Ecosystem Functional Types as an EBV to characterize functional diversity in Sierra Nevada
Lead Author: University of Granada and University of Almería
Introduction
A better understanding of ecosystem functioning and functional diversity is key to the management of nature and its services (Jax 2010) and to determine a planetary boundary to promote sustainability (Chapin et al. 2010) and a safe operating space for humanity (Steffen et al. 2015). The Convention on Biological Diversity (CBD 2013) has adopted the Essential Biodiversity Variable framework (Pereira et al. 2013) to coordinate monitoring programmes worldwide, including an ecosystem function class and a focus on satellite remote sensing (Pettorelli et al. 2017, 2018). Ecosystem Functions address how systems perform (Petchey & Gaston 2006), and provide the links between biological diversity (Chapin et al. 2010; Pereira et al. 2013), ecosystem services (Balvanera et al. 2006) and ecological resilience (Mouchet et al. 2010). In addition, ecosystem functional variables are particularly interesting for monitoring programs since environmental changes are particularly noticeable at the ecosystem level (Vitousek & Hooper 1994). Functional attributes show a quicker response to disturbances than structural ones (Milchunas & Lauenroth 1995) and they can be easily monitored using remote sensing at different spatial scales, over large extents, and under common protocols in space and time (Paruelo et al. 2001).
In Sierra Nevada (SE Spain), a Mediterranean high-mountain biodiversity hotspot (Myer et al. 2000), vegetation studies have been developed under a compositional perspective (phytosociological method) or successional perspective (vegetation series). These studies have been very useful for describing the vegetation heterogeneity at mesoscale (Loidi 2017), they have been also the basis for the characterization of habitats of interest for conservation (Directive 92/43/EEC), and they served for the development of forest restoration policies (Valle et al. 2003). However, these approaches are difficult to monitor the effects of management actions, to understand the environmental gradients at protected area scale that underlie biodiversity, and to evaluate the role of ecosystems in providing benefits to society (Cabello et al. 2019).
Here, we propose Ecosystem Functional Types (Alcaraz-Segura et al. 2013) as a functional classification of vegetation that synthetically characterizes ecosystem functioning and allows to assess functional diversity at regional scale. First, we analyzed the spatial patterns of three ecosystem functional attributes (i.e., annual primary production, and the seasonality and phenology of carbon gains), as well as their integration into a synthetic mapping of ecosystem functional types (EFTs). Second, to identify hotspots of ecosystem functional diversity in Sierra Nevada, we used two ways: richness and rarity of ecosystem functional types. Finally, we showed the most stable and variable zones between years (either by directional changes or by fluctuations) in terms of ecosystem functioning, we evaluated the inter-annual variability in ecosystem functioning from two measures, the number of EFTs that were observed during the period 2001-2016 at pixel level, and the inter-annual similarity in the composition of EFTs at landscape level. In all cases, to make it easier for the reader to interpret the spatial patterns found, we provided a comparison with the structural vegetation types of Sierra Nevada.
Methods
To characterize ecosystem functioning of Sierra Nevada Protected Area, we used time series of satellite images of Enhanced Vegetation Index (EVI, MOD13Q1-MODIS product), from 2001 to 2016. Three functional attributes that describe ecosystem functioning were calculated from the EVI seasonal curve: an estimator of annual total amount (annual Mean), a descriptor of seasonality (the seasonal coefficient of variation sCV), and an indicator of phenology (the date of the maximum MMAX). To synthesize the functional heterogeneity of all these continuous metrics into a discrete classification, we identified EFTs following Alcaraz-Segura et al. (2013): i.e., the range of values of each metric was divided into four intervals, giving a potential number of 64 EFTs (4 × 4 × 4). For DMAX, the four intervals agreed with the four seasons of the year. For Mean and sCV, we extracted the first, second, and third quartiles for each year and then calculated the inter-annual mean of each quartile for the 16-year period (Figure 1). 
Figure 1: Ecosystem Functional Types workflow. Times-series of EVI images are used to produce three meaningful descriptors of ecosystem functioning: primary productivity, seasonality and phenology. These three continuous ecosystem functional attributes (EFAs) are then combined into a discrete classification of Ecosystem Functional Types (EFTs).
To characterize functional diversity we used EFT richness and rarity. Richness was calculated for each year by counting the number of different EFTs within an 4×4-pixel moving window across the study area (Alcaraz-Segura and others 2013). Then, an average richness map across all years was obtained. EFT rarity was calculated as the relative extension of each EFT compared to the most abundant EFT (Equation 1) (Cabello and others 2013).
Rarity of EFTi = (Area_EFTmaxArea_EFTi)/Area_EFTmax (Equation 1)
where Area_EFTmax is the area occupied by the most abundant EFT and Area_EFTi is the area of the i EFT being evaluated, with i ranging from 1 to 64.
To identify which were the most stable areas and the areas with the greatest inter-annual variability in ecosystem functioning, we calculated the number of different EFTs that occurred in the same pixel along the 2001-2016 period. As an additional measure of the interannual variation that considered the movements in EFT composition that may occur at the landscape level, the dissimilarity index (1Jaccard similarity index; Jaccard, 1901) was used in 4x4 pixels MODIS mobile windows (924 x 924 m; ?1 km2)..
Results and discussion
Ecosystem functional attributes showed a clear altitudinal pattern (Figure 2).
Thus, the lowest primary productivity values occurred in the cryo- and oromediterranean (Figure 3 a and b), in grasslands, high mountain rocks and pastures.
The highest values were observed in the supra- and mesomediterranean areas associated with oak groves, holm oaks, coniferous repopulations and native pine forests of Pinus sylvestris subsp. nevadensis. At the eastern and western ends of the protected area, in thermo- and mesomediterranean areas (Figure 3 a) intermediate values of productivity were obtained, being medium-high in the western zone, and medium-low in the eastern zone, both dominated by scrub, and mid-mountain pastures (Figure 2 a and b). For seasonality (Figure 2 c and d), understood as the coefficient of variation of EVI, we found an inverse pattern to that of productivity, i.e. high values in the cryo- and oromediterranean that decrease as we go down in altitude towards the supra-, meso- and thermomediterranean areas (Figure 3 a). The crioro- and oromediterranean were characterized by more seasonal ecosystems, such as pastures, high mountain rocks, borreguiles, and high mountain scrubland, where snowfalls are the determining limiting factor. In the supra- and mesomediterranean (Figure 3 a) we also find high values of seasonality due to the presence of oak groves. In addition, there are native pine forests of Pinus sylvestris subsp. nevadensis and mid-mountain pastures with average seasonality values, and holm oaks, coniferous plantations and mid-mountain scrub with low seasonality (Figure 2 d). When descending more in altitude, towards the eastern part of the protected space, in the meso- and thermomediterranean areas, the values of seasonality became medium-high again (Figure 2 c).With respect to phenology, in the cryo- and oromediterranean, in the vegetation types of grasslands, canchales, rocks and borreguiles, the moment of maximum greenness of the vegetation in summer (July-August) dominated (Figure 2 e). In the supra- and mesomediterranean, associated with mid-mountain grasslands and scrublands, autochthonous pinewoods, oaks and holm oaks, the moment of maximum greenness of the vegetation used to occur in late spring (May-June). However, some western and southern valleys showed small areas with maximum greenness during the early autumn and winter months, even becoming winter in the eastern end of the semi-arid thermo-Mediterranean (Figure 2 f).
Figure 2: Ecosystem Functional Attributes and their values in natural vegetation types of Sierra Nevada national park. Vegetation types: Rocky fields, Borreguiles, Scrub High Mountain (Scrub HM), Grass High Mountain (Grass HM), Scrub Mid Mountain (Scrub MM), Pinus sylvestris (Pin Syl), Holm Oak (H Oak), Oak and Crops.
As a result of the combination of the three functional attributes of the canopy, mean productivity, seasonality and phenology, represented in Figure 2, we obtained the EFTs map (Figure 3 e) that includes a synthetic characterization of the spatial patterns of ecosystem functioning. The most abundant EFT presented the maximum greenness in spring, with productivity values from low to intermediate and under all possible seasonality values: Aa1, Ba1, Cb1, Cd1, Bb1, and Cc1 accumulated 37% of the surface of the Sierra. By other hand, the rarest EFT were Bc2, Ca4 and Ba3, characterized by medium productivity, medium or high seasonality and a maximum of summer, winter and autumn respectively (Figure 3 f).
Figure 3: Ecological classifications of Sierra Nevada. a) Bioclimatic areas and b) area occupied by each category; c) Vegetation types and d) area occupied by each category; e) Ecosystem functional types based on the Enhanced Vegetation Index (EVI),and d) Relative abundance of each EFT. The classes of EFTs are indicated in the legend. Vegetation types: Rocky fields, Borreguiles, Scrub High Mountain (Scrub HM), Grass High Mountain (Grass HM), Scrub Mid Mountain (Scrub MM), Pinus Sylvestris (Pin Syl), Holm Oak (H Oak), Oak and Crops.
The greatest EFTs richness was observed in the supra- and mesomediterranean, particularly in the southern face of the national park (Figure 4 a), where the number of vegetation series is also greater than in the rest of the bioclimatic areas (Valle et al. 2003). The presence of EFTs richness hotspots mainly in the mid-mountain, and particularly in the southern face, could be related to two factors. On the one hand, many Mediterranean mountains show high values of beta diversity up to 1750-1800 m (Wilson & Schmida, 1984), when there is an important structural and compositional replacement of their vegetation. On the other hand, in the middle mountain and especially in its southern face there is a very diverse mosaic of different types of natural vegetation mixed with different types of reforestation, traditional crops and uses (Camacho et al. 2002), which gives them the qualifier of multifunctional landscapes from the point of view of the provision of ecosystem services (García-Nieto et al., 2013; Cabello et al., 2019). In contrast, the lowest EFTs richness were located in the oro- and cryoromediterranean areas, and in the eastern semi-arid thermomediterranean extreme, where soil and climatic conditions (Martín Peinado et al., 2019) diminish floristic diversity although endemicity increases (Fernández Calzado et al., 2012).
Rarity of EFTs was highest in the cryoromediterranean, the area with the highest concentration of endemisms (Cañadas et al., 2014; Peñas et al., 2019) (Figure 4 c). Cryo-Mediterranean vegetation are developed under very particular ecological conditions that determine uncommon types of ecosystem functioning (rarity 0.6; Figure 4 d), such as, for example, in relatively mobile rocks located on steep slopes, where the percentage of rarity or compositional endemicity rises to 80% (Blanca & Algarra, 2011). EFTs rarity was also very high in the eastern end of the semi-arid thermomediterranean, with a high concentration of endemisms typical of the Desert of Tabernas (Mota et al., 2004) (Figure 4 c). Lowest rarity appeared in the oromediterranean, due to the great extension of this bioclimatic area in the Sierra Nevada (Figure 3 a and b), made that its functioning not appear as rare, and increasing again in the supra- and mesomediterranean (Figure 4 c).
Interannual variability was higher in the supra- and mesomediterranean areas, coinciding with the altitudinal range where interannual climate variability is most affected (e.g., they may present a lot of snow in cold years and be affected by drought in dry and warm years). Eastern part of national park also showed high interannual variability, due to the greater climate fluctuation and where small changes in precipitation produce large changes in the dynamics of primary production (Cabello et al., 2012). On the other hand, the most stable vegetation types interannually, i.e., those that changed the least during the period, were located in the meso-oromediterranean and crioromediterranean, specifically, the oak and borreguil vegetation types, ecosystems that are not subject to anthropic presence (e.g., low forest management and low presence of livestock). Dissimilarity pattern increased towards lower floors, finding the highest values of dissimilarity (or major change) in areas where changes in land use and management are most present (Zamora et al., 2015).
Figure 4: Functional diversity patterns based on the Enhanced Vegetation Index (EVI). a) Spatial richness of EFTs from a 4x4 pixel MODIS mobile window (?1 km2) and b) EFT richness values per vegetation type; c) Spatial rarity patterns of EFTs and d) values per vegetation type; e) Interannual variability of EFTs for the period and f) values per vegetation type; g) Interannual dissimilarity of EFTs or 1 - Jaccard coefficient for the period and h) values per vegetation type. Vegetation types: Rocky fields, Borreguiles, Scrub High Mountain (Scrub HM), Grass High Mountain (Grass HM), Scrub Mid Mountain (Scrub MM), Pinus Sylvestris (Pin Syl), Holm Oak (H Oak), Oak and Crops.
Conclussion
This study provides a characterization of ecosystem functioning of the Sierra Nevada Protected Area through the analysis of time series of satellite images of spectral indices that capture the photosynthetic activity of the vegetation. The combination of functional attributes in a synthetic classification of Ecosystem Functional Types integrates in a single map the spatial and temporal heterogeneity of carbon gains by vegetation. On the other hand, the use of EFTs as biological entities allowed analyze spatial patterns and inter-annual variability in functional diversity at the ecosystem level and revealed the existence of hotspots of functional diversity, as well as more stable areas and others with greater variability between years.
References:
Alcaraz-Segura, D., Paruelo, J. M., Epstein, H. E. & Cabello, J. (2013). Environmental and human controls of ecosystem functional diversity in temperate South America. Remote Sensing, 5(1), 127-154.
Balvanera, P., Pfisterer, A. B., Buchmann, N., He, J. S., Nakashizuka, T., Raffaelli, D., & Schmid, B. (2006). Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecology letters, 9(10), 1146-1156.
Blanca, G. & Algarra, J.A. (2011). Flora del Espacio Natural de Sierra Nevada. In: L. Serra Laliga (Ed.), Jornadas Estatales de Estudio y Divulgación de la Flora de los Parques Nacionales y Naturales (pp. 2136). Alcoi: CAM.
Cabello, J., Alcaraz-Segura, D., Ferrero, R., Castro, A. J. & Liras, E. (2012). The role of vegetation and lithology in the spatial and inter-annual response of EVI to climate in drylands of Southeastern Spain. Journal of Arid Environments, 79, 76-83.
Cabello, J., Pacheco, M., Rodríguez, M.D. & Torres T. (2019). Valores y argumentos para la conservación de la flora y vegetación de Sierra Nevada. In: Biología de la Conservación de Plantas en Sierra Nevada. In: Peñas, J. & J. Lorite (eds.), Biología de la Conservación de plantas en Sierra Nevada. Principios y retos para su preservación. Editorial Universidad de Granada.
Camacho Olmedo, M.T., García Martínez, P., Jiménez Olivencia, Y., Menor Toribio, J. & Paniza Cabrera., A. (2002). Dinámica evolutiva del paisaje vegetal de la Alta Alpujarra granadina en la segunda mitad del S. XX. Cuadernos Geográficos, 32, 2542.
Cañadas, E. M., Fenu, G., Peñas, J., Lorite, J., Mattana, E., & Bacchetta, G. (2014). Hotspots within hotspots: Endemic plant richness, environmental drivers, and implications for conservation. Biological Conservation, 170, 282-291.
CBD (2013). Essential Biodiversity Variables. Seventeenth meeting Montreal, 14-18 October 2013 (https://www.cbd.int/doc/meetings/sbstta/sbstta-17/information/sbstta-17-inf-07-en.pdf).
Chapin III, F. S., Carpenter, S. R., Kofinas, G. P., Folke, C., Abel, N., Clark, W. C., ... & Berkes, F. (2010). Ecosystem stewardship: sustainability strategies for a rapidly changing planet. Trends in ecology & evolution, 25(4), 241-249.
Fernández Calzado, M. R., Molero Mesa, J., Merzouki, A. & Casares Porcel, M. (2012). Vascular plant diversity and climate change in the upper zone of Sierra Nevada, Spain. Plant Biosystems-An International Journal Dealing with all Aspects of Plant Biology, 146(4), 1044-1053.
García-Nieto, A. P., García-Llorente, M., Iniesta-Arandia, I. & Martín-López, B. (2013). Mapping forest ecosystem services: from providing units to beneficiaries. Ecosystem Services, 4, 126-138.
Jax, K. (2010). Ecosystem functioning. Cambridge University Press.
Martín Peinado, M., Jiménez Morales, M.N. & Fernández Ondoño, E. (2019). Los suelos de Sierra Nevada, su relación con la litología y la vegetación. In: Peñas, J. & Lorite, J. (eds.), Biología de la Conservación de Plantas en Sierra Nevada. Principios y retos para su preservación. Editorial Universidad de Granada.
Milchunas, D. T., & Lauenroth, W. K. (1995). Inertia in plant community structure: state changes after cessation of nutrient?enrichment stress. Ecological Applications, 5(2), 452-458.
Mouchet, M. A., Villéger, S., Mason, N. W., & Mouillot, D. (2010). Functional diversity measures: an overview of their redundancy and their ability to discriminate community assembly rules. Functional Ecology, 24(4), 867-876.
Paruelo, J. M., Jobbágy, E. G., & Sala, O. E. (2001). Current distribution of ecosystem functional types in temperate South America. Ecosystems, 4(7), 683-698.
Peñas, J., Cañadas, E., & Del Río, J. (2019). Fitogeografía de Sierra Nevada e implicaciones sobre su conservación. In: Peñas, J. & Lorite, J. (eds.), Biología de la Conservación de plantas en Sierra Nevada. Principios y retos para su preservación. Granada: Editorial Universidad de Granada.
Pereira, H. M., Ferrier, S., Walters, M., Geller, G. N., Jongman, R. H. G., Scholes, R. J., ... & Coops, N. C. (2013). Essential biodiversity variables. Science, 339(6117), 277-278.
Petchey, O. L., & Gaston, K. J. (2006). Functional diversity: back to basics and looking forward. Ecology letters, 9(6), 741-758.
Pettorelli, N., Nagendra, H., Rocchini, D., Rowcliffe, M., Williams, R., Ahumada, J., ... & Chauvenet, A. (2017). Remote Sensing in Ecology and Conservation: three years on. Remote Sensing in Ecology and Conservation, 3(2), 53-56.
Pettorelli, N., Schulte to Bühne, H., Tulloch, A., Dubois, G., Macinnis?Ng, C., Queirós, A. M., ... & Sonnenschein, R. (2018). Satellite remote sensing of ecosystem functions: opportunities, challenges and way forward. Remote Sensing in Ecology and Conservation, 4(2), 71-93.
Steffen, W., Richardson, K., Rockström, J., Cornell, S. E., Fetzer, I., Bennett, E. M., ... & Folke, C. (2015). Planetary boundaries: Guiding human development on a changing planet. Science, 347(6223), 1259855.
Valle, F., et al. (2003). Mapa de series de vegetación de Andalucía.
Vitousek, P. M., & Hooper, D. U. (1994). Biological diversity and terrestrial ecosystem biogeochemistry. In Biodiversity and ecosystem function (pp. 3-14). Springer, Berlin, Heidelberg.
Wilson, M. V., & Shmida, A. (1984). Measuring beta diversity with presence-absence data. The Journal of Ecology, 1055-1064.
Zamora, R., Pérez-Luque, A.J., Bonet, F.J., Barea-Azcón, J.M. y Aspizua, R. (editors). (2015). La huella del cambio global en Sierra Nevada: Retos para la conservación. Consejería de Medio Ambiente y Ordenación del Territorio. Junta de Andalucía. 208 pp.
You can click here to download a preesentation of the storyline.
Last update: May, 2019