Interaction between agro-ecosystems and natural grasslands: stone graining and loss of natural ecosystems
Lead Author: CNR
Contributors: Valeria Tomaselli, Giuseppe Veronico (CNR-IBBR); Palma Blonda, Cristina Tarantino (CNR-IIA); Fasma Diele, Carmela Marangi, Angela Martiradonna (CNR-IAC).
Photo © Gianfranco Maiullari
Natural dry grasslands and scrubland on calcareous substrates are habitats that tipically characterize the Alta Murgia National Park in the southern Italian region called Apulia. Natural areas are alternated with semi-natural pastures, mainly grazed by sheep, and wide cereal crops that characterize the arid and bare Murgia land; two main habitats (according to the Annex I of 92/43 EEC Directive) characterize this site: "Pseudo-steppe with grasses and annuals of the Thero-Brachypodietea" (6220*), which includes thermo-Mediterranean xerophile annual grasslands rich in therophytes of calcareous, oligotrophic soils, with inclusions of some perennial communities (belong to classes Poetea bulbosae, Lygeo-Stipetea) and "Eastern sub-mediterranean dry grasslands (Scorzoneretalia villosae)" (62A0), this habitat incorporates sub-Mediterranean xeric grasslands with perennial important species of interest with endemic, rare species (Festuco-Brometea class). These habitats provide several ecosystem services that refer to the biodiversity maintenance (pollination, animals and plants refuge function, etc.) and to the support of ecotourism, education and research. Murgia natural landscape is closely linked with human presence that is expressed by crops, in particular cereals, and livestock breeding, in particular ovine. Good agricultural practices mainly supply food but also create manifold benefits for associated agro-biodiversity (ecological niches, habitat heterogeneity in rural landscape, health crops, etc.).
The regional Authority, in charge for compliance to the Habitat directive, is most concerned of the degradation of such natural habitats (loss, fragmentation, quality depletion), particularly in connection to the conservation of some raptors species (globally threatened and priority species according to the Bird Directive).
In recent decades the area, which is characterized by the presence of unique highly diverse ecosystems and also of endemic and threatened species, has been undergoing an awfully accelerated process of habitat fragmentation and contamination both within and at its borders by a number of combined pressures. Among these:
• the Common Agricultural Policy (CAP), which drove transformation of natural grasslands and semi-natural pastures into agricultural (cereal crops) areas by stone (rock) graining (clearance). In this process, rocks are crumbled through by mechanical means and are mixed with soil, with the aim of creating a substrate suitable for crops. Among the secondary effects, it induces soil erosion, sediment deposition and contamination of the aquifer;
• the illegal waste and toxic mud dumping on natural areas converted into illegal landfills, causing heavy metal contamination of soils and aquifer systems;
• the increasing of traditional legal and illegal mining;
• the expansion of wind farms;
• the spread of invasive species
All these factors of risks cause an alteration of the delicate balance of natural habitats, and indirectly have an impact also on the entire crop production cycles and livestock; as a consequence, the most important ecosystem services have been compromised, not only agriculture for human well-being, such as provision of crop products and dairy products (i.e. milk, cheese, etc.), but also a big fallout on the entire grassland ecosystem in terms of habitat fragmentation and biodiversity loss. Pedological studies have showed that the structural characteristics of soil have worsened in the last decades, organic matter percentage has decreased, whilst the fine inert particles increased. This means that, during heavy rains, part of these soils tend to slide down from the slopes and to accumulate in depressions, where anthropic artefacts are often present. The amount of soil that flows down also increases when plowing is made along the lines of maximum slope (known locally as "rittochino" plowing ).
The crop type influences level and severity of the erosion process: cereal crops, very common in Alta Murgia and covering the soil surface for several months per year, raise the vulnerability of the sites subject to rock clearance in terms of soil loss, especially during the most extreme weather events.
Rock clearance affects also the anthropic landscape; in facts, ancient structures such as dry stone walls, beaten roads, rural tracks, wells, water and snow stone tanks, etc. and other rural artefacts are been "grounded" by unscrupulous farmers to give place to few square meters of naked land for agricultural purposes.
Illegal dumping (toxic sludge, special hospital waste, leather waste products, ferrous scraps, etc.) causes water and soil poisoning with serious consequences for the agricultural sector: wide areas become no longer cultivable; high percentages of chrome, copper, zinc and cadmium poison the soil. Overall pollution causes a decrease of agricultural yield (dairy products, vegetables, cereals, etc.), a quality decline of Murgia typical products and a field's abandonment.
This storyline focuses on two aspects relevant for the management of the protected area:
a) the use of remote sensing for monitoring land cover and habitat changes induced by anthropic pressure and climate change
b) the combined use of remote sensing and modelling to detect and control invasive species
Recent developments
(1) Monitoring land cover and habitat changes
The objective of the activity was the use of intra-annual Time Series (TS) Landsat images to exploit the additional contribution of high temporal frequency of acquisition (16-days) and generate the time series of Modified Soil Adjusted Vegetation Index (MSAVI) for Land Cover and Habitat change mapping. The focus was on monitoring the extension and the changes of grasslands ecosystems in the Murgia Alta site, according to the European Union Biodiversity Strategy to 2020.
Initially, the grassland LC mapping was performed by supervised classification using Support Vector Machine (SVM) classified with a RBF kernel function (after tuning of parameters after Cross-Validation) for a 10-classes problem. Two different mapping for 2011 and 2017 were obtained considering as input to the classifier an intra-annual MSAVI TS for each year.
Then the land cover change detection for grasslands from 2011 to 2017 was approached by comparing different algorithms as Post Classification Comparison (PCC), Cross-Correlation Analyses (CCA) and BFAST. The mapping of different habitat in grasslands environment was performed by supervised classification using Support Vector Machine (SVM) classified with a RBF kernel function for a 5-classes problem. The mapping for 2017 was obtained considering as input to the classifier an intra-annual MSAVI TS.
The use of intra-annual time series allowed to consider the phenology evolution of vegetation LC classes: grasslands was among the better identified LC classes (high accuracy ≈90% for 2011 used as reference T1 map for change detection). Regarding the validation, although for the LC maps ground truth can be obtained by existing data or new in-field campaign with the aid of the earthtrack application developed by Richard Lucas, the validation of the change maps is an open issue. The approach used in the case study was the convergence of evidence by means of results obtained from different methods. For those pixels of changes detected by both CCA and PCC methods, information about where they occurred in a certain period and the specific arrival transition can be known. Bfast can add information about the year/month when changes occurred. The results confirmed that the intra-annual TS perform better than 4 multi-seasonal VHR images for the habitat mapping. All the maps and data can be can be ingested in EODESM.
The use of spectral indexes from Sentinel-2 is a further development of the study.
(2) Detecting invasive species from space
Alien plants can modify biodiversity and functioning of ecosystems causing their degradation by diminishing both abundance and survival of native species. On-going climatic and anthropic changes are making invasive species spreading a global issue whose monitoring and management would require powerful remote sensing data and techniques. CNR-IIA, in the framework of the ECOPOTENTIAL project, studied new solutions for the automatic detection of Ailanthus altissima (Mill.) Swingle, one of the most widespread and harmful invasive plants in both the USA and Europe.
In Murgia Alta National Park, Ailanthus can grow mainly in semi-natural and natural grasslands fields due to shepherds’ practices abandonment or at the edges of cultivated fields, mainly herbaceous areas. Within such fields, the invader is generally controlled by the farmers through regular ploughing or other agricultural practices. Occasional fires occurring in this area can also be favoured by the proliferation of this plant.
The study introduces a novel investigation approach of the Ailanthus altissima species by analysing multi-spectral and multi-temporal Very High Resolution (VHR) satellite data (i.e., WV-2, characterized by 8 spectral bands in the visible/near infrared electromagnetic spectrum and 2 meters spatial resolution) as an alternative to the unavailability of hyperspectral data from aerial campaigns (characterized by a finer spectral resolution essential for the discrimination at species level). The technique used relied on a two-stage hybrid classification process to obtain the Ailanthus. altissima mapping: the first stage applied a knowledge-driven learning scheme to provide a land cover map (LC), including deciduous woody vegetation and other classes, without the need of reference training data; the second stage exploited a data-driven classification to: i) discriminate pixels of the invasive species found within the deciduous vegetation layer of the LC map (two-classes problem); ii) determine the most favourable seasons for such recognition.
Since training data to discriminate Ailanthus altissima from other deciduous plants are required only in the second stage, the use of the first stage LC mapping as pre-filter can reduce not only classification complexity but also time and costs involved by in-situ reference data collection.
The best encouraging Overall Accuracy (OA) value of 97.96% for the Ailanthus altissima mapping was obtained considering the July and October WV-2 images as input to the Support Vector Machine classifier in the second stage.
Although the methodology proposed and the data used would require further applications for the mapping of Ailanthus altissima and other invasive species in different sites, the use of multi-temporal VHR data and the hybrid classification approach may offer new opportunities for invasive plant monitoring and follow up of management decisions.
Graphical abstract from:
C. Tarantino, F. Casella, M. Adamo, R. Lucas, C. Beierkuhnlein, P. Blonda. (2018). “Ailanthus altissima mapping from multi-temporal very high resolution satellite images”, ISPRS Journal of Photogrammetry and Remote Sensing, 147, 90-103, https://doi.org/10.1016/j.isprsjprs.2018.11.013.
(3) Modelling the spread and the control of invasive species
Invasive species management is one of the most important topics in natural resource management, due to the environmental and economic damage that they cause. A recent report from the European commission estimates that invasive species cost the European Union at least 12 billion per year. This figure includes costs for key economic sectors, as agriculture, fisheries, aquaculture, forestry and health sectors as well as damages and management costs. Moreover, invasive species are also a major cause of global biodiversity loss. Hence, enormous benefits for the hosting ecosystems can be obtained by controlling invasive species or even eliminating them, when feasible and advisable. Unfortunately, this is a costly endeavour which requires careful planning to ensure cost-effectiveness, especially in protected areas, where resources are often too scarce to face all the pressures generated by internal and external drivers. Very often, it is necessary to apply control actions of widespread invasive species again and again, since wind and roads are constantly transferring the infection from outside. It is then of outmost importance to develop cost-effective tools for both monitoring and controlling the spread of the invasive species in a variety of scenarios, potentially including also climate change effects.
Within ECOPOTENTIAL, we combined remote sensing techniques and mathematical modelling to support the control of one of the so called tree of heaven (Ailanthus altissima).
The approach, which is a result of a collaboration of CNR (IAC, IIA, ISPA) and colleagues of CSIRO and University of Ferrara, has been tested on the Alta Murgia National Park, where an ongoing LIFE project is carrying out an eradication program, providing data and expert knowledge useful to build up the model. The initial map of presence of Ailanthus altissima, as well as the land cover map of the site, which are input to the model, have been generated by CNR-IIA, by applying remote sensing techniques to time series of very high resolution images. The model developed by CNR and CSIRO is based on a differential equation system and produces the optimal allocation of resource in both space and time, needed to dynamically control the spread of the species. A further relevant assumption of the model is the constraint of a maximum available budget to perform the eradication task.
The model has been implemented in open source software, and the workflow is available on the ECOPOTENTIAL Virtual Lab.
Details on the model on:
C.M. Baker, F. Diele, C. Marangi, A. Martiradonna, S. Ragni, Optimal spatiotemporal effort allocation for invasive species removal incorporating a removal handling time and budget, Natural Resource Modeling,31(4), 2018, doi: 10.1111/nrm.12190.
References
http://www.parcoaltamurgia.gov.it/
http://www.altramurgia.it/index.php?option=com_frontpage&Itemid=85
http://www.paesaggio.regione.puglia.it/
http://www.sit.puglia.it/portal/sit_portal
Andriani G.F. & Walsh N., 2008. An example of the effects of anthropogenic changes on natural environment in the Apulian karst (southern Italy). Environ Geol n. 58 (2009), pp. 313–325. Springer.
Calò F. & Parise M., 2006. Evaluating the human disturbance to karst environments in southern Italy. Acta Carsologica n. 35/2, pp. 47–56, Ljubljana
Campedellia T., Londi G. , La Gioia G. , Frassanito A.G. & Tellini Florenzano G., 2015. Steppes vs. crops: is cohabitation for biodiversity possible? Lessons from a national park in southern Italy. Agriculture, Ecosystems and Environment n. 213 (2015), pp. 32–38. Elsevier.
Canora F., Fidelibus M.D., Sciortino A. & Spilotro G., 2008. Variation of infiltration rate through karstic surfaces due to land use changes: A case study in Murgia (SE-Italy). Engineering Geology n. 99 (2008), pp. 210–227. Elsevier.
Caprioli M. & Tarantino E., 2006. Identification of land cover alterations in the Alta Murgia National Park (Italy) with VHR satellite imagery. Int. J. Sus. Dev. Plann. Vol. 1, n. 3 (2006), pp. 261–270.
Carbonara S., 2007. Il territorio agricolo tra politiche di settore e pratiche urbanistiche. Atti di convegno. XXXVII Incontro di Studio del Ce.S.E.T. - Ferrara 19-20 novembre 2007.
Ciani E., Tedone L., Terzi M., De Cillis F.M., Castellana E. & Fracchiolla M. 2012. Characterization of plant diversityad Author: of pastures and volatile organic compound analysis in ewe’s milk from a typical farm system in the Alta Murgia national Park (southern Italy): opportunities for a sustainable land use. Italian Journal of Agronomy, volume 7: e19. Page Press.
Diele F., Marangi C., and Ragni S., 2012a. Exponential Runge-Kutta integrators for modelling Predator-Prey interactions. AIP Conference Proceedings 1479, 1181; doi: 10.1063/1.4756361
Diele F., Marangi C., and Ragni S., 2012b. Implicit - symplectic partitioned (IMSP) Runge-Kutta schemes for predator-prey dynamics, AIP Conference Proceedings 1479, 1177; doi: 10.1063/1.4756360
Fidelibus M.D., 2007. Meccanismi di trasporto degli inquinanti in acquiferi carsici sotto eventi estremi (Alta Murgia, Puglia). Periodico Geologi e territorio n. 3-4 (2007), pp. 77-85.
Forte L., Perrino E.V. & Terzi M., 2005. Le praterie a Stipa austroitalica Martinovsky ssp. austroitalica dell’Alta Murgia (Puglia) e della Murgia Materana (Basilicata). Fitosociologia n. 42 (2) (2005), pp. 83-103.
Macchia F., Cavallaro V., Forte L. & Terzi M., 2000. Vegetazione e clima della Puglia. In : Marchiori S. (ed.), De Castro F. (ed.), Myrta A. (ed.). La cooperazione italo-albanese per la valorizzazione della biodiversità. Bari: CIHEAM, 2000. pp. 33-49 (Cahiers Options Méditerranéennes; n. 53).
Perrino E.V. & Wagensommer R.P., 2013. Habitats of Directive 92/43/EEC in the National Park of Alta Murgia (Apulia—Southern Italy): Threat, Action and Relationships with Plant Communities. Journal of Environmental Science and Engineering A 2 (2013), pp. 229-235. David Publishing.
Perrino E.V., Brunetti G. & Farrag K., 2014. Plant Communities in Multi-Metal Contaminated Soils: A Case Study in the National Park of Alta Murgia (Apulia Region - Southern Italy). International Journal of Phytoremediation, n. 16: 9, pp. 871-888.
Perrino P., Laghetti G. & Terzi M., 2006. Modern concepts for the sustainable use of Plant Genetic Resources in the Mediterranean natural protected areas: the case study of the Alta Murgia Park (Italy). Genetic Resources and Crop Evolution n. 53 (2006), pp. 695–710. Springer.
Regione Puglia - Ufficio Parchi e Tutela della Biodiversità (a cura di), 2014. Grastepp: tra Gravine e Steppe. Azioni per la conservazione della biodiversità nel Parco Nazionale dell’Alta Murgia e nel Parco Naturale Regionale Terra delle Gravine. Rapporto Finale. Never Before Italia srl, Castellana Grotte (BA).
Terzi M., Di Pietro R. & D’Amico F.S., 2010. Analisi delle Specie Indicatrici applicata alle comunità a Stipa austroitalica Martinovsky e relative problematiche sintassonomiche. Fitosociologia vol. 47 (1), pp. 3-29.
****
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 641762.
Last update: March 2020