Structural alterations in land use have been intrinsically associated with a growing demand for food [1
]. Increasingly, contemporary processes of progressing urbanization and industrialization have been aggravating the conflicts between different functional land types [3
]. As land systems represent a critical intersection between economic and ecological systems [4
], land distribution patterns are becoming more vulnerable to a variety of environmental and social issues. Out of the one-fifth of the world’s total land surface, which is potentially suitable for crop production, more than half is already actively cultivated [5
]. Further agricultural expansion is hampered by natural and geographical factors [6
], pervasive land-use change impacts [7
], high economic costs [8
], and infrastructure constraints [9
]. At the same time, according to DeFries et al. [10
], Ajani [11
], and Lambin [12
], agricultural production tends to face increasing competition for land with other types of land use. Over recent decades, many scholars and practitioners, including Platt [13
], Briggs and Yurman [14
], Vining et al. [15
], and Sioen et al. [16
], among others, have been reporting the irreversible removal of substantial areas of land previously used for agriculture to urban, industrial, infrastructure, and other types of use instead. Urbanization and industrialization intensify competition between agricultural and non-agricultural land-use practices [17
]. Along with industrial development and urban sprawl, there are significant alterations of land use far beyond city limits that result in arable land loss [18
Generally, at a regional scale, agricultural lands do not strictly compete with other categories for the same land areas due to the specific climate, soil, and topographical requirements for farming. However, in land-abundant and climate-diverse countries, the geographical distribution of agricultural land use tends to adjust to better match land quality [19
]. Russia is aa good example of aa country that can be used to demonstrate this fact. Agriculture abandonment in vast northern and eastern areas has occurred in parallel with a concentration of intensive agriculture in fertile lands in the southern, western, and central regions of the country. In Russia, agricultural lands only represent 12.96% of the total national land fund (cropland at 7.16%, rangeland at 3.99%, hayfields at 1.40%, fallow at 0.28%, and perennial plantings at 0.11%). Per-territory concentrations of agricultural land vary from 75.32% in the Southern Federal District and 70.96% in the North Caucasian Federal District to only 4.05% in the Northwestern Federal District and 1.30% in the Far Eastern Federal District.
We clarified the definitions of the main terms used in this study as follows:
District—A type of supraregional administrative division of Russia, which includes several territories based on a geographical principle (currently, eight federal districts exist).
Land distribution—how lands of particular categories are spread out in a country, district, or territory.
Land fund—the total of available land resources in a country, district, or territory.
Land fund composition—a division of a land fund into land categories.
Land use—the total of arrangements, activities, and inputs that people undertake in a certain land cover type.
Territory—an umbrella term to designate various types of administrative divisions of the Russian Federation (oblasts, krais, republics, autonomous districts, and autonomous republics).
The disproportions of agricultural land distribution are, to some extent, caused by economic factors, not only geographic and natural conditions. Similar to most post-socialist countries, Russia has experienced dramatic changes in land ownership and land tenure since the early 1990s. Among the principal transformations, Lerman and Shagaida [20
] have outlined the privatization of agricultural land, rights to agricultural land for individual landowners, and the removal of prohibitions on buying and selling land. The land market has responded positively to the liberalization with an increase in transactions between individual landowners [20
]. However, the domination of shared and joint land ownership has weakened the role of the state in controlling land use [21
] and has increased the fragmentation of public land property into many scattered units [22
]. Almost twelve million land shares (certificates) were distributed between rural individuals and former employees of collective and state farms [23
]. According to Trukhachev et al. [23
], Lerman and Shagaida [20
], Rozhkov [24
], and Visser et al. [25
], land reform in Russia has significantly contributed to structural variations in the composition of land funds. The proportion of agricultural land in the total land fund has been declining due to a loss of arable land, particularly in the vast areas of the Far Eastern Federal District and the Siberian Federal District [26
]. From 1990 to 2000, the rate of land abandonment in Russia was above 30%, one of the highest among the economies in transition [27
]. Milanova [28
] reported a decrease in the cropped area for all crops during the 1990s due to the changes in land tenure and stagnation of the agricultural sector. A drastic decline in livestock production resulted in a reduction of hayfields and rangelands. Vast areas of arable land were abandoned due to land degradation. In some territories in the central, northern, and eastern parts of the country, humus content dropped by 50%. Prishchepov et al. [29
] revealed the correlation between the spatial distribution of abandoned croplands and natural factors, such as inadequate precipitation and shorter growing periods, in both Siberia and eastern parts of the country. As many farms were situated in the boreal zone, some of the abandoned lands have experienced shrub and tree encroachment [30
Many experts report an aggravated environmental degradation of agricultural lands due to over-exploitation [31
]. The changes in land cover and land use in forest-steppe and steppe vegetation zones (agriculture-oriented territories of southern Russia, the European center, and southern parts of Ural and Siberia) have been driven by extensive farming. Milanova [28
] and Milanova et al. [33
] reported that up to 90% of lands in some territories were converted to crop production. However, where environmental concerns of land use are mentioned in either federal or regional legislation, they predominantly relate to reducing industrial emissions or waste disposal in urban and suburban areas, not to agricultural land use [34
]. Over 40 million hectares of cropland is now abandoned in Russia, and another 58 million is eroded. Land degradation, along with desertification due to irrational land use, poses serious environmental, economic, and social threats in the long-term. Griewald et al. [34
] argued that the land use context in Russia did not support a transition towards sustainable land management, i.e., a “use of land resources, including soils, water, animals, and plants, for the production of goods to meet changing human needs, while simultaneously ensuring the long-term productive potential of these resources and the maintenance of their environmental functions” [35
]. The urban expansion causes shrinkage of arable and other categories of agricultural land [36
], which are transferred to various non-agricultural types of land use. A considerable amount of agricultural land loss due to urbanization and industrialization takes place on fertile soil [37
] and irrigated lands [38
]. In return, the increase in agricultural land acreage occurs on soils that are lower in terms of their fertility. Prishchepov et al. [39
], Brueckner [40
], and Brown et al. [41
] raised concern over the growing concentration of arable land in smaller and more fragmented locations in proximity to urban and industrialized areas. Erma et al. [42
] reported many cases where residential settlements occupied agricultural land in southern and central parts of the country, which are known as the breadbasket regions of Russia.
With increased variability in the composition of land funds, a reliance on research in this area has become more critical. In a series of empirical studies, many authors, including Verburg et al. [43
], Van Doorn and Bakker [44
], Nainggolan et al. [45
], and Diogo and Koomen [46
], among others, have attempted to construct hypotheses about the relationship between proximate driving forces and agricultural land-use patterns. The problem is that the established hypotheses do not adequately explain the causality between land-use processes and the compositions of land funds at different regional scales. In transition economies, including Russia, where land reforms have dramatically changed the distribution of the land inventory in recent decades [42
], variations in agricultural lands due to the pressure of non-agricultural land use have remained under-investigated. The composition of agricultural land funds has commonly been considered out of a non-agricultural context [47
], instead of exploring the interactions between the proportions of agricultural, urban, infrastructure, and industrial lands. Most of the studies have applied a proportion of agricultural land in a land fund as a core territorial specification without further testing for alternative non-agricultural land use variables [4
]. Therefore, in regional studies, a knowledge gap has emerged in terms of how the variations in the compositions of land funds may be tracked with an aim to optimize agricultural land use. A more explicit focus on the relationships between land categories is required to be able to explain and predict land system dynamics in diverse locations [49
]. With this background, in the case of Russia, this study aimed to contribute to the body of knowledge on regional scale land uses by identifying structural variations in the compositions of territory land funds and revealing the interdependencies between the proportions of agricultural, on the one side, and urban, industrial, and other types of land on the other.
The results, as expected, demonstrated that the compositions of the land funds in Russia vary across territories. Echoing Bichler et al. [100
], Chu [101
], Smith et al. [102
], and Bakker et al. [103
], we found that the distribution of agricultural lands is largely affected by natural factors, while agricultural lands are spread unevenly across the country. At a regional scale, belt-type concentrations of cropland suggest an agriculture-focused land distribution pattern in the southern and central areas of Russia. This is consistent with the observations of Rounsevell et al. [104
] and White and Engelen [105
], who revealed that agricultural land use tends to become concentrated in locations, reflecting the influence of natural factors and neighboring land distribution patterns. Nevertheless, in particular territories, the proportion of agricultural lands in the land funds do not match the type of agricultural land activity.
Emulating earlier studies by Mazurkin and Mihailova [70
], Shishkina et al. [74
], Mazurkin [69
], and Buckett [71
], we revealed that the application of a land activity parameter could result in creating a picture of land distribution patterns that are different from that which might be expected from the knowledge of the proportions of individual land categories. Therefore, land distribution change maps are not sufficient to capture specific finer-scale variations in the compositions of land funds at a regional scale. In Russia, land tenure and demand for land have been the principal economic proxies to map agricultural land distribution. According to Shagaida [107
], the demand for agricultural land varies significantly across Russia’s territories, depending on the degree of land consolidation. In the course of land reform, the previously dominant state farms have transformed the organizational form of their land use but still have persisted as the backbone of the agricultural sector [34
]. In the embryonic land market in the 1990–2000s, the establishment of new land tenure patterns had not involved immediate changes in the distribution of land from big ex-Soviet agricultural enterprises to individual owners [107
]. Since land certificates do not specify land plots, most of the shareowners have not withdrawn their land property from joint use by former collective farms. Over 70% of land in Russia is still used by large enterprises for rent, 25% is contributed to the capital of large enterprises, and only 4% is retained by private owners [109
]. In the breadbasket southern and central European territories of Russia, large agricultural holding companies have aggregated even more agricultural land property when compared to the Soviet period [110
To a large extent, the existing demand-based distribution matches the land activity map (Figure 3
), as the highest demand for land is identified in the central parts of the country close to Moscow. This demand primarily exists due to non-agricultural businesses. For type I and II territories, this correlates well with the finding of strong links between the proportions of agricultural land categories, on one side, and those of residential, industrial, and infrastructure lands on the other. In type III southern locus (Krasnodar), Lerman and Shagaida [20
] reported high demand for land among corporate farms. In that classification, type I and II territories are considered as less developed areas in terms of agricultural production (sometimes even as “agriculturally depressed regions” ([20
] p. 20)), where corporate farms tend to reduce their holdings and abandon land plots. Our results, on the contrary, demonstrated that in the south of European Russia, where the concentration of croplands is the highest, agricultural land activity is lower compared to many other territories of the country.
In the territories where a high proportion of croplands coexist with low agricultural land activity, many of the variations in the composition of a land fund could be explained by socio-economic factors. Van de Steeg et al. [111
] and Gärtner et al. [112
] confirmed that the distribution of agricultural land strongly correlates with the level of rural development, proximity to economic and market centers, urbanization, and the demand for agricultural land from non-agricultural industries. Our study revealed correlations between the proportions of agricultural and urban lands across type I–III territories, which could represent losing agricultural land due to urban development. In type II territories, the compositions of agricultural land funds are more affected by urban development than the compositions of type I and III. These results supported the findings of Daniels [113
], Su et al. [114
], Yeh and Huang [115
], and Dredge [116
], i.e., the proximity to urban development can be a powerful predictor of changes in agricultural land use. Many scholars, including Parsipour et al. [117
], Li et al. [118
], and Al-Kofahi et al. [119
], among others, agree that the accelerating urbanization has been causing increasingly harmful effects on agricultural lands. In the case of Russia, we did not reveal the acceleration of agriculture land loss in urbanized type I–III territories. What was revealed, however, was the strengthening of the correlation between the variations in the compositions of agricultural land funds and residential, industrial, and infrastructure lands. As Zubair et al. [120
] and Lucero and Tarlock [121
] forecasted, such stronger associations would continue to put increasing pressure on agricultural lands and result in more fragmented agricultural land use in the future.
Along with urbanization, an orientation of a land fund composition towards agricultural production is determined by the population density [111
]. In urbanized type I and II territories, agricultural land use is affected by the variations in the acreage of residential lands. In agriculture-oriented Krasnodar, Rostov, and Stavropol, the changes in agricultural land fund compositions are mainly linked with those of lands for transportation and communication. This result was consistent with what Ramadani and Bytyqi [123
], Li et al. [118
], and Al-Kofahi et al. [119
] reported when assessing the effects of more significant concentrations of the population on the lower proportions of agricultural lands in a land fund.
Reversely, Meyfroidt et al. [124
] and Nguyen et al. [125
] revealed that in the industrialized areas in Russia, where the density of population is lower, the concentration of abandoned lands is higher. There is an array of studies that have reported a link between industrial growth, changes in agricultural land distribution, and the degradation of farming opportunities internationally. Explicitly, Oyebanji et al. [126
] confirmed the existence of a positive long-term relationship between industrialization and land loss in Nigeria. Deng and Li [127
] revealed that the soil sealing effect has resulted from industrial and infrastructure construction in China, while Müller and Sikor [128
], Milanova et al. [33
], and Müller et al. [129
] linked changes in agricultural land distribution and agricultural abandonment in EU countries with unfavorable environmental conditions due to increasing industrialization. The expansion of urban and industrial infrastructure not only triggers agriculture-to-urban and agriculture-to-industry land transfers but also leads to the overexploitation and degradation of remaining agricultural lands [127
]. Many areas in Russia may soon face a reduction in farming opportunities due to various kinds of environmental pollution. Many experts tend to explain the unprecedented increase of barren land in Russia (by four million ha during the past two decades) by the intensive exploitation of mineral resources and industrial construction [39
]. Kashtanov [131
] and Dobrovolski [132
] associated the expansion of industrial infrastructure with long-term and irreversible losses of cropland in Russia. In support of the earlier findings of Sorokin et al. [130
] and Solgerel et al. [133
] concerning the close relationships between industrial development and arable acreage, strong correlations between the proportions of croplands, perennial plantings, and industrial lands are revealed in both urbanized type I and II territories and sparsely populated yellow belt areas.
Distinct from urbanization, industrialization may affect agricultural land use in remote areas. According to Sorokin et al. [130
], most of the abandoned lands are located in the north of Russia. This agrees well with our finding of strong correlations between the variations in the acreage of croplands, disturbed lands, and barren lands in the north locus of the yellow belt. Prishchepov et al. [39
] and MacDonald et al. [134
] also reported abandoned agricultural land concentrated in remote and isolated industrialized areas in northern Russia. Nakvasina et al. [135
] claimed that the proximity to urban areas might be used as a critical criterion to transfer disturbed and barren lands back into agricultural use. However, we did not identify strong correlations between the variations in agricultural land fund compositions and residential lands for type III territories.
In diverse land activity patterns across the Russian territories, changes in the compositions of agricultural and non-agricultural land funds depend on the degree of industrial development. As mentioned by Postek et al. [136
] and Prishchepov et al. [39
], agricultural land loss due to increasing industrialization causes the fragmentation of arable lands as smaller locations with lower productivity. However, according to Popov [137
], fragmentation is not a problem in agriculture-oriented areas due to the excessive lease of agricultural land. The issue is particularly topical in territories where arable land is scarce, however [138
]. Nefedova [140
] reported that in northern and eastern parts of Russia, agricultural land distribution is extremely fragmented. Our results demonstrated that in the Russian North and Far East, low activity of cropland is coupled with the prevalence of hayfields in the composition of the agricultural land funds there. High intragroup correlations between the proportions of cropland, rangeland, hayfields, and perennial plantings in type IV territories confirm the observations of King and Burton [141
], Tan et al. [142
], and Dhakal and Khanal [143
], i.e., the fragmentation results in the competition between the categories of agricultural lands.
We performed our analysis in the short-term, but it is commonly known that land transformations (particularly, for croplands and annual crops) can be rapid, whereas transformations are slower in grassland-livestock oriented areas and permanent crop areas. Nationally, the ongoing loss of croplands may not have an immediate effect on the agricultural output of Russia. Still, this represents enormous environmental, economic, and social costs that will be hard to absorb in terms of a long-term perspective [144
]. Griewald et al. [34
] and Hunt et al. [145
] outlined five principal drivers of long-term change in agricultural land use in Russia, environmental drivers being one of them. Our findings would allow one to expect that the evolution of land-use change will be affected by the pressure exerted on ecosystems by various land management types [34
]. While some authors, including Diputra and Baek [146
] and Mahcene et al. [147
], reported little evidence that industrialization causes a significant increase in disturbed land acreage, our results suggested that lower activity of agricultural land categories is correlated with a higher activity of barren land, disturbed land, and industrial land. Weaker, but still significant, correlations between the proportions of agricultural and industrial land categories are revealed in type I and II territories here. In type IV territories, the contributions of croplands and perennial plantings to regional land funds are also linked with variations in the acreage of barren lands.
Among the drivers of land-use change, in the long run, there are also economic, social, technological, and policy-related factors. Bukvareva et al. [148
] stated that current land-use policies in Russia pay little attention to the environmental costs associated with the re-use of abandoned lands. In light of the economic recession that Russia has been experiencing since the mid-2010s, farmers tend to reinforce the exploitation of all available lands to ensure sufficient income inflow. Often, this is done regardless of whether some lands are of high environmental value or are socioeconomically marginal [29
]. In the short-term perspective, we did not reveal an increase in the acreage of croplands due to the use of other categories of agricultural land. To some extent, however, the correlations between the proportions of agricultural land categories are identified in type III territories. In these yellow belt areas, land reclamation programs will require substantial investments for clearing forested land, liming, and other works. In the short-term, high reclamation costs along with poor soil quality may reduce expected economic returns [149
]; however, in the long run, the incentives for reclamation may grow as both the availability and quality of croplands in type I and II territories degrade. Such a perspective highlights the need for a deeper investigation of the variations in land fund compositions within a sustainable agricultural land management approach as a component of the broader economic and environmental system [150
In recent decades, there has been increasing concern for ensuring the effective utilization of agricultural land due to the limited area of highly productive arable land and the growing demand for food and farming products internationally. In Russia, an orientation of state policy towards the growth of agricultural production, along with a low level of environmental awareness among farmers, has impeded the prospects of sustainable land management as an integral aspect of land use planning. The degradation of agricultural lands due to irrational use has posed environmental, economic, and social threats to the national development objectives of land management in many territories of the country. As most studies in Russia have focused on land changes between the categories of agricultural land, the influence of agriculture-to-urban and agriculture-to-industry transfers has been downplayed.
We conducted this work, intending to study such variations by revealing the interdependencies between the proportions of agricultural land categories, on the one hand, and urban, industrial, and other types of land on the other. First, land distribution was mapped based on a share of agricultural lands in a composition of a land fund and, second, by a “land activity rating” of Russia’s territories. Such a two-step approach to mapping allowed us to find that the proportions of agricultural lands in the composition of a land fund do not appropriately reveal the variations in the activities of agricultural land categories. In the territories, where agricultural lands dominated in the structure of a land fund, the agricultural land activity could be depressed by high proportions of non-agricultural lands. In urbanized and densely populated territories, the composition of the agricultural land fund was predominantly affected by the changes in the acreage of residential and industrial lands, as well as the lands for transportation and communication. In industrialized but underpopulated territories, the acreages of croplands and perennial plantings were strongly correlated with those of disturbed and barren lands. We also found that lower land activity tended to increase the variations within the agricultural land fund, which might indicate intercategory competition for more fertile, more productive, and better-located agricultural lands.
By establishing and testing the five-stage algorithm, we attempted to solve the scientific problem of low awareness in the causality between land-use processes and the composition of the land funds at regional scales. As distinguished from previous studies in the area, we investigated variations in the compositions of a land fund as interactions between the proportions of agricultural and non-agricultural lands. Practically, in territory-scale studies, such an approach might complement regionally adapted monitoring networks by targeting the mismatches between the cadaster-based mappings of agricultural land distributions and ranking-based activities of agricultural lands. Theoretically, such an algorithm allows one to capture the complex relationships of a variety of land categories and the resulting correlations between their proportions, therefore, being applicable for studying territorial land-use patterns, the simulation of agricultural land distribution systems, and the extrapolation of current trends into the future. Potentially, the algorithm is suitable for numerous locations. However, one of the limitations of the current study was that it used the Russian system of land statistics, which is built on thirteen land categories. Due to the different sources of land use data in different countries, an adjustment of the array of land categories to a national land reporting system is needed when implementing the method in a broader international context. Another limitation that could potentially challenge cross-country comparisons is the different sizes of territorial units. Russia’s case demonstrates that this problem may arise even within one country, where territories substantially differ in size. In an attempt to overcome a data discrepancy obstacle, we conversed cadastral classification data into land-rating values. To address the diversity of territories, we used an agricultural land activity parameter. This allowed us to adjust agricultural and non-agricultural-oriented ranking systems to make them comparable. Nevertheless, further research is needed to assess to what extent the approach would be able to appropriately picture variations in agricultural land activity patterns in the conditions of information asymmetries among countries.