LEaRNForME

Wiki quality control

Has flag	N/A

Name, responsible organisation and contact person

Has full name	Landslide Erosion and Runoff: Neural Forest Estimation ModEls
Has acronym	LEaRNForME
Has wiki contact person	Gianfranco Scrinzi
Has wiki contact e-mail	gianfranco.scrinzi@entecra.it

Software identification

Has software	LEaRNForME.Software

Description

Has description	LEaRNForME is an instrument for the land planning able to recognise the role of the vegetation cover in controlling some hydro geological instability phenomena. This evaluation will give the opportunity to introduce environmental issues into forest planning.
Has modelling scope	Forest indicators
Has temporal scale	Long term (strategic), Short term (operational)
Has spatial context	Spatial with no neighbourhood interrelations
Has spatial scale	Forest level, Stand level
Has objectives dimension	Multiple objectives
Has related DSS
Has goods and services dimension	Market wood products, Non-market services
Has decision making dimension	Single decision maker
Has forest management goal	multi-functional
Supports tree species	all the Italian forest types
Supports silvicultural regime	user defined

Concrete application

Has typical use case	one forest stand managed at a time
Has user profile
Has country	Italy
Has references about examples of application
Has number of users	<=10
Has number of real-life applications	<=10
Has utilisation in education	presentation/demo
Has research project reference
Has tool dissemination	scientific and technical publications, web site, conferences

Decision support techniques used in the DSS

Has decision support techniques	LEaRNForME.Decision support techniques

Support of Knowledge Management

Has knowledge management processes	LEaRNForME.Knowledge management process

Support of social participation

Has support for social participation	LEaRNForME.Support of social participation

DSS development

Has DSS development	LEaRNForME.Description of DSS development

Documentation

Has website	http://www.targetstars.org/ricerca_ex_isafa/riselvitalia43/index_eng.htm
Has online demo	http://www.targetstars.org/ricerca_ex_isafa/riselvitalia43/index_eng.htm
Has manual	Yes
Has technical documentation	Yes
Has reference	Andrenelli et al., 2007; Scrinzi et al., 2006; Gregori et al., 2007; Scrinzi et al., 2005

ADDITIONAL INFORMATION (needs to be migrated using the "edit with form" link)

Scope of the system

The schema of LEaRNForME

A procedure has been conceived and implemented in order to support the forester in classifying each homogeneous land unit for:

• predisposition to instability phenomena (termed "propensity") such as shallow landslides, water erosion and runoff generation;
• vegetation cover functionality in contrasting these events.

Both aspects have to be described and modelled by means of distinctive sets of variables; neural network analysis is then applied to a comprehensive set of study cases. The resulting evaluations of predisposition and functionality are combined into four different indexes with descriptive and planning significance:

• equilibrium level, a rough estimation of the balance between tendency and cover protection;
• protection value, assessment of the ability of the vegetation in controlling land degradation;
• constraint level, grade of limitation with respect to timber-oriented management compatible with the assessed protection value;
• action priority, preliminary screening of land units requiring ameliorative practices.

At present only the models devoted to the protection from shallow landslides and soil water erosion are operating; analogous procedure for runoff generation is in progress. To try the online-models go to this Web page: http://www.isafa.it/scientifica/assestamento/riselvitalia43/modelli_eng.htm/.

System origin

LEaRNForME was developed during 2001-2007 by Gianfranco Scrinzi, David Galvagni and Giacomo Colle in the Forest Monitoring and Management Unit (Unità di ricerca per il Monitoraggio e la Pianificazione forestale) (http://mpf.entecra.it/) of Italian Council for Research in Agriculture (Consiglio per la Ricerca e la Sperimentazione in Agricoltura) (http://sito.entecra.it).

Support for specific issues

LEaRNForME assists the forestry on site where shallow landslide, water erosion and runoff generation are instability phenomena.

Support for specific thematic areas of a problem type

• Forest planning.
• Territorial planning.
• Protection.
• Prevention.

Capability to support decision making phases

• Intelligence: the system use a fully automatic, congruent, progressive and recordable procedure.
• Design: forest planner input data to evaluate propensity and protective functionality of forest.
• Choice: results give operational and descriptive indexes for sustainable management.
• Monitor: implemented for test areas.

Typical spatial extent of application

The typical spatial extent of application is the forest management plan. There isn’t a base spatial unit for the output, it’s depend on what extension data input refer. But the minimum reasonable unit is the forest plot. Definitely we can say that the typical spatial extent of application is the single property scale.

Forest data input

The forest data input are about vegetation cover and forest management disturbances.

Type of information input from user (via GUI)

The user provides basic input data about the site conditions (lithology, aspect, slope, vegetation cover, erodibility, land use, practices, disturbances, etc.). In the shallow landslides model, the users input data to evaluate propensity to instability phenomena concerning: lithology, strata bedding, aspect, slope, climatic aggressiveness, drained area and seismicity. To evaluate vegetation cover protective functionality, the user input data about category of vegetation types, vegetation types, dominating vegetation cover, cover gaps, secondary vegetation cover, forest management disturbances, other disturbances, forest engineering practices, forest engineering practices. In the soil water erosion model the user input data to evaluate the propensity to instability phenomena concerning erodibility, climatic aggressiveness, heat load index, topographic factor. To evaluate vegetation cover protective functionality, the user input data concerning land use categories, land use, dominating vegetational cover, bushes cover, grasses cover, dead materials cover, agro-forest management disturbances, practices, slope.

Models

The system use Artificial Neural Network (ANN) models.

Definition of management interventions

The shallow landslides form

The soil water erosion form

The intersection of the vegetation functionality and land propensity to an instability phenomenon through complex mathematical functions gives four useful indices for soil conservation purposes. The user has to interpret this following indices.

• equilibrium level: this index provides a rough estimation of the balance between land tendency to the analysed geomorphologic phenomenon and cover protection. The “balanced” situation corresponds to conditions of almost equivalent levels of Propensity and Functionality; in these cases one can suppose that the protective action of the vegetation cover is adequate to face the onset of the degradation phenomena under examination. When the warrant provided by the vegetation increases the situation may be classified as “quiet”. On the opposite, the increase of Propensity generates a more or less unbalanced situation which may be defined as “uncertain” or “worrying” according to the algebraic difference between the two indexes;
• protection value: it assesses the ability of the vegetation in controlling land degradation. The protection value is classified into three classes (“high”, “fair” and “negligible”) according to a set of sigmoid functions;
• constraint level: this index expresses the degree of limitation with respect to timber-oriented management in relation to the assessed protection value. The classes of constraint are three: “none or negligible”, allowing the adoption of even very intensive cultural systems; “intermediate”, in which only cutting and harvesting techniques that assure the integrity of root apparatus and soil should be allowed; “maximum”, where only cares devoted to the enhancement of the protective functionality of the vegetation cover should be permitted (woodlands destined primarily to the environmental protection);
• action priority: this index assumes some planning importance by representing a preliminary screening of land units requiring ameliorative practices both for increasing the vegetation cover and for runoff control. The conceptual approach is addressed to prevent the degradation phenomena rather then to reclaim compromised situations. The model provides three classes of “action priority” (“high”, “intermediate” and “negligible”).

Typical temporal scale of application

Planning and operational level.

Types of decisions supported

• Management and sylvicoltural actions.
• Strategic decisions.
• Operating control decisions.

Types of outputs

The outputs of LEaRNForME Web system are qualitative indices. The outputs of LEaRNForME in stand-alone system are maps and table.

Spatial analysis capabilities

LEaRNForME in stand-alone system integrates weighted overlay capabilities using GIS functions.

System requirements

The Web application runs on multiplatform environment, only an Internet access and a Web browser is required. The stand-alone application requires ArcView 3.x and MS Access.

Architecture and major DSS components

LEaRNForME is both a stand-alone and a Web application. LEaRNForME is a modular system.

Usage

Forest management and planning, educational.

Computational limitations

No limitations for the Web application; in the stand-alone GIS application it depends on hardware and software used.

User interface

The user has to fill the tables for each instability phenomena: in some case he has to select between the proposed values in the pull down menu. It’s not necessary specific computer knowledge for using the system, but geological and forestry comptences is required to choose the appropriate classes for each variable.

Documentation and support

Information about LEaRNForME can be found at the Ri.Selv.Italia project website.

Installation

Installation not needed.

References

External resources

• Andrenelli M.C., Colle G., Galvagni D., Giannetti F., Gregori E., Scrinzi G., Zorn G.– Assessing the protective role of the forest cover against hydrogeological disturbances:a GIS-based tool for forest planning - IUFRO International Conference: "NATURAL HAZARDS AND NATURAL DISTURBANCES IN MOUNTAIN FORESTS" - Trento, ITALY (September 18th-21th, 2007);
• Gregori E., Andrenelli M. C., Zorn G., 2007 – Soil and hillslope management using scenario analysis and runoff-erosion models: a critical evaluation of current techniques. International Conference in Florence, ITALY (7th-9th May 2007);
• Andrenelli M. C., Galvagni D., Gregori E., Scrinzi G., Zorn G. Colle G., , 2006 – Propensione all’erosione idrica del suolo: un approccio valutativo in ambiente gis mediante l’utilizzo di reti neurali. - Convegno Nazionale SISS : "Suolo Ambiente Paesaggio", Imola 27-30 giugno 2006;
• Scrinzi G., Gregori E., Giannetti F., Galvagni D., Zorn G. Colle G., Andrenelli M. C., 2006 – Un modello di valutazione della funzionalità protettiva del bosco per la pianificazione forestale: la componente stabilità dei versanti rispetto ai fenomeni franosi superficiali. Forest@ 3 (1): 98-155;
• Scrinzi G., Gregori E., Giannetti F., Galvagni D., Zorn G., Colle G., Andrenelli M. C., 2005 - Boschi e fenomeni franosi superficiali: modello per la valutazione dell'azione protettiva. Estimo e Territorio (Ed agricole - il Sole24ore), anno LXVIII - n.12 - dicembre 2005, pp.30-47;
• Scrinzi G., Gregori E., Giannetti F., Galvagni D., Zorn G. Colle G., Andrenelli M. C., 2005 – Un modello di valutazione della funzionalità protettiva del bosco per la pianificazione forestale: la componente stabilità dei versanti rispetto ai fenomeni franosi superficiali., 66 p.

Every documentation about LEaRNForME and research processes can find in the http://www.targetstars.org/ricerca_ex_isafa/riselvitalia43/index_eng.htm