We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

Multiobjective analysis of open areas invaded by forest with open source software: the case of the SATURN project

00:00

Formal Metadata

Title
Multiobjective analysis of open areas invaded by forest with open source software: the case of the SATURN project
Title of Series
Number of Parts
351
Author
License
CC Attribution 3.0 Unported:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language
Production Year2022

Content Metadata

Subject Area
Genre
Abstract
Multiobjective analysis of open areas invaded by forest with open source software: the case of the SATURN project In northern Italian mountainous regions, forests are invading pastures and abandoned cultivated surfaces leading to an important land-use change phenomenon and reducing those open areas that are fundamental for ecological purposes [1]. The research here presented, focuses on a multiobjective and contemporary assessment methodology of two or more multicriteria analyses applied in the identification of the most suitable areas for agricultural purposes between those surfaces that have been invaded by forests carried out using Free and Open Source Software for Geospatial (FOSS4G) software. The analysis of the areas was determined by taking into account their intrinsic characteristics and their spatial location in relation to the territory and started from previous studies on land use in the Autonomous Province of Trento (Italy). The pilot areas are three municipalities that are part of Trento’s Province: the municipality of Trento - the Province’s capital, the municipality of Pergine Valsugana and seven municipalities that are part of the Piana Rotaliana region. Almost 88% of the Municipalities are located at an altitude of more than 600 m above sea level reflecting the peculiar topography of the province made up of valleys and high mountains with high percentages of steep slopes [2]. In Trento, the overall density is 742 inhabitants per square kilometers and the pressure on urban and peri-urban areas is nine times higher than the rest of the province [51]. 20% of Trento’s territory is classified as agricultural and 50% as forest or pasture land. About 70% of the territory is covered by silvopastoral -agricultural areas, the remaining 30% is categorized as urban. The repartition of the province’s surface is similar to the one of the city of Trento: 61% of the territory is covered by forests, 33.6% by agricultural areas, and only 5% by other types of land use. Collective bodies and public actors manage most of these silvopastoral -agro-forestal areas whose ownership is collective and is managed following the “uso civico” rights, a customary right embedded within the properties of communities and villages [52]. Therefore, profit is not their main aim. This study has been part of the SATURN European project [3] funded by EIT Climate-KIC (November 2018-December 2021). Three city-regions have been involved: the Trentino region in Italy, Birmingham in the United Kingdom, and Gotheborg in Sweden. The project aimed to reintegrate natural resources into cities' climate change adaptation strategies and to expand and nurture its model by creating a broader initiative involving an increasing number of stakeholders. Geospatial data set was georeferenced and managed with GRASS and QGIS and the files were collected combining data freely available at the Autonomous Province of Trento as well as self produced during the project. The comparative analysis and methodology were carried out by means of QGIS 3.8 Geographic Information System that has been used to complete the analysis in order to develop a methodology that can be widely used by territorial operators and Public Administrations. Through a series of multi-criteria analyses [4] of the agricultural and ecological vocation of a given region, and more specifically of abandoned agricultural areas, it was possible to create initial maps assigning values according to specific considered aspects. To lead these analyses, it was necessary to collect and select a significant amount of georeferenced data and then standardise them. Synthesis analyses have been useful to compare the ecological and agricultural aspects and to integrate them in synthesis maps, which can be used in the future for land management and planning. In order to validate the model and to verify the results, on-site inspections were carried out both in Valsugana and in Val d'Adige. Technicians and experts have been involved in the research through focus groups, organised within the SATURN project, which allowed some general criticalities of the territory to emerge, and through the completion of a questionnaire proposed within the thesis work. Through these questionnaires, it has been possible to identify the most important criteria for assessing a plot of land from an ecological and agricultural point of view. The obtained results showed how the classical approach, based on single criteria analysis, differ from the multicriteria approach for its potential to produce a more precise and clearer classification output of the aspects considered, showing the two multi-criteria analyses and their dependence on a single final map. Significant advantages have been taken from the use of this method in terms of data and information exchange between the stakeholders and in terms of a deepen understanding of the characteristics of the areas that have been analysed. The proposed methodology and the script that has been developed can be used in order to better plan forest management and as a basis for future territorial plans. Moreover, the multicriteria approach, which initially provides for a separate analysis of the research layers and then integrates them into a single final output, may represent a starting point for ecosystem evaluations. Preserving the ecosystem of an area, or rather the mosaic of ecosystems that make it up, is in fact of fundamental importance, as is succeeding in creating an eco-sustainable environment. In order to achieve this, it is necessary to have a spatial planning process that is as accurate as possible and that evaluates all the ecological criteria in a diversified manner with respect to the criteria of the object of research, so as to be able to identify key elements. The model presented can be replicated by changing the current research object, i.e. agricultural assessment, and keeping the ecological assessment instead. Future development will foresee the transformation of the Python script into a plug-in for Qgis, guaranteeing greater functionality for those who wish to use it.
Keywords
CASE <Informatik>Open setMathematical analysisForestOpen sourceIntegrated development environmentCivil engineeringExecution unitGreen's functionMultiplicationObservational studySystem identificationSurfacePlanningContext awarenessKey (cryptography)ResultantAreaIntegrated development environmentForestObservational studyConservation lawSystem identificationGoodness of fitOpen setComputer animation
SurfaceFundamental theorem of algebraForestIntegrated development environmentPlanningAreaMathematicsIntegrated development environmentComputer animation
MultiplicationClassical physicsFunction (mathematics)Mathematical analysisElement (mathematics)Integrated development environmentForestData managementSoftware maintenanceData integrityProcess modelingLarge eddy simulationPairwise comparisonData recoveryPerspective (visual)InformationSystem programmingParameter (computer programming)Archaeological field surveySatelliteFunction (mathematics)MereologyMathematical analysisAreaProjective planeSinc functionMultiplicationForestDifferent (Kate Ryan album)NetzwerkverwaltungClassical physicsFocus (optics)Process modelingINTEGRALIntegrated development environmentPerspective (visual)Pairwise comparisonWeightObject (grammar)Level (video gaming)Element (mathematics)Software maintenanceData recoveryOpen setArchaeological field surveyParameter (computer programming)Data managementComputer animation
AreaForestMereologyAreaProjective planeAutonomic computingUniform resource locatorMetre
Observational studyMereologyCASE <Informatik>AreaComputer animation
Feld <Mathematik>2 (number)AreaWater vaporForestMereologyStreaming mediaCASE <Informatik>Roundness (object)Observational studyPeripheral
CASE <Informatik>Field (computer science)Observational studyForest2 (number)Near-ringMIDIFlow separationComputer animation
Axonometric projectionUnified threat managementTransverse waveEllipsoidProcess modelingTransformation (genetics)ForestThermal expansionAreaTotal S.A.Time zoneSurfaceDistanceComputer networkData bufferPerformance appraisalCategory of beingGoodness of fitAbelian categoryVector potentialDatabase normalizationLimit (category theory)Constraint (mathematics)Maxima and minimaWeightParameter (computer programming)Local GroupIncidence algebraError messageDivisorLatent heatSample (statistics)Mathematical analysisRaster graphicsScripting languageGrass (card game)outputMultiplicationInformationDifferent (Kate Ryan album)Cellular automatonWebsitePoint (geometry)Characteristic polynomialFormal verificationPlot (narrative)Equals signTemporal logicElement (mathematics)Variable (mathematics)Operator (mathematics)Axiom of choiceInformationLimit (category theory)Mathematical analysisCategory of beingMappingConstraint (mathematics)Performance appraisalScripting languageAreaProcess modelingParameter (computer programming)Function (mathematics)Latent heatSampling (statistics)Transformation (genetics)Expert systemGroup actionGraph (mathematics)Error messageDivisorArchaeological field surveyWeightPoint (geometry)Software developerFunktionalanalysisMultiplicationUnified threat managementTime zoneGoodness of fitEllipsoidVector potentialTexture mappingCASE <Informatik>Observational studyGrass (card game)Type theorySingle-precision floating-point formatView (database)Interactive televisionElement (mathematics)Flow separationValuation (algebra)Replication (computing)System administratorRational functionAxiom of choiceVariable (mathematics)Context awarenessVulnerability (computing)ResultantPlug-in (computing)Operations support systemDigital photographyoutputThermal expansionCellular automatonProjective planeForestSoftware testing2 (number)Dependent and independent variablesNichtlineares GleichungssystemSimilarity (geometry)Different (Kate Ryan album)Graph coloringMatrix (mathematics)Field (computer science)Raster graphicsBinary multiplierDifferenz <Mathematik>Computer animation
Transcript: English(auto-generated)
Thank you. Good evening, everyone. I'm going to present the results of our research that we have conducted between 2020 and 2022 at the University of Trento. The aim of our study was to explore a multi-objective assessment method for the identification of most suitable areas for agricultural purposes
between host surfaces that have been invaded by forests. In the end, the method proved to be used to add in other environmental assessment for territorial planning purposes. The context in which we are moving is the European mountainous regions that are invading by forest in the last decades.
These areas, these open areas, are very important and play a key ecological role for biodiversity conservations. This phenomenon is particularly evident in the Trentino region in Italy, where an important land use changes is happening, reducing those open areas that are fundamental for ecological purposes.
To preserve these open areas is increasingly important to include policies for their protection in the regional planning, and above all, recognize their importance for the natural environment. It is also very important to quantify the phenomenon and classify those areas according
to their ecological vocation. Forest planning and land management are key elements in preserving these areas and ensuring the maintenance and the integrity of ecosystems. The literature identifies multi-criteria and multi-objective methods for environmental assessment, since they can examine multiple competing land use.
Through these approaches, different criteria and analyses can be compared both contemporarily and separately, and this methodological process provides a more precise and clear classification output than a classical multi-criteria analysis. Our research focused on how it was conducted during the SATUM
project, which was an EIT climate kick project that started in 2019 and ended in 2021. In the SATUM project, three areas were part of the SATUM project, the city of Gothenburg in Sweden, the city of Birmingham in the UK, and the Trentino region in Italy.
In the Trentino region, we have three main areas, which are the municipality of Trento, the municipality of Pergine, and the Piana-Rotaliana valley. The objective of our research was to play a multi-criteria analysis to apply that to identify the former open areas recognized
by forests that are most promising from an agricultural and ecological recovery perspective. The aim was to identify a method to run a multi-objective analysis that provide a simultaneous comparison of multi-criteria analysis using geographic information
system. The method was made up of four main stages, criteria definition in first stage, then a criteria weight, and then the comparison between multi-criteria analysis and multi-objective analysis. The tools we used was a QGIS 3.16,
and we used also questionnaires and surveys to better define the weight of the parameters that we choose. As mentioned before, the research has been conducted during the SATUM project in the autonomous province of Trento between 20 and 21.
The autonomous province of Trento is an Italian alpine region in northwest Italy, encompassing about 60,000 hectares with a population of roughly half a million people. The geography of the province, which is made up of valleys and high mountains, with large percentage of steep slopes,
is reflected in the fact that 88% of the municipalities are located at a height of more than 60 meters above sea level. Because of these geographical features, farming has always been challenging, and terraces were built to solve the problem.
Nowadays, agricultural machinery is frequently unsuitable for usage in such location. And therefore, most of the farming part of the farmer areas are moved in the valley floor, where it's easier to use the machinery. As I said before, we have three main areas
in our case studies, which is made up of nine municipalities. Seven municipalities are part of the Rotalliana Valley. One municipality is the Trento City, and the last municipality is Pergine Valsugana.
The city of Trento, which is also the capital of the region, is located along the Adige Valley. The major water courses that flow through the city of Trento are the Adige River, which runs through it from north and south, and the Fersina Stream, which runs through it from east to southwest.
In recent decades, the city has undergone a strong urban expansion toward the hillside, and Trento is now a very important tertiary center, with a high level of industries, activities, and peripheral areas. But also, the agricultural part is important, because about one third of the area
is covered by agricultural fields. And also, we have a very expanded forest. The second case study is represented by Pergine Valsugana, which is the third largest municipality in Trentino. The town is located in the middle of Trentino,
near or close to Trento, and Pergine Valsugana includes several hamlets. In Pergine Valsugana, the agricultural sector is very, very important, and also here, we have a very high percentage of the surface of the municipality that is covered by forest.
The third case study is represented by the Piana Rotaliana Valley, which is a very large plan for the Trentino region, in which the agricultural sector is the main economic sector. In fact, we have a very high percentage of fields that are covered by vineyards.
The data collection of our work started from an automatically induced classification maps obtained from the autophotos of 1954 and 2015, highlighting the areas of forest expansion, thanks to our previous research. The estimated surface involved in this transformation
is approximately over 3,000 hectares, corresponding to 11% of the total surface of the Trentino. UGIS 84 is the reference ellipsoid use for the map processing, and the projection use is UTM. The municipalities of interest for the project
are located in zone 32. The first step was the definition of a criteria that can be used for scoring the classification of the goodness or suitability of the area studies. The criteria has been divided into three categories. The first category represent the criteria suitable
for the evaluation of areas for agricultural restoration. Then we have criteria for ecological valuation. And at the end, we have created a third category that collect criteria used for both ecological assessment and agricultural potential. We have also defined constraints because in particular regarding the criteria used
for the agricultural assessment, we have many physical and political limits. Instead, the category regarding the ecological assessment are no subject to limitations. Firstly, all criteria were analyzed separately
to understand the particularity of the case, and then they have analyzed the contemporaries. After the definition of the criteria and the constraints, we have weighted the parameters that could be affected by the subjectivity of a researcher
because we don't have physical or normative limits to assign the weights. So to reduce the incidence of this error factor, questionnaires and surveys have been administered to specific samples of experts and technician. Specifically, we had two groups. The first group made up of five experts,
and the second group was made up of more than 100 people that respond to these questionnaires and surveys. After weighting the parameters, we developed two equation to calculate the agricultural importance of the areas analyzed
and also to quantify and assess the ecological aspects of the areas. We run three different methods to test and assess these areas. And then at the end,
the methods have been compared and discussed. The first two methods have been run through a QGIS raster calculator, and the third has been operated with a script. The first two have been compared before separately and then simultaneously. The first method consists in the substraction
between the ecological and agricultural criteria analysis previously normalized and multiplied. From this first analysis, it has been possible to determine which the two parameters ecological or agricultural prevails, and more specifically, the more positive the value,
the more the agricultural aspects prevails. Whereas the more the value is negative, on the contrary, the more of the ecological aspect prevails over the agricultural one. The second method is similar to the first, but instead of subtracting the values, we multiplied the agricultural criteria
to the ecological criteria. The result represents the importance of the area under the two combined aspects. From this second analysis, it has been possible to determine how the two aspects interact with each other. The third method, which is the more innovative,
compares the two multi-criteria analyses performed with raster data in GRASS through the specifically developed tools script. The script's inputs are the individual multi-criteria analyses, and in this specific case of the research, the inputs was the multi-criteria analysis
of the agricultural and ecological values previously performed. The script differs from the simple multiplication between the two rasters, so from this second method, because it is possible to know which value predominates over the other, because here in the matrix, as you can see,
we can see contemporarily the ecological value and the agricultural value of each cell. So different information and colors, graduations are associated with each cell, which corresponds to the map we developed. To verify the results, we did nine field visits,
and through them, we validated the model. So we went in these areas and verified the information we developed through the previous analysis. So the results. The first method, which is the agricultural
minus the ecological criteria value, give us the possibility to determine which of the two parameters, ecological or agricultural, prevailed over in the former open areas considered.
And as I said before, in this method, it is not possible to analyze contemporarily the two aspects. So we can only say that if the ecological value is more than the agricultural aspects or the contrary. The second method determines on the contrary
how the two aspects interact with each other, but through this method, it's not possible to discriminate which of the two parameters prevails over the other. So furthermore, the map can be used to identify are suitable for both agricultural restoration, but we cannot say which is an area
that is, for example, higher from an ecological point of view or agricultural point of view. The third method instead has been developed to obtain a multi-criteria analysis. So through this method, we can know which value prevails
and also how the two value interact between each other. The limits of our method are mainly three. The first is a limitation that is related to the dating of the maps. In fact, the temporal evaluation is a crucial element
in this map analysis. So if a map turns out to be dated, the model may suffer and become less accurate. The second limit is the assignment of weights that is another important element to be considered. In fact, it is the element most affected by the subjectivity of the operator.
And finally, we have a third limitation that could be considered as the last, that the method can be related to the choice of criteria. The type of analysis and variables involved in the multi-criteria evaluation has been chosen according to the quantity and quality of data available. Final consideration regards that the obtained results
showed how the multi-objective approach differs from the classical approach based on single-criteria analysis, with potential to produce a more precise and clearer classification output of the considered aspect, showing that the multi-criteria analysis and their dependence on a single final map.
And finally, the multi-criteria approach, which initially provides for a separate analysis of research layers, and then integrates them into a single final output may represent a starting point for the ecosystem evaluation. So connected to this last point, we have future developments that can be improved.
That is a replication of this model in other contexts, which can be, for example, a methodology for a risk assessment, in which we can, for example, assess three different aspects, hazard, vulnerability, and asset value in a given areas. Moreover, a future development could be the transformation
of a script into a plug-in for the QAGIS, guaranteeing greater functionality for whose wish to use it, for example, public administration in their planning purposes. Thank you for your attention.