Keywords

1 Introduction

Land use is a human activity serving manifold needs. Among others, it delivers food, biomass or agrofuels from agricultural production, or creates space for housing and business. Land use is constantly changing over time due to human land use decisions. For agriculture, it is determined by socioeconomic (e.g., market prices, regulations), sociocultural (e.g., public attitudes), biophysical (e.g., climate), ecological (e.g., pollinator availability), and landscape factors (e.g., landscape connectivity) (Kristensen et al. 2016; Sluis et al. 2019; van Vliet et al. 2015). Farmers respond differently to these factors and adjust their land use decisions (e.g., changing crop rotation, abandoning alpine grasslands). Land use and its changes can be analyzed at different spatial scales (e.g., field, farm, landscape). The sum of field scale land use changes in a given area results in changes at the landscapeFootnote 1 scale (e.g., landscape composition and connectivity, water availability, grassland species abundance and diversity).

Land use change is related to two major contemporary challenges: the climate and biodiversity crises (IPBES 2018; IPCC 2019), which pose considerable risks to future agricultural production and ecosystems functioning at field to global scales. Due to its high relevance, land use change has been extensively researched by different scientific disciplines in past decades using the “land use system” as an analytical framework. It is composed of factors causing land use change and of feedbacks from land use change on the factors – all in all, making it highly complex, stochastic, and dynamic. Thus, scholars have called for an analysis of the land use system at the landscape scale to cover important socioeconomic, sociocultural, biophysical, and ecological processes (Cordingley et al. 2016; Sayer et al. 2013). The analytical scale typically determines the level of detail of the land use system considered and the spatial resolution of input data. Smaller scales (i.e., farm, field, landscape) allow for consideration of detailed land use system factors and high resolution input data, while they are limited in spatial coverage compared to larger scale studies.

The definition of the land use system, its representation with scientific methods (e.g., computerized land use models), and data on the factors composing the land use system are the three components that determine integrated landscape assessments (ILAs). ILAs build on data, theories, and methods from different disciplines and consider relevant factors with high spatial and temporal resolution. They can be used to reveal the effects of land use change on land use subsystems, i.e., ecology (e.g., species composition) or natural resources (e.g., topsoil organic carbon stock), thus complementing landscape monitoring. ILAs may also support analyses of socioeconomic effects (e.g. regional employment) and allow to reveal the effects of past land use change (i.e., ex-post studies). Their strength, however, is to assess alternative land use options as well as possible land use decisions and their effects under current or future conditions of land use system factors (ex-ante or future studies). Thus, ILAs can provide valuable information for policy design, land use decision-making, and technological development in the context of the climate and biodiversity crises. As alpine regions are considered very important in this respect, there are many examples of ILAs for case study landscapes in Austria, each focusing on particular challenges related to land use change and both crises (Karner et al. 2019, 2021; Schönhart et al. 2014, 2016, 2018). However, these studies neither present a definition of the land use system nor the full set of data available for ILAs in Austria.

Based on a literature search on i) frameworks and theories aiming to describe land use systems (and more general socioecological systems) as well as land use decision-making, and ii) empirical studies describing factors influencing land use at the landscape scale, we develop a conceptualized land use system at the landscape scale. It includes relevant farm internal factors influencing land use decisions as well as factors representing the socioeconomic, sociocultural, and ecological environment of farms. Feedbacks between factors are also considered. Our land use system at the landscape scale shall allow us to identify data needs, relate data sources to land use system factors and identify data gaps for Austria as a case study. For this reason, we reviewed empirical studies describing factors influencing land use at the landscape scale and data needs of published ILAs in Austria. We also researched available data sources relevant for ILAs in Austria. As an overarching objective, this article shall support ILAs in Austria.

In the following section, we present a conceptualized land use system at the landscape scale and the interrelations and feedbacks among factors. We then describe the three main groups of factors in detail, i.e., farm scale, socioeconomic and sociocultural, and environmental factors. We specify the required data for a typical ILA for Austria. On the one hand, data are required to parameterize the ILA methods and to validate the modelled land use changes. On the other hand, data are needed to reveal the effects of land use decisions and hence to be able to consider feedback loops. We present a table for each factor group. The tables provide an overview of the data sets related to the conceptualized land use system, refer to data sources available for Austria, and specify the typical temporal and spatial resolution relevant for ILAs.

2 The Conceptualized Land Use System at the Landscape Scale

We present a conceptual land use system at the landscape scale centered around land use decisions at farm scale (Fig. 1). It is considered as a general system description, but its institutional components (e.g., the role of agricultural policy) may be more typical for Europe than for other world regions. The three groups of main factors, i.e., farm scale, socioeconomic and sociocultural, as well as environmental factors, are presented in the subsequent sections.

Fig. 1
figure 1

Conceptualized land use system at the landscape scale. Own illustration

Full size image

Farmers’ land use decisions are taken from a set of options available to a farm. Decisions on the one hand cover management practices such as crop rotation, fertilization, irrigation and pesticide application rates, tillage and soil management, and maintenance of landscape elements. Decisions may also include land cover changes, i.e., conversion of grassland to cropland or the abandonment of alpine grassland. Decisions on livestock management (e.g., herd size, feed, manure management) are included in land use decisions for area-based livestock production systems. Farmers’ investment decisions can also be part of a bundle of land use decisions.

Land use decisions are taken at different temporal scales. For instance, farmers’ investment decisions are typically long-term and can create sunk costs and path dependencies. It should be considered that they affect farm scale factors in the short term (e.g., crop rotation) (Berger and Troost 2014).

Land use decisions have effects on the environment and society. Some of these effects become feedbacks to other factors, as indicated by the arrows in Fig. 1. Farm scale factors can be influenced by land use changes and their effects on an individual farm (Meyfroidt 2013). Socioeconomic and sociocultural factors as well as environmental factors react to land use changes at landscape scale. Frequently, changes in the provision of ecosystem services create such feedbacks (e.g., Lambin et al. 2000). For instance, the removal of landscape elements (i.e., land use change) leads to a feedback on ecological factors (e.g., biodiversity) as well as on cultural factors (e.g., if the quality of life of the people who live in or visit this landscape declines, this can put pressure on policy instruments). Feedbacks are also likely for the climate. For example, higher fertilizer use causes more greenhouse gas emissions and hence contributes to climate change. Feedbacks on socioeconomic factors include a reduction in labor demand due to widespread investments in new technologies.

The land use system shown in Fig. 1 also captures feedback loops via cascades of different factors. For instance, increasing fertilizer use (i.e., a land use decision) has negative effects on biophysical factors such as water quality. These effects might cause the adoption of new regulations (e.g., the Nitrate Action Program), restricting future behavioral options, and ultimately alleviating negative environmental effects. For simplicity, the system boundaries of the conceptual land use system are limited to the landscape scale. Effects outside a given landscape (e.g., water quality at large watershed level) are not considered.

3 Factor Groups and Data Sources

3.1 Farm Scale Factors

Farm scale factors directly influence land use decisions. As part of it, farmers’ behavioral factors include cognitive factors, internal social factors, and dispositional factors (Dessart et al. 2019). For the data sources available for Austria, see Table 1.

Table 1 Farm scale factors and data in Austria. Own illustration
Full size table

Cognitive Factors

are proximal to specific land use decisions and may thus vary for different land use decisions such as climate change adaptation or biodiversity management. They are related to learning and reasoning (Dessart et al. 2019). For climate change and biodiversity assessments, they include the appraisal of the respective risk (i.e., perceived probability and severity of climate change or biodiversity loss), and the appraisal of land use practices to combat these risks (i.e., perceived efficacy, self-efficacy, and costs; Grothmann and Patt 2005; Mitter et al. 2019). The appraisal of land use practices can limit the behavioral space of a decision. For instance, an environmentally friendly farmer might not consider using certain pesticides or fertilization rates.

Internal Social Factors

describe aspects related to farmers’ interactions with other participants in the land use system (e.g., other farmers or farm advisers). They include injunctive norms (i.e., social approval), descriptive norms (i.e., social comparison, conformism, or conditional cooperation), and signaling motives (i.e., the need for social status) (Dessart et al. 2019).

Dispositional Factors

are relatively stable, internal factors of farmers such as their farming objectives, risk tolerance and resistance to change as well as decision-making rules (Dessart et al. 2019; Schlüter et al. 2017). For the latter, Schlüter et al. (2017) distinguish between random, optimization, satisficing, imitation, social comparison, and habitual rules. Müller-Hansen et al. (2017) differ between forward-, backward- and sideward-looking decision rules. Kropf and Mitter (see chap. Factors Influencing Farmers’ Climate Change Mitigation and Adaptation Behavior: A Systematic Literature Review) behavioral factors related to climate change adaptation and mitigation and provide some further details.

Land use decisions are also influenced by other factors at farm scale:

Farmers’ Sociodemographic Characteristics

provide details about farmers (e.g., age, education, gender, knowledge and skills, household size, off-farm employment).

Farm Characteristics

include farm size, farm type, cultivation system, main production activities, farm succession, on-farm technologies, employees, labor and land productivity, owned and rented land, and liquidity.

3.2 Socioeconomic and Sociocultural Factors

In contrast to farm scale factors, socioeconomic and sociocultural factors are independent from an individual farm and describe the social environment of agriculture at larger scales, i.e., from the landscape to the global scale, that influences farmers’ land use decisions. These include economic, political and institutional, technological, external social and cultural factors as well as infrastructure. For data sources available for Austria, see Table 2.

Table 2 Socioeconomic and sociocultural factors and data in Austria. Own illustration
Full size table

Economic Factors

are regional to global aspects that can be considered part of the economic system and affect the economic performance of farms. These include, for instance, off-farm employment opportunities, input and output prices, investment costs, regional and national supply and demand, the land market, international trade of agricultural inputs and commodities, and marketing channels (Malek et al. 2019; Mitter et al. 2020; Plieninger et al. 2016; van Vliet et al. 2015).

Political and Institutional Factors

describe determinants that formally regulate land use decisions and are defined by administrations, parliaments or other institutions. They can be defined at different scales, for example, by municipalities (e.g., zonal plans), federal states (e.g., standards, laws), nations (e.g. regulations, laws), supra-national entities (e.g., regulations, standards) or other countries (e.g., regulations, agreements). They include, for instance, environmental/food/animal welfare/social standards, public payments (e.g., for the environmentally friendly use of land in marginal areas), zonal plans, trade agreements, taxes, and tariffs (Malek et al. 2019; Mitter et al. 2020; Plieninger et al. 2016; van Vliet et al. 2015).

Technological Factors

refer to elements that determine technological endowment options of farms and include agrotechnological change as well as the availability and accessibility of specific technologies (Malek et al. 2019; Mitter et al. 2020; Plieninger et al. 2016; van Vliet et al. 2015).

External Social Factors

describe the social environment of farming and include, for instance, the farming network, demographic characteristics at larger scales (e.g., regional to national population, age, education), and the pace of urbanization (Groeneveld et al. 2017).

Cultural Factors

are informal institutional aspects that include, e.g., public attitudes, values, beliefs, and the traditional gender division of labor (Plieninger et al. 2016).

Infrastructure

in context of land use describes basic physical and organizational structures and facilities that enable farming in a particular region. Infrastructure comprises, for instance, financial (e.g., banks), technical (e.g., energy and water supply, wastewater treatment), digital (e.g., high-speed internet), transport (e.g., public transport, railways, road network), and social infrastructure (e.g., childcare, schools, information and extension services) (Malek et al. 2019).

3.3 Environmental Factors

Environmental factors refer to the natural environment and include biophysical factors and climate change, ecological factors and biodiversity, and landscape factors. These determine the quality of farm level characteristics such as the fields of a farm and its interaction with the natural environment. For Austrian data sources, see Table 3.

Table 3 Environmental factors and data in Austria. Own illustration
Full size table

Biophysical Factors and Climate Change

refer to determinants that are part of the biophysical environment, e.g., climate, soil and water conditions. These include, for instance, rainfall, temperature, soil structure, topography as well as water quantity and quality. Some of these factors are variable over time and may be influenced by land use resulting in feedback loops (e.g., climate conditions), while others are rather stable (e.g., topography).

Ecological Factors and Biodiversity

describe the living environment such as flora and fauna below and above ground. A number of ecosystem services related to agriculture (e.g., pollination, natural pest control) depend on these factors. Typical ecological factors include species richness (i.e., the number of different species) and species abundance (i.e., the population size of one species).

Landscape Factors

describe the character of a landscape, apart from biophysical and ecological factors. They comprise the elements and structure of a landscape (e.g., landscape patterns).

4 Discussion and Conclusions

The importance of data in land use science was stressed about two decades ago, for instance, by Lambin et al. (2000), National Research Council (1992), and Verburg et al. (2004). As of today, there is a variety of data on different factors influencing land use decisions. Most of the data sets identified for Austria relate to the socioeconomic and biophysical factors, while publicly accessible data sets on behavioral, social, cultural, and ecological factors are scarce and rarely accessible. This coincides with the methodological focus of ILAs that frequently cover socioeconomic and biophysical factors and give limited consideration to behavioral and ecological factors. Some data could be obtained from previous studies, as shown in Table 1. However, most behavioral, social, cultural, and ecological data i) show the existing range of, e.g., values or species, but do not provide quantitative information on their distribution in (entire) Austria, which would be required for ILAs, ii) are not centrally stored and accessible (often due to data protection agreements), but are dispersed in literature, and iii) are often tailored to specific aspects of decision-making, e.g., climate change adaptation (Mitter et al. 2019), biodiversity-friendly land use (Schmitzberger et al. 2005), organic farming (Darnhofer et al. 2005), soil management (Braito et al. 2020), or farm succession (Engelhart et al. 2018).

However, filling all data gaps for ILAs involves high costs. Novel methods such as remote sensing may close some of the data gaps in the long run. Additional cost-efficient methods include, for instance, expert evaluations of missing data, citizen science approaches and scenario analyses. An expert evaluation was conducted, e.g., by Hainz-Renetzeder et al. (2015) in the Seewinkel region. In several workshops, they asked a heterogenous group of experts to semi-quantitatively evaluate the actual and potential provision and the status of different environmental goods in order to compensate for lacking soil, water, and biodiversity-related data. This allows, for instance, to evaluate the effect of land use change on environmental factors. Citizen science approaches can tackle two problems, i.e., the provision of data and stakeholder participation. Scenarios allow to partially compensate for the lack of specific data, to analyze a range of future developments (e.g., the Shared Socioeconomic Pathways for agriculture and food systems; Mitter et al. 2020), and to account for data uncertainty and validity (e.g., climate scenarios; Kirchner et al. 2021). In both, expert evaluations and scenario analyses, participatory methods and stakeholder engagement can contribute to successful implementation (see e.g., Mitter et al. 2014; Reed 2008).

The data needs of an ILA are mainly determined by the scope of the study, i.e., the chosen focus on the land use system, and the research context. For instance, if an assessment aims to reveal the impacts of climate change scenarios on land use and environmental effects (e.g., Karner et al. 2021), socioeconomic factors become secondary. Furthermore, certain factors might not be relevant in a particular landscape. However, neglecting important factors of the land use system at the landscape scale when assessing land use change effects might lead to biased recommendations. Thus, close cooperation with regional stakeholders is essential to represent key factors in the ILA, as also stressed, e.g., by Kaim et al. (2020).

One challenge for researchers is to evaluate the quality of the data and hence their inherent uncertainty. Central data repositories such as those operated by the Climate Change Centre Austria (CCCA) and the Austrian Social Science Data Archive (AUSSDA) have the advantage of providing centralized access and minimal quality control of the data (e.g., a metadata check). Other data that are plausibility-checked and partly quality-assured (e.g., some Statistics Austria data sets) have to be purchased by researchers.

To overcome the problem of data availability, dispersion, and accessibility, we call for i) a wider use of open access data storages with defined quality standards, ii) government support to maintain, merge, and expand existing data repositories such that they include qualitative and quantitative data at different spatial and temporal scales and resolutions, iii) commonly agreed quality standards for data storage to ensure high data quality, iv) government support for data storage, e.g., to meet commonly agreed quality standards, and v) free access to publicly funded data sources (e.g., FADN data) as well as training of researchers and students in data retrieval, processing, manipulation, and interpretation.