A Systematic Review of Developments in Farmland Cover in Chile: Dynamics and Implications for a Sustainable Future in Land Use

Abstract

Farmland covers present diverse characteristics, methods, and techniques to monitor and evaluate crops in other geographic areas. This study systematically reviews Land Use/Land Cover Change (LULCC) in agricultural land in Chile through a systematic review of the scientific literature. Using the PRISMA 2020 method, the Web of Science (WOS) database was consulted using the keywords “Landuse”, “Landcover”, “Agriculture”, and “Chile”. We applied six exclusions criteria and constructed a matrix to select relevant aspects, such as title, year of publication, study area and period, methods used, and principal results. In our review, we identified four studies that focused specifically on agricultural land dynamics, mainly in south-central Chile. Chile was selected as the study area due to its geographical diversity, which poses significant challenges for decision-making in land use regulation. These results underscore the need for more spatially informed data on farmland dynamics to inform decision-making, particularly during the alternatives evaluation stage. In this phase, it is essential to assess the impacts on and potential of the territory in order to define suitable economic activities. Although there are numerous studies on LULCC, most emphasize changes in native forests, underscoring the need to address LULCC more comprehensively by considering other land categories, such as agricultural land, shrublands, grasslands, and others. This evidence is crucial for designing practical land management tools and identifying areas that have been extensively studied but lack sufficient research.

1. Introduction

Land use and land cover have undergone significant global changes, driven by increasing demands for food and resources to meet population needs [1]. Farmland cover within the framework of LULCC reflects a complex historical–territorial transition involving varying degrees of alterations affecting natural vegetation, soil, and animal habitats [2]. These changes challenge understanding common patterns across scales [3].

In Europe, Ref. [4] states that agricultural land cover change is explained by differences in land rent between settled and farming areas, transformations into forests in economically struggling rural regions, and increased productivity and changes in agricultural competitiveness. This situation has generated the fact that in the 2015–2030 period, more than 20 million hectares of agricultural land are at high risk of potential abandonment. In the United States, the United States Geological Survey (USGS) Trends assessment shows complex LULLC patterns, with high rates of change in forested regions with active forestry and dynamic agricultural lands that intensify where there is competition for other land uses [5], where according to [6], between 1973 and 2000, agricultural land decreased by approximately 90,000 km². In Southeast Asia, authors like [7] have identified agricultural expansion and forest retreat in protected areas such as wildlife sanctuaries, nature reserves, and protected forests, where about 10% of agricultural expansion and forest retreat occurred in protected areas in the last 32 years.

In Chile, it is also estimated that protected areas will be intervened by land use dynamics, since according to [8] by the year 2100 it is estimated that agricultural land will occupy approximately 8% of the areas surrounding conservation sites, reshaping the landscape to create optimal areas for economic activities [9]. On the other hand, Chile’s free market oriented economic model means that agricultural activity is not subsidized to guarantee production and soil management [10], as is the case in Europe with the Common Agricultural Policy, which, through the subvention of public funds, supports the promotion of sustainable agricultural practices [11].

Additionally, significant demographic changes, including the industrialization of productive activities, have caused a marked decline in the rural population. This demographic shift has led to the progressive abandonment of agricultural mosaics and the expansion of secondary forests [12], generating negative impacts such as the disappearance of traditional agricultural practices, losses of natural habitats, and invasion of exotic species [13]. Farm use is widespread in Chile’s Central Valley, a geomorphological unit historically occupied by human settlements. Despite geographical and climatic challenges, such as water scarcity and its detrimental effects on food production [14], the Central Valley provides optimal conditions for the cultivation of Mediterranean food plants and cereal production. Most cattle and sheep farming occur further south [15]. Agricultural practices can be found in all geomorphological units, with 54% concentrated in the Araucanía, Biobío, and Maule regions [16]. Another 8.4% is in the extreme north and the Coquimbo region, while no more than 1.8% occurs in the Aysén and Magallanes regions [17].

Land use has significantly modified soil structure [18], as prolonged land use leads to lower organic matter levels and greater soil degradation [19]. In Chile, the most significant loss of native forest was due to burning trees for grazing and crop growth [9]. Industrialization and demographic changes have significantly reduced the rural population, leading to the abandonment of farms and the expansion of secondary forests [12]. Historical maps of LULCC have been designed to explain current landscapes based on historical processes [20]. This has identified trends such as expanding agricultural land in ecologically valuable areas [21] and abandoning agricultural land in southern Chile [13].

In Chile, the vegetational cadaster is currently a primary cartographic source for LULCC analysis. However, its usefulness in identifying farming activity is limited due to its focus on dense vegetation cover, which results in less precise agricultural land classification than forest cover [22]. Additionally, variations in the surface area of certain categories across different cadastercadaster years result from technological advancements in the acquisition and processing of digital cartographic materials, such as aerial photographs and satellite images [23]. Some institutions in Chile, such as the Natural Resources Information Centre (CIREN) and the National Statistics Institute (INE), address this challenge by periodically creating registers specific to farmland, such as the Fruit Cadaster [24] and the Farming and Forestry Census [25]. These initiatives aim to enhance the accuracy and coverage of agricultural land data.

One of the challenges in agriculture worldwide is to understand the state of farm systems, given the significant carbon loss due to practices like slope cultivation, excessive pesticide use, and soil erosion [26]. In this context, authors such as [27] argue that coordination between spatial territorial planning and the Sustainable Development Goals (SDGs) is necessary. Such coordination can facilitate the development of indicator systems tailored to the specific conditions of a given territory. Similarly, Ref. [28] highlights that the most critical SDGs for agricultural development are those related to production, food security, and environmental quality.

In Chile, Ref. [29] state that soil degradation due to LULLC has adverse effects on agricultural productivity, rural livelihoods, biodiversity, and food security. Soil biota is particularly impacted by degradation processes, including soil erosion and physical and chemical deterioration, all of which exert negative pressures that compromise the diversity and functionality of soil organisms [30]. This problem is particularly serious in Chile, as rural land remains unregulated at the legislative level, allowing for overexploitation due to unrestricted property rights [31]. However, the recently enacted National Rural Development Policy (PNDR) and National Land Management Policy (PNOT) are expected to introduce regulatory oversight in rural development [32]. Currently, both policies are in the implementation phase, with coordinated efforts among different levels of government and civil society actors to achieve their objectives. Nevertheless, these policies have certain gaps in addressing land overexploitation, as they establish general guidelines but lack enforcement mechanisms and clear sanctions for those who overexploit the land [33]. Finally, Chile lacks a dedicated public agency for land monitoring and evaluation, resulting in oversight being limited to regions of high agricultural and forestry interest [34], which hinders effective control of soil overexploitation.

Identifying farmland LULCC is critical for designing territorial plans, as agriculture shapes the rural landscape [13] and affects the provision of ecosystem services [35]. The authors hypothesize that most of the research on land cover change in Chile focuses on land use dynamics related to large areas of dense vegetation, due to the extensive area changes that large forest covers have undergone in south-central Chile since the 1970s. This hypothesis is supported by a general database search, where a large portion of LULCC-related articles in Chile include the term “Forest” in their titles. The objective of this paper is to provide a comprehensive descriptive review of agricultural land use and land cover change (LULCC) in Chile, examining the spatial dynamics and large-scale patterns of these changes. Additionally, it aims to identify the types of studies and methods employed in LULCC analysis. The results of this review will highlight existing evidence from scientific research on farmland monitoring, offering valuable insights to address Chile’s challenges in integrating LULCC data into decision-making processes.

2. Materials and Methods

The protocol used for the present systematic review was registered at OSF (Open Science Framework) and can be found here: https://doi.org/10.17605/OSF.IO/26NV8 (accessed on 27 January 2025). This review curated scientific articles on Land Use and Land Cover Change (LULCC) involving farmland use/cover. The Web of Science (WOS) database was queried using the English keywords: “Landuse”, “Landcover”, “Agriculture”, and “Chile”. We decided to use keywords only in English because, after reviewing articles published in WOS, we observed that all results are in English or at least include an abstract in English. Therefore, even if an article is published in Spanish, it would still appear in the search results due to the presence of an English abstract. We chose to use WOS as the search engine, as this database is widely recognized in the scientific community for its comprehensive coverage and its role as a key source for evaluating scientific production. We believe that including other databases, such as Scopus or Google Scholar, could lead to duplicate results and increase bias in the inclusion of articles. The search for articles was carried out between March 2022 and November 2023. English keywords were chosen as WOS abstracts, and titles are in English, even for Spanish manuscripts. We focus on Chilean regions, as our interest lies in analyzing information monitoring at the national level. This will allow us to discuss the challenges of decision-making, recognizing that each country faces unique realities and challenges. We examined all types of articles addressing LULCC in Chile that included farmland terms like “crop”, “meadows”, or “grasslands” (due to minimal distinction between agricultural land and grassland). The search was extended to 2022 using the PRISMA method [36], following the guidelines outlined in the checklist available online at: https://www.prisma-statement.org/prisma-2020-checklist (Supplementary Materials). The characteristics of the studies that were searched are as follows:

1.

Only documents of type “Article” were considered, and “Book Chapters” were considered. Documents such as “Proceeding Paper”, “Review Article”, “Editorial Material”, among others, were not considered. This choice was made because the aim was to identify original and applied works with specific results.

2.

The search for scientific articles was conducted up to the year 2022 to avoid the inclusion of new articles in the process of writing this review. This considers that the article review period was between March 2022 to November 2023.

3.

The search considered all Web of Science categories, since LULC papers are crosscutting to all areas of knowledge.

We identified 1305 WOS articles ordered by relevance, where we assume that the most cited and essential articles are found in the first places of the search, a first review of the title, the abstract, and then the results of the manuscript.

With the results obtained, six exclusion criteria were applied (

Table 1. Exclusion criteria for manuscript selection.

Number of Reasons	Reason for Exclusion	Register
1	Human selection. We examined the initial 718 out of 1305 WOS search results, as subsequent articles were repetitive, covered diverse subjects, and had lower views and citations.	587
2	Manuscripts that do not present an analysis of land cover change in Chile as a study area.	331
3	The study area does not correspond to an area of Chile	155
4	Articles whose results only addressed one LULCC data point.	162
5	The first author presents the same results in another manuscript.	5
6	The results are not specialized. That is, articles analyzing land cover change figures but not using maps.	11

and Figure 1), leading to the exclusion of 587 articles for being repetitive or addressing other topics in which LULC is mentioned without any in-depth analysis. For instance, in the study by [37], the authors carried out a productive reconversion of a watershed but only briefly mentioned LULC in certain sections without analyzing these changes in detail. Similarly, Ref. [38] evaluated landslide susceptibility in a watershed in Chile but did not incorporate LULC into their assessment. In total, 331 articles were excluded because they did not consider LULC analysis (e.g., [39], where the authors analyze a protected site in Chile, without considering the LULC); 155 articles were excluded because they did not consider a territory of Chile (e.g., [40], where the authors analyze the Natura 2000 network (N2000) in mainland Spain and the Balearic Islands, a study area far from Chile); 162 articles were excluded because they covered only one year of study (e.g., [23] where the authors analyze the capacity of the Chilean national forest monitoring system to perform a temporal follow-up of the forest area, considering for this purpose the methodology of the Chilean forest cadastercadasters; however, the authors do not make comparisons of the cadastercadasters between years.); 5 articles were excluded because the authors presented the same results in other manuscripts (e.g., Ref. [41] used the same study area and results as [42], incorporating classification trees and logistic regression models in their more recent work to analyze vegetation change trajectories); and finally, 11 articles were excluded because the LULC results were not spatialized (e.g., Ref. [43] spatially analyzed the degree, extent, and distribution of anthropogenic disturbances in the Magallanes Region of southern Patagonia, Chile. However, their analysis did not account for land use and land cover change (LULCC). Considering all these exclusions, we reviewed a total of 54 manuscripts. We constructed a matrix to extract key aspects from each article, including the title, year of publication, study area and period, methods used, and main results (

Table 2. Matrix to select relevant aspects of each article.

Importance Percentage	Aspect	Description
10%	Titte	We analyze the title of the article to identify the main theme
30%	Year of publication	We identified the year of publication of the article to be until 2022
30%	Study area and period	We identified that the study area be in a region of Chile, and that the period of analysis be at least two different dates
10%	Methods used	We review the methods used for LULCC analysis
20%	Principal results	We identified the main results that were achieved in each project

). In this process, we prioritized the study area (30% weight) and the study period and year of publication (30% weight), as these criteria are the most relevant for identifying the periodicity and geographical scope of the manuscripts. The main results were given a weight of 20%, considering that this section contains the spatial analysis outcomes. Finally, the methods used (10% weight) and the article title (10% weight) were assigned lower priority, as these elements can vary significantly between manuscripts and may or may not explicitly include terms such as “LULCC” or “Land use”. We developed a standardized data extraction form to structure the review results. The review authors applied the form independently, considering the six exclusion criteria and reviewing the elements indicated above. Any discrepancies in judgments of risk of bias or justifications for decisions were resolved by discussion to reach a consensus between the two review authors. The main discrepancies arose in interpreting the exclusion criteria, considering papers that, for example, mentioned LULCC data but were never used in the results, or some papers that were not applied articles but corresponded to literature reviews.

To improve the robustness of the results, the 54 articles were synthesized in a table and classified by methodology, analysis periods, geographic area of analysis, and topic of study (

Table 3. General studies of farmland cover in LULCC in Chile.

Methodology	Period	Geographical Area	Topic	Source
Supervised classification	1975–1980–1985–1990–1995–2000–2005–2010	Maule, Ñuble and Biobío region	Forest dynamics	[44]
Vegetational cadaster	1997–2014	La Araucanía region	Agricultural dynamics	[45]
Satellite products	2015–2019	Mountain’s areas, from the Metropolitan region to the Los Lagos region	Forest dynamics	[46]
Vegetational cadaster	1997–2009–2016	The Longaví watershed, Maule region	Watershed dynamics	[47]
Vegetational cadaster	2008–2015	The Andalién watershed, Bíobio region.	Watershed dynamics	[48]
Supervised classification	1990–2003–2017	Quillota, Valparaíso region	Urban dynamics	[49]
Supervised classification	1976–2001–2016	Arauco, Bíobio region	Forest dynamics	[50]
Supervised classification	1984–2000–2018	Ñuble region–Coyhaique, Aysen region	Integrated Land Use and Land Cover Dynamics	[51]
Supervised classification and Vegetational cadaster	2001–2002, 2014–2015, 2016–2017	Three watersheds on the Quepe River, La Araucanía region.	Watershed dynamics	[52]
Supervised classification and Vegetational cadaster	1975–1989–1999–2009–2016	Vichuquén watershed, Maule Region	Watershed dynamics	[53]
Satellite products and vegetational cadaster	1992–2011	Continental Chile	Integrated Land Use and Land Cover Dynamics	[8]
Supervised classification	1987–2001–2015	Budi and Lingue watersheds, La Araucanía and Los Ríos Region	Forest dynamics	[54]
Aerial photographs and satellite imagery	1955–1975–2014–2017	Constitución, Maule Region	Forest dynamics	[55]
Supervised classification	2001–2016	Boca Maule watershed, Bíobio region	Watershed dynamics	[56]
Supervised classification	1986–2011	A watershed in Biobío Region	Watershed dynamics	[57]
Supervised classification	2002–2017	Two watersheds on Valparaíso Region	Watershed dynamics	[58]
Supervised classification	1975–2016	Two watersheds on Valparaíso Region	Watershed dynamics	[59]
Supervised classification and aerial photographs	1960–2014	Coastal Range of the Maule, Ñuble and Biobío Region	Forest dynamics	[60]
Vegetational cadaster	1994–2007–2014	Ñuble-Biobío Regions	Integrated Land Use and Land Cover Dynamics	[61]
Vegetational cadaster	2000–2016	O’Higgins, Maule, Valparaíso and Metropolitan Region	Agricultural dynamics	[62]
Vegetational cadaster	1994–2007	Quepe watershed, La Araucanía Region	Integrated Land Use and Land Cover Dynamics	[63]
Vegetational cadaster and aerial photographs	1998–2008	Biobío Region	Integrated Land Use and Land Cover Dynamics	[64]
Supervised classification	2001–2012	Maipo watershed, Metropolitana Region	Integrated Land Use and Land Cover Dynamics	[65]
Satellite products	2001–2013	From Coquimbo to Los Rios region	Agricultural dynamics	[66]
Vegetational cadaster	1996–2011	Urban area between the Metropolitan and the Valparaiso region	Integrated Land Use and Land Cover Dynamics	[67]
Supervised classification	1985–1999–2001–2004–2006–2009–2011	South of the Los Rios and North of Los Lagos region	Integrated Land Use and Land Cover Dynamics	[68]
Supervised classification	1986–2001–2011	Rupanco watershed, Los Lagos region	Watershed dynamics	[69]
Supervised classification	1986–2011	25 watersheds in the coastal range of Maule, Biobío and La Araucanía region	Watershed dynamics	[70]
Aerial photographs	1983–2007	Premountainous valley in Pucón, La Araucanía region	Forest dynamics	[71]
Aerial photographs	1994–2007	Quepe watershed, La Araucanía Region	Integrated Land Use and Land Cover Dynamics	[72]
Supervised classification	Spring 2014–Summer 2014–Fall 2014–Winter 2014	Continental Chile	Integrated Land Use and Land Cover Dynamics	[73]
Vegetational cadaster	1993–2008	Malleco-Vergara watershed, Araucanía Region	Forest dynamics	[74]
Supervised classification	1986–2001–2011	From Valparaiso to Los Lagos region	Forest dynamics	[12]
Supervised classification	2000–2008	Vergara watershed, Bíobio region	Watershed dynamics	[75]
Supervised classification	1986–2003	Districts of Valdivia, Corral, Paillaco and La Unión, Los Ríos region	Forest dynamics	[76]
Supervised classification	1975–1992–2001–2011	Catapilco, Valparaíso Region	Integrated Land Use and Land Cover Dynamics	[77]
Supervised classification	1973–1987–1999	La Araucanía Region	Forest dynamics	[78]
Supervised classification	1985–2011	Coastal Range in La Araucanía and Los Ríos Region	Forest dynamics	[79]
Supervised classification	1975–1999–2010	Santiago, Metropolitan Region	Urban dynamics	[80]
Supervised classification	2001–2009	Santiago, Metropolitan Region	Urban dynamics	[81]
Aerial photographs and satellite imagery	1962–2005	Six watersheds in Bíobio and La Araucanía region	Forest dynamics	[82]
Supervised classification	1998–2001–2006	A watershed on Biobio region	Integrated Land Use and Land Cover Dynamics	[83]
Supervised classification	1986–1999–2008	Nahuelbuta Range and municipality of Angol, Araucanía Region	Forest dynamics	[84]
Supervised classification	2000–2010	Metropolitan area of Concepción, Biobío Region	Urban dynamics	[85]
Supervised classification	1989–1999–2009	Metropolitan District of central Chile, Metropolitan Region	Forest dynamics	[86]
Documentary and archaeological sources	1550–2007	From Maule to Los Lagos region	Integrated Land Use and Land Cover Dynamics	[9]
Supervised classification	1985–1999–2007	Temperate landscapes and forests in the Los Lagos Region	Forest dynamics	[87]
Supervised classification	1975–1990–2007	Coastal Range of the Maule and Biobío Region	Forest dynamics	[88]
Aerial photographs	1994–2004	Boroa watershed, La Araucanía Region	Integrated Land Use and Land Cover Dynamics	[89]
Supervised classification	1985–2007	Municipality of Ancud, Los Lagos Region	Agricultural dynamics	[13]
Supervised classification	1975–2008	Mediterranean bioclimate zone, Valparaiso, O’Higgins and Metropolitan Region	Integrated Land Use and Land Cover Dynamics	[42]
Supervised classification	1989–2003	Andean precordillera zone, Maule Region	Forest dynamics	[90]
Supervised classification	1975–1990–2000	Maule River–Cobquecura, Maule and Biobío Region	Forest dynamics	[91]
Aerial photographs	1978–1991–1998	Cities of Los Ángeles and Chillán, Biobío and Ñuble Region	Urban dynamics	[92]

). Each article’s materials and methods section were reviewed to identify the method, geographic area, and analysis periods, specifying how the LULC analysis was performed, the area of study, and the periods to detect changes. As for the topic of the study, it was interpreted by the reviewers at the time of completing the form, identifying the main land cover on which the article analysis was focused.

The results of this review are presented in two points: (1) Period and scale of the scientific manuscripts published on farmland LULCC in Chile, and (2) Methodology of the studies of farmland LULCC. The discussion presents the dynamics of agricultural use/cover in Chile, and the challenges for the Integration of Agricultural Land Geoinformation with Spatial-Temporal Basis in Chile. Finally, the conclusions reflect the results of this review and its implications.

3. Results

3.1. Period and Scale of the Scientific Manuscripts Published on Farmland LULCC in Chile

Table 3 shows the findings from the bibliographic search, encompassing 54 scientific manuscripts on farmland LULCC in Chile. The timeline spans from 1550 to 2022, covering the Coquimbo to Los Lagos regions, excluding national scale works by [8,73] (Figure 2). Notably, LULCC records from 1550 are based on the work of [9], which relied on chroniclers’ accounts and vegetation estimations. Apart from this study, the earliest available records date to 1955, identified by [55] through supervised classification techniques. The Biobío region stands out as a focal point in the published literature (Figure 2), likely because it was the initial site of forest plantation expansion in Chile. Additionally, this region is located near biodiversity hotspots, characterized by a high concentration of endemic species [93] (Figure 2). The analysis reveals that the dynamic forest is a frequently studied LULCC type, with 19 papers focusing on temporal analyses of native forest cover change. Fifteen papers adopt an integrated approach, exploring land use changes for landscape and territorial assessment, incorporating various land cover types, soil properties, land potential, and impacts of productive activities.

Furthermore, eleven papers examine LULCC effects on hydric systems, five delve into urban land use dynamics and spatial models, and four address agricultural dynamics. Articles specifically focusing on agricultural land dynamics include studies by [13,45,62,66]. These studies address fruit crop dynamics, agricultural land abandonment patterns, temperature as a factor, agricultural land as a fire driver, and predictive analyses of crop production. The study by [45] examines the dynamics of agricultural land and fruit crops by projecting future land cover and fruit crop distribution using Markov chains. The findings indicate a significant decline in agricultural land due to the expansion of forest plantations and urban areas, while fruit crop areas have expanded considerably. Additionally, Ref. [13] analyzed rural land abandonment patterns using a spatially explicit statistical model to identify the key drivers of land abandonment. The study concluded that land abandonment results from a complex interaction of geophysical and socioeconomic factors, with significant implications for landscape planning. Furthermore, Ref. [62] investigated temperature and cropland as drivers of wildfires by employing structural equation modeling to evaluate predictors of fire activity, concluding that central Chile will be highly vulnerable to wildfires in the near future due to climate change and human activities. Finally, Ref. [66] performed predictive analyses of crop production, demonstrating that combining publicly available satellite data with optimal linear regression (OLR) and deep learning (DL) models can effectively predict anomalies in vegetation productivity in Chile, where decision makers currently lack an early drought prediction system. Although these articles provide valuable insights into the dynamics of agricultural land in Chile, not all of them are complementary. For example, the studies by [62,66] were conducted at an interregional level, overlapping in five regions of the country. These studies indicate that the areas with the highest agricultural productivity in central Chile are also the most prone to fires caused by agricultural activity. However, this finding should be interpreted with caution, as [62] reports that the accuracy of predictions decreases with increasing time and latitude, highlighting a limitation in the model’s performance over time and space. In contrast, the studies by [13,45] cover different geographical areas and time periods, limiting their comparability. Specifically, Ref. [13] focused on an island in southwestern Chile, while [45] analyzed a region in southern Chile.

A review of all the studies reveals that, geographically, researchers conducted twenty studies in river basins, nine at an interregional scale, eight considering geomorphological boundaries, four at a regional scale, two at a national scale, two at a provincial scale, and two at a local scale.

The studies reviewed are highly heterogeneous and present diverse methods and types of analysis, so we present below the general characteristics we were able to identify using a summary approach.

Table 4. Time Periods and Scales of the Maps from the Analyzed Studies.

Period	Num. of LULCC Maps	Scale of Maps
Pre-1950	1	National (0); Interregional (1); Regional (0); Local (0)
1950–1960	2	National (0); Interregional (0); Regional (1); Local (1)
1961–1970	1	National (0); Interregional (0); Regional (0); Local (1)
1971–1980	13	National (0); Interregional (5); Regional (1); Local (7)
1981–1990	23	National (0); Interregional (7); Regional (2); Local (14)
1991–2000	28	National (1); Interregional (7); Regional (3); Local (17)
2001–2010	46	National (0); Interregional (13); Regional (2); Local (31)
2011–2020	38	National (5); Interregional (9); Regional (2); Local (22)

shows the number of land cover and land use maps produced in Chile within scientific articles. Out of the 54 articles analyzed, we identified 152 LULCC maps. Researchers made 46 maps between 2001 and 2010, covering periods before 1970 and including products at national, interregional, regional, and local scales. Specifically, there were 6 LULCC maps at the national level, 42 at the interregional level, 11 at the regional level, and 93 at the local scale (including communes, localities, geomorphological units, and watersheds).

Authors such as [8,73] conducted national-scale analyses of Land Use and Land Cover Change (LULCC) across continental Chile, using satellite products and vegetation cadasters. Both studies categorize agricultural land and emphasize database accuracy and spectral index dynamics throughout the seasons. Ref. [8] additionally projects future scenarios for LULCC, anticipating a national decrease in farmland by 2030, 2050, and 2080, with significant losses in the central-southern region and gains in the northern part. This review also identified 11 studies on hydrological systems and watershed dynamics among the 93 local-scale LULCC maps.

Ref. [94] highlight watershed boundaries as optimal for natural resource management, planning, and monitoring due to their direct relevance to water resources and support for sustainable development at the local scale.

3.2. Methodology of the Studies of Farmland LULCC

Most studies utilized supervised classification of satellite images to identify changes in land use/cover (e.g., [12,58,59,78,79]). These studies generally treat farmland as a broad land cover category. Other approaches include spatial sensor products for modeling spatial changes (e.g., [57,61,62,66]), vegetation registers (e.g., [9,63,74]), and photointerpretation of satellite images and air photographs (e.g., [60,89,92]). Early change analysis typically began around 1973, when satellite images became accessible, but [60] analyzed the period 1960–2014 using air photographs. Ref. [9] extended their study period from 1550–2007 based on historical accounts and documents, though spatial precision varied. Scale differences among studies range from interregional to regional, communal, and down to hydrographic basin or territorial unit levels.

4. Discussion

4.1. Dynamics of Agricultural Land Use and Land Cover Changes in Chile: Methods and Approaches

The reviewed articles examine agricultural land use and cover changes since 1550, identifying trends in gains and losses. Ref. [9] agricultural land existed before European colonization, expanding further with large-scale plantations that caused extensive forest destruction and fragmentation. In the mid-20th century, import substitution policies increased domestic agricultural production but negatively impacted native forests [12]. By the 1970s, agricultural expansion in southern Chile became a primary driver of native forest loss [78,87,90]. These dynamics responded to population growth and open market policies, incentivizing agriculture and forestry with export-oriented crops, new irrigation technologies, and infrastructure improvements [42].

Ref. [12] studied agricultural land cover changes from the Valparaíso region to the Los Lagos region, totaling approximately 21,339,600 hectares. The research utilized supervised classifications of Landsat images from 1986 to 2011 (

Table 5. Parameters of changes in farmland cover between 1986 and 2011. FL: Farmland loss in the period (%), NAFL: Net annual farmland loss (ha), FLR: farmland loss rate for the period (ha/year), FLR: farmland loss rate for the period (%/year).

Period	FL (%)	NAFL (ha)	FLR (ha/year)	FLR (%/year)
1986–2001	−26.76	−1047	−69,800	−1.58
2001–2011	14.82	681	68,100	1.60

). They calculated Agricultural Cover Loss (ACL) at −26.76% from 1986 to 2001, reversing to 14.82% from 2001 to 2011. The study also assessed Net Annual Loss of Farmland (NALF) in hectares and as a Farmland Loss Rate (FLR), indicating negative values from 1986 to 2001 and positive values from 2001 to 2011.

Other reviews of LULCC in Chile, such as [93], reported changes from the O’Higgins region to the Los Lagos region from 1970 to 2010, focusing on native forest loss rather than farmland cover changes. They documented 92,488 hectares of native forest converted to agricultural land from 1970–1990, 44,288 hectares from 1990–2000, and 39,518 hectares from 2000–2010 [93]. Ref. [9] attributed these trends to early land clearing for agriculture and livestock, while subsequent periods saw reduced forest conversion due to regulatory incentives promoting activities like exotic forest plantations [12,93].

Authors such as [78] note that productive activities historically exploit optimal environmental conditions and accessibility, including plantations of exotic forestry species and farmland. However, in profitable agricultural regions of central-southern Chile, native forest fragmentation has been identified. Conversely, areas with poorly drained soils are unsuitable for agriculture, contributing to rural poverty [87,88]. Human-induced changes also impact LULCC in farmland. Ref. [89] observed that drainage incentives in coastal wetlands expanded agricultural areas in south-central Chile’s watersheds.

In contrast to agricultural expansion, studies like [13] have analyzed agricultural land abandonment in Chile, using spatially explicit statistical models to identify causes and economic impacts, particularly in remote areas [13,87]. Ref. [51] argue that agricultural abandonment stems from biophysical, socioeconomic, and cultural factors, while [13] emphasize property abandonment as a primary driver in Chile. Ref. [85] highlight the challenge of distinguishing between agricultural abandonment and crop rotation practices due to seasonal changes in land cover.

Excluding satellite products, most land cover records in Chile rely on the vegetational cadaster, updated regionally, focusing on forest communities’ location, size, and condition. However, data collection varies regionally, limiting spatial detail. CONAF’s register identifies general land use (e.g., farmland) and sub-uses (e.g., crop rotation/grasslands) without specific crop details. In this context, studies that integrate vegetation cadastercadaster data with satellite imagery (e.g., [52,53,55]) are valuable. However, these studies should explicitly report the Minimum Mapping Unit [95] to prevent issues related to differences in scale and information levels. Among the studies focusing on agricultural land LULCCs, some have been developed using cadastral information (e.g., [45,62]), while others have relied on satellite imagery (e.g., [13,66]).

4.2. Challenges for the Integration of Farmland Geoinformation with a Spatio-Temporal Basis in Chile

The results of this review highlight that most analyses of Land Use and Land Cover Change (LULCC) in Chile focus on the conversion from native forest cover to forest plantations in the central-south zone. Economic incentives and laws like legal decree 701, active in Chile from 1974 to 2012, have predominantly influenced forestry dynamics [96,97]. In contrast, our review found limited specific studies on farmland dynamics in the central-southern region, with notable examples including [13,45,62,66]. There have been other reviews of use change works in Chile, such as [93], where different studies were examined, and an analysis was performed based on latitude of changes, time, and vascular plant richness using generalized linear models. However, Ref. [93] focus on the dynamics of large vegetational densities, such as native forests and forest plantations, focusing the analysis on the Chilean biodiversity hotspot. There are other studies, such as that of [29], in which, although not a literature review, the authors discuss five important soil priorities for Chile, where it is mentioned that maintaining an updated soil registry at the national level is fundamental for research, management and future decision-making, and where the need for a legal framework for land use regulation is also mentioned. Finally, there is the work of [31], which analyzes the current legislation associated with land use in Chile; however, in terms of spatial analysis, they work only in a watershed in the Biobío region.

Understanding changes in farmland is crucial due to their high productivity and the potential negative impacts of mismanagement. Additionally, Ref. [98] emphasize that only 25% of Chile’s soil cover has been adequately studied, indicating a critical lack of soil knowledge. Thus, there is a pressing need to analyze agricultural land cover dynamics beyond the highly productive south-central region, where geographical factors such as sparse vegetation density and complex dynamics like crop rotation pose challenges for obtaining comprehensive spatial data [99].

The review carried out in this study aims to contribute by highlighting the works that consider the dynamics of agricultural soils in Chile, helping to identify priority areas for future research. Additionally, it provides scientific evidence on agricultural land dynamics for the alternative’s evaluation phase, enabling the assessment of territorial impacts and potential to define suitable economic activities. Moreover, authors such as [100] emphasize that Geographic Information Systems (GIS) techniques and the development of multitemporal models are essential for decision-making, as they facilitate monitoring progress toward Sustainable Development Goals (SDGs), particularly SDG 2 (Zero Hunger) and SDG 1 (No Poverty).

This qualitative review has some limitations, such as the analysis of the studies included in the literature review, which depends on the interpretation and criteria of the researcher, which introduces possible subjectivity into the selection and interpretation of the results. Another significant limitation of the review is the non-inclusion of other databases, such as Scopus; however, we decided to focus on WOS due to its high-quality indexed journals. However, we consider that this review of articles corresponds to a first effort to explore the dynamics of agricultural land in Chile.

Authors like [101] emphasize that monitoring farmland presents challenges due to the need for spatially resolved data, leading to discrepancies in classification algorithms across different spatial scales. The review also reveals that studies focusing on croplands are typically conducted in small study areas, as [102] noted, where homogeneity in practices and environmental conditions facilitate more accurate modeling and local knowledge validation. For instance, studies such as [8,73] include northern Chile in their LULCC analyses; however, they do so at a national scale encompassing continental Chile, where agricultural areas are generally classified, allowing for comparisons with other land cover categories.

Fruit growing is a critical specialization in Chilean agriculture, playing a vital role in the economy by contributing up to 2.6% of the national GDP in recent decades and accounting for nearly 40% of the gross agricultural product [103]. The primary agricultural zone in Chile spans from 29° to 41° S in the central regions, encompassing approximately 95% of the country’s cultivated land [104]. This land is predominantly used for fruit production, vineyards, and horticulture [66]. Fruit crops are increasingly recognized as a high-growth economic cluster, leveraging Chile’s diverse geography and climate to produce high-quality fruits and various export products [105]. Ref. [45] noted a decline in agricultural land and grasslands between 1997 and 2013 in southern Chile, attributed to expanding forest plantations and urban areas. However, fruit crop areas increased from 2000 to 2019 despite the decline in regional agricultural land.

LULCC studies are fundamental for territorial analysis and for addressing global environmental change since they are spatial expressions of the links between the biosphere and economic structures, which have produced marked changes in the environment and society [106], enabling prospective models [8]. In this context, spatially analyzing and assessing land cover anticipates future problems and identifies priority areas for potential management and policy interventions through land and restoration programs. However, it is essential to analyze the characteristics and dynamics of the study area because this situation varies globally. In some studies conducted in Chile, like [45], urban areas replaced much fewer agricultural lands than forest plantations.

Having detailed LULCC data for farmlands at various scales will contribute to global discussions on issues like climate change, which affects crop yields by altering hydrological patterns, changing water levels and precipitation, and modifying temperature [58,107]. In Chile, this has been addressed through the adaptation plan for the Farming and Forestry sector [108]. However, authors like [29] emphasize the need for a legal framework that guarantees soil protection and appropriate use to preserve this non-renewable natural resource for future generations, highlighting its role in ensuring food security and mitigating climate change.

In Chile, land use regulation has primarily focused on creating instruments to regulate urban land use and establish rules for converting land use from natural ecosystems to anthropogenic uses [29]. Measures have been implemented to regulate rural land use management, such as Decree No. 4363 (Forestry Law) of 1931, which first demonstrated the state’s interest in promoting reforestation for erosion control. Law 20,412 is significant in the agricultural sector, introducing an incentive system for farmland recovery primarily applied in central-southern regions [109]. However, compared to the farming industry, the forestry sector has seen more legal provisions and development [110]. This situation underscores the importance of extending land cover change analyses beyond forestry to include new categories like cultivated land. Understanding the dynamics of these land cover classes is crucial for formulating territorial planning instruments to ensure adequate natural resource management. As mentioned above, this review is a first effort to identify the sectors of Chile where the dynamics of agricultural land use change have not been studied in depth, allowing researchers to focus their efforts on new geographic areas where it is relevant to know the state of land use change, such as northern Chile. We also recommend that new revisions of land use change are made, incorporating more recent years, covering new databases, and adding gray literature such as thesis documents, books, and notes, among others.

5. Conclusions

This review emphasizes the predominant focus of LULCC studies in Chile on forest dynamics, particularly in the central-southern region, starting in 2001. Four studies have explored farmland dynamics, examining aspects such as fruit cultivation, agricultural abandonment, drivers of change analysis, and predictive agricultural production. The emphasis on forest studies can be attributed to the abundant availability of spatial data for that region, including vegetation surveys and satellite imagery. The productivity and management intensity in these areas also underscore the importance of spatial analysis. A significant challenge lies in the complexity of spectral identification for agricultural coverage, which varies across Chile’s diverse geography. Many reviewed manuscripts operate at watershed scales, focusing on the water resource potential of farming activities. However, farmland dynamics at the watershed scale still need to be explored in the current body of research.

The results of this review highlight that economic factors often outweigh the benefits of agricultural production, leading to the widespread abandonment of agricultural land. Studies like that of [111] in the European Union similarly demonstrate that regions with significant mountainous terrain face higher risks of land abandonment due to challenges in accessibility, cultural heritage preservation, and societal preferences. These factors are also evident in Chile, as shown by [13], who identify proximity to transportation routes as a key accessibility issue and highlight the role of cultural heritage in shaping local identities through subsistence agriculture.

This study highlights the impact of spatial information on decision-making for sustainable agriculture. However, Territorial Planning Instruments (IPT) have not kept pace with these changes [112]. This is particularly evident in northern Chile, where scarce vegetation complicates spatial analysis. Ref. [113] predict that ongoing low precipitation will worsen water shortages, affecting productivity and conservation. On the other hand, land use regulation in Chile focuses on urban land [29], overlooking the global drivers of agricultural conversions affecting rural landscapes [3,114]. Analyzing changes in rural land cover is key to territorial management, conservation, and planning. A robust monitoring system is crucial to track modifications and identify dynamics, while effective regulations are needed to manage agricultural land impacts and measure compliance with the SDGs, which can facilitate the development of a system of indicators adapted to the specific conditions of a territory.