nineteenth and early twentieth centuries (Beldytsky, 1901; Belousov, 1915; Krylov, 1926; Onchukov, 1901; Teploukhov 1856; Varopai, 1880) for the number of inhabitants and list of settlements.

Data Handling and Analysis

The ICVs provided precise distance to the SAB site from the nearest settlement, size of the site, and dates of the case. At each SAB plot, the species (spruce, fir, birch, aspen) and condition (cut down, burned, or standing) of every damaged tree was used to assess economic damage. All damaged wood was categorized according to size and variety: tree-length, brushwood, sticks, poles, firewood, and deadwood.

Until 1918, measurements of length, area, and volume were recorded in now obsolete units of measurement (Shevtsov, 2017), which we converted to the modern metric system. Similarly, we added 12 days to the dates recorded according to the Julian calendar to conform with the Gregorian calendar.

Results and Discussion

Overview of the Cases of Violations Registered in the AIC

Population, Forest Cover, and Land-Use Patterns at the End of the Nineteenth Century

In Perm province, SAB was allowed only in farmers’ forests, which were not large enough to provide needed timber and firewood as well as accommodate annual rotation and fallow of SAB plots. As village populations expanded, greater pressure was put on farmers’ forests and fewer areas of unused land were available per household. In response to increased incidences of illegal SAB, in 1884 a special law allowing the farmers of the Perm province to take additional forest plots for SAB was issued, giving the state legal control over timber felling and use of fires. However, the farmers ignored the new law and continued to practice SAB illegally in the forests.

However, even the new law (Tagantsev, 1886; Faleev, 1912) had no specific penalty clauses for SAB in state forests. If a farmer was caught felling a tree stand, then he was punished only for unauthorized felling, and had to pay double the cost of the felled or damaged timber and a fine. If the farmer managed not only to cut down the forest stand, but also to set fire to the site, then in addition to unauthorized felling, he was charged with careless handling of fire and an additional fine was imposed. In cases where a farmer did not have the means to pay fines, he was sent to public works or held under arrest for three days.

We estimated the overall number of SAB cases and illegal summer slashes and their associated settlements in specific years from the AIC. Between 1885 and 1894, 175 SAB cases were recorded in Kolvinskiy forest district (Table 2), mostly summer slashes (123), the first stage in the SAB cycle. Of these, 15 plots were located near 35 settlements, with no exact location mentioned for 23 plots. In only 52 records were latter stages of SAB registered. The maximum number of SAB cases was registered in 1887 and 1892 and of summer slashes 1887 and 1894. Only summer slashes were registered in 1890 and 1891.

Table 2 The number of SAB cases and summer slashes recorded in the AIC between the years 1885 and 1894

Our AIC analysis demonstrated that despite total prohibition of SAB agriculture, farmers in most settlements along the rivers Kolva and Pechora persisted with SAB agriculture in violation of forest regulations (Fig. 3). Most of the cases were registered by settlements located along the Kolva river. Only 13 cases were located along the Beryozovaya river, 11 cases along the Pechora river, six cases along the Unja river, and only five cases along the Visherka river.

Fig. 3
figure 3

The series of maps shows the number of SAB cases by year registered by the settlements mentioned in the archive data between the years 1885 and 1894. Black dots mark settlements

Analysis of the Protocols of ICV

The forest service identified the locations of violations by the smoke of the fires. Most protocols (65%) recorded the stage of burning or gathering the remains of the burned forest, while 17% reported the stage of the forest cutting, 15% reported the stage of crop sowing, and only 3% reported the harvesting stage.

Site Selection and the Features of SAB Process

The distance from settlements to possible SAB sites was generally rather short, despite the absence of roads suitable for cartage: the shortest registered distance was 1.07 km, while the longest was 7.5 km: the mean distance was 3.4 km, and 10 sites were located closer than 2.5 km, nine at a distance of 2.5 to 5 km, and another five at 5 to 7 km.

Farmers preferred plots dominated by dark coniferous spruce and fir, with an admixture of birch and aspen for the SAB sites (Table 3). However, most of the SAB sites were young secondary forest. Both the age of the forest and the distance to the site were important when choosing a SAB site. Farmers had to look for a compromise: the site should be close enough to the settlement, but also not easily detected by the forest guard.

Table 3 Summary of the ICV protocols: plot size, distance to the settlement, characteristics of the stand on the SAB plots and types of damage registered in the protocol

Selective cuttings were also common in forests with huge birches and aspens (more than 25 cm in diameter). Large trees were left standing during forest cutting and burning and remained standing on the SAB plot as dead wood until they fell naturally. Ruma (1903) suggested that these trees were left for shading cereals. However, it is more likely they were left because the felling of mature trees requires more time, and once felled trees do not have time to dry out in one season, and will not burn with less mature growth.

Timing of SAB

Our analysis of the archive documents revealed that a two-year SAB cycle was the most common in the area: one year for identification of a suitable site and forest cutting; the second year for burning and clearing. However, one case noted that the forest was felled in 1889 but burnt only three years later in 1892.

Trees were usually cut on the chosen plots between July and October of the first year. Through the summer of the next year farmers burned fallen trees. The burning likely started earlier when the spring was dry and warm. The earliest registered date of burning is 1st June 1888. All registered burnings occurred in June and July. If the summer months were dry, the burnings were arranged in those months. Thus, in 1888 the burnings were recorded in the first half of June as well as during the whole of July, the driest and hottest periods in the piedmont of the Northern Ural.

The choice of the crop depended on the farmers’ needs. We found reports of several rye and turnip plantations on SAB sites. According to other studies in that area, the most popular crops were barley and rye, occasionally along with turnip and flax to get the maximum yield in the first year (PRM, 1957; Sokolov, 1909; Varopai, 1880). The SAB ban resulted in the sites being used only once before being abandoned.

The Sizes of SAB Plots

The archives provide information on the size of 21 burned and subsequently cultivated plots: size varies from 32 × 11 m to 120 × 110 m, with a median size of 80 × 55 m. The plots were mostly of elongated shape, but one plot had one side five times longer than another (134 × 25 m). The area of the plots varies from 0.03 ha to 2.70 ha, with a median of 0.55 ha. A similar average area of 0.33 ha was also noted in Olonets province (now the Republic of Karelia) (Krivonozhenko, 2020).

SAB Cultivation in the Piedmont of Northern Ural at the End of the Nineteenth Century

The western piedmont of the Northern Ural has a harsh climate with early frosts, which posed a risk to farming, and the difficult orographic conditions allow for only a small amount of agricultural land, traditionally occupied by local tribes of Mansi pushed east by the Russian population (Aleinikov, 2017, 2019a, b; Sokolova, 2009). The area was very sparsely populated and even up to the middle of the nineteenth century there was enough space available for new settlements of Russian farmers who sailed up the rivers to clear suitable land with fire. Until the prohibition of free settlement in the state forest in 1880, farmers’ allotments were not limited in size, and farmers could settle freely, often with another household, on land for cultivation (Sbornik, 1889; Shildaev, 1917).

However, the farmers faced harsh climatic, soil, and orographic conditions that limited what they were able to cultivate (Materials, 1901; Krylov, 1926; Milovanovich, 1926; Chagin, 2017a). The amount of fertiliser produced by the limited number of livestock was sufficient to increase the yield of crops by 3–4 times more than what was seeded (Sbornik, 1889). Agricultural land was in increasingly short supply due to the rapid growth of the population at the end of the nineteenth century (Aleinikov, 2021; Chagin, 2017b), when the local population turned to SAB cultivation in state forests, which produced yields as large as 25–30 times more than seeded (Sovetov, 1867; Varopai, 1880). It is likely that their ability to practice SAB agriculture was favoured by the very limited resources of the forest service and the vast territories in which the SAB sites were not always detected.

Another factor favouring the continuation of SAB cultivation was the low commercial value of the forest in the region due to poor access to transportation. Thus, while 30% of the forest was accessible in Kolvinsky forest district, in Chusovsky and Upper-Pechorsky only 15% could be used (Bokov, 1898). A portion of the timber was used by the salt-making plants and barge construction, and a small amount was sold on the southern markets and to the local population (Batuev, 1902; Aleinikov et al., 2018). The wood of the Upper Pechora basin was not used commercially at all. The local population of Upper Pechora had no economic motivation to protect the forests, compared to populations in Scandinavia who gradually abandoned SAB agriculture due to an increase in the commercial value of the wood (Storaunet et al., 2013).

The number of SAB violations recorded varies greatly around different rivers. The number of cases over 10 years was 10 times fewer around the Pechora and Beryozovaya rivers, and 20 times fewer around the Visherka and Unja rivers that those around the Kolva river (Table 1). The difference was likely due to the population density on the Kolva river, which was two to four times greater than that of other rivers.

The number of SAB plots varies significantly from year to year, most likely due to weather conditions. The largest number of settlements (15) with associated SAB plots was registered in 1887, whereas in 1886 only four settlements had reported SAB cases. The archives also contain information on forest fires, indicating the largest number of forest fires recorded in 1885, although in1886 no fires were registered. The maximum number of SAB cases was registered in 1894, possibly due to a reorganization of Kolvinskoe forest district to strengthen forest supervision. However, the relationship between the number of SAB cases and climatic conditions needs further investigation.

Environmental Aspects of SAB

The analysed archival materials throw light on ecological aspects of SAB agriculture in the study area. The forest composition data we derived from the ICV reveal that most of the SAB plots (17 out of 21) were secondary communities. Young dark coniferous communities and small-leaved forests, likely restored communities after a previous SAB or forest fire, were recorded in 23 of the 21 cases. The other11 cases were in mixed forests with the upper layer formed by birch and aspen, with spruce, fir, and Siberian pine forming the second or understory, such as might appear after fires or previous SAB events 70–100 years earlier (Aleinikov et al., 2018). Vegetation communities on all plots mentioned in the ICV had already been modified by previous land-use practices and fires.

Until the middle of the nineteenth century, farmers used the most valuable wood for their own needs (Sonni, 1839; Sovetov, 1867). However, with the complete ban on SAB agriculture, even freshly cut logs near a settlement could attract the attention of the forest authorities, and farmers chose to leave all cut wood on the site.

It is important to note the specific approach of local farmers to tree felling: all conifer trees smaller than 25 cm in diameter were cut down and burned (firs and spruces burn much better than birch and aspen and are also more valuable for fertilizing the soil) while the more solid trees (mostly birches and aspens) were left standing. Such practices were described in neighbouring territories, inhabited by Komi-Zyryans and Komi-Permyaks (Belitser, 1958; Oborin, 1956) where, due to long-time exploration, a particular ecosystem called the bushland formed in contrast to the hemi-boreal forests (Tomson et al., 2018). As a result, specific communities with single standing old and burnt birch and aspen formed in the piedmont of Northern Ural.

The total prohibition of SAB agriculture most likely led farmers to create smaller SAB sites than before the ban (e.g., 400–500 m to 100–150 m according to Belitser, 1958). This slowed the transformation of forests and contributed to their preservation, but at the same time increased fragmentation of plant communities led to a combination of plant communities specific to SAB agriculture forming in the state forests around the settlements, and post-fire communities developing in the adjacent territories. Our data show most burnings were registered during June and July, when the weather was warm and dry.

Most SAB sites were located within a 5 km radius of the settlements. In the absence of roads, the farmers were forced balance the costs of being caught by forest guards against the costs of their labour. At the same time, the short distances to the settlements indicate that the forest remained healthy while sustaining sufficient SAB sites. Since cuttings were officially allowed in all forests in the earlier populated areas of Vologda province, by end of the nineteenth century all potential sites around settlements had already been exploited, forcing farmers to seek SAB sites and fuel for burning much further from their settlements (Kolonist, 1913; Lyubomudrov, 1889).

SAB Agriculture Cases and Uncontrolled Forest Fires

A number of earlier studies have shown that in the eighteenth and nineteenth centuries, SAB agriculture was considered extremely harmful for forests, since it was assumed to be the cause of more widespread catastrophic fires (Sonni, 1839; Teploukhov 1856; Lyubomudrov, 1889; Kolonist, 1913; Tretyakov, 1932). In some archive cases there are notes on fires caused by burning to clear plots for agriculture. In certain years, along with numerous reports of SAB, numerous forest fires were also noted. The AIC does not indicate the causes of the fires, but possible causes are discussed in some documents. For instance, in 1885, a large fire was recorded in an area of about 1,700 hectares during a severe drought that lasted more than a month due, according to forestry workers, to careless supervision of local residents, who were actively working on hayfields along rivers and fields at the time (see also Aleynikov et al., 2017). However, even in dry summers, most farmers cleared at least a small SAB, which could lead to catastrophic fires. For example, in 1934, about 10,000 hectares of dark coniferous forests burned in the Pechora River basin because of an out-of-control SAB fire (Aleinikov et al., 2018). It is reasonable to assume that agricultural activity in the forest, irrespective of population density, is the main cause of the unwanted spread of fires, especially during dry years (Drobyshev et al., 2015). During relatively wetter years, occasional small fires may have occurred around settlements but did not spread further to the forests.

Our analysis of the AIC showed that SAB was most often found either during felling in the first year or during site burning the following year. Once the ban came into effect in the last third of the nineteenth century, farmers made greater efforts to avoid accidental forest fires, which would have led to identification of the SAB area and themselves, and chose burning technologies with lower risks of causing accidental forest fires. The fragmentary data from archival sources do not unambiguously confirm or rule out any connection between SAB cultivation and forest fires.

Conclusions

The historical documents (AIC and ICV) we examined relating to SAB agriculture in the Northern Urals are important sources of information about contemporary society, environment, and land use. Our analysis showed that along the western macro-slope of the Northern Urals, SAB plots were actively used for short-term benefits of greater yields in the difficult orographic and climatic conditions of the area. Once the practice was banned, farmers had to balance the costs of violating the law against these short-term benefits.

The identified SAB sites were generally small (median = 0.5 ha) and were located within 7.5 km of the settlements. After the total SAB ban, sites were used only once before being abandoned, forming a mosaic of communities at different stages of post-burn succession in the vicinity of the settlements. In the wet and cold conditions of the Northern Urals piedmont, the trees on a SAB site could only be burned in warm and dry springs, when farmers in neighbouring villages would simultaneously burn their sites. Although the archival data we present do not unambiguously confirm or rule out the significant role of SAB in the fire dynamics of dark coniferous forests, it can be argued that it is one of the most common causes of fires close to settlements.

Our study provides a new perspective on the history of SAB in boreal forests. Despite the available evidence of the destructiveness of the SAB system for boreal forests, the territories in the east of the European part of Russia were minimally affected by its use due to the late settlement of the territory, the low population density, and the prohibition of SAB outside designated farmers’ forests. Our analysis of archival documents provides further understanding of the transformation of boreal forests across their distribution, as well as facilitating the identification of remaining intact boreal forests for studies of their ecosystem functions and services. The identification of old SAB sites can enable assessment of the long-term environmental consequences of SAB agriculture in boreal forests, including plant community structure and composition, and physical and chemical soil characteristics.