**5. Conclusions**

366 International Perspectives on Global Environmental Change

regions of the world including the Canadian Boreal forest (Alcaraz-Segura et al., 2010a) and South America (Baldi et al., 2008), the NDVI increases observed in Sierra Nevada with GIMMS during the 1980's and 1990's agreed with other independent datasets. Alcaraz-Segura et al. (2010b) showed that the positive NDVI trends that Sierra Nevada displayed in previous studies with the GIMMS dataset were observed for the 1981-1999 period using other independent datasets such as PAL (Pathfinder AVHRR Land), FASIR (Fourier-Adjustment, Solar zenith angle corrected, Interpolated Reconstructed), and LTDR (Land Long-Term Data Record) datasets. Positive NDVI trends were also observed in Sierra Nevada during the 1989-2002 period using the MEDOKADS (Mediterranean Extended

The EVI decrease observed at the beginning of the growing season during the 2000-2009 period in Sierra Nevada oak woodlands (Figures 3 and 4), is also in contrast with the NDVI seasonal increase in autumn, winter, and spring that was reported for the 1982-2006 period using GIMMS images of the entire Park (see Figure 2 in: Alcaraz-Segura et al., 2008a). Such contrasting trends lead to think that the increase of spring vegetation greenness that occurred throughout de eighties and nineties (Alcaraz-Segura et al., 2008a) ended around the year 2000 when the spring started to return to lower greenness values. Yet, the trends towards greater vegetation greenness in autumn and winter reached during the eighties and nineties (Alcaraz-Segura et al., 2008a) was maintained after the year 2000, since we did not find significant EVI trends in these seasons. The strong EVI decreases at the beginning of the growing season and the presence of some EVI summer increases during the senescence period lead to think that the growing season of southern oak woods (Figure 4) might be starting later but strengthening

An important outcome of our work is that significant functional changes, i.e. a significant decrease of vegetation greenness at the beginning of the growing season, took place in Sierra Nevada oak woodlands without implying significant trends in the annual averages. Despite the EVI annual mean, an estimator of annual primary production, is extensively used as an integrative descriptor of ecosystem functioning and status, our work highlights the importance of studying variables beyond the annual summaries (like seasonality and phenology) as significant trends in particular months of the year may not significantly affect the EVI annual mean but may have broad ecological consequences in critical periods such as

Since satellite images are regularly captured over large regions and under common protocols, the spectral vegetation indices represent an adequate approach to implement ecosystems monitoring programs in protected areas and to promote adaptive management actions (Alcaraz-Segura et al., 2008a; Alcaraz-Segura et al., 2008b; Cabello et al., 2008). Our work provides interesting information for the prioritization and the orientation of management actions for the Pyrenean oak forests of Sierra Nevada National Park. First, we provided a regional functional reference characterization of all oak woodlands of the Park for the 2001-2009 period. Our monitoring approach uses EVI-derived descriptors of ecosystem functioning that may allow managers to detect the spatial and temporal anomalies (Oyonarte et al., 2010), and to guide specific management actions in particular areas. The spatial and temporal deviations from the baseline conditions detected could be alerting of inconspicuous "within-state" changes in the forests as a result of cumulative impacts (Vogelmann et al., 2009). However, to improve the ecological significance of this

Daily One-km AVHRR Data Set) archive (Martínez & Gilabert, 2009).

towards the summer (with the exception of Poqueira; Figure 4c).

**4.2 Application to forest monitoring and management** 

the start of the growing season.

Our approach shows how satellite based monitoring systems can be very useful to assess the effects of environmental changes on protected areas and to orientate adaptive management actions. Overall, this study provides a reference characterization against which to assess changes in ecosystem functioning of the oak woods of Sierra Nevada, and identifies functional changes that occurred during the 2001-2009 period. Such information helps to fill the lack of knowledge about these woodlands, as demanded by the Spanish Ministry of Environment (García & Mejías, 2009). In practical terms, it allows the incorporation of ecosystem functional aspects of ecosystems to nature conservation and to the maintenance of ecosystem services, in particular those related to carbon sequestration in this protected area. Our results imply that conservation and management policies cannot be only based on static situations, since ecosystems are changing. In addition, annual summaries are not enough as monitoring indicators, since functional changes may occur at key seasonal stages without affecting the annual means.

Satellite-Based Monitoring of Ecosystem Functioning in Protected Areas:

37°10'0"N

**Legend DIA MAX Dia\_Max** <Null> 1 ene 17 ene 2 feb 18 feb 6 mar 22 mar 7 abr 23 abr 9 may 25 may

1 ene 17 ene 2 feb 18 feb 6 mar 22 mar 7 abr 23 abr 9 may 25 may

1 Jan 17 Jan 2 Feb 18 Feb 6 Mar22 Mar7 Apr23 Apr9 May25 May10 jun

10 jun 26 jun 12 jul 28 jul 13 ago 29 ago 14 sep 30 sep 16 oct 1 nov 17 nov 3 dic 19 dic

26 jun 12 jul 28 jul 13 ago 29 ago 14 sep 30 sep 16 oct 1 nov 17 nov 3 dic 19 dic

10 Jun26 Jun12 Jul 28 Jul 13 Aug29 Aug14 Sep 30 Sep 16 Oct1 Nov17 Nov3 Dec 19 Dec **a) DMAX**

37°5'0"N

37°0'0"N

37°10'0"N

**Legend Tendencia EVI Tendencia** -0.009 - -0.005 -0.005 - 0 0 0 - 0.005 0.005 - 0.009


37°5'0"N

37°0'0"N

3°25'0"W 3°20'0"W 3°15'0"W

**c) Sen's slope of the 2001-2009 EVI\_mean trend**

3°25'0"W 3°20'0"W 3°15'0"W

trend d) Mann-Kendall p-value of the 2001-2009 EVI\_mean trend.

´ <sup>036</sup> 1.5 km

´ <sup>036</sup> 1.5 km

Recent Trends in the Oak Forests (*Quercus pyrenaica* Willd.) of Sierra Nevada (Spain) 369

37°10'0"N

**Legend DIA MIN Dia\_Max** <Null> 1 ene 17 ene 2 feb 18 feb 6 mar 22 mar 7 abr 23 abr 9 may 25 may

1 ene 17 ene 2 feb 18 feb 6 mar 22 mar 7 abr 23 abr 9 may 25 may

1 Jan17 Jan 2 Feb 18 Feb 6 Mar22 Mar7 Apr 23 Apr9 May25 May10 jun

10 jun 26 jun 12 jul 28 jul 13 ago 29 ago 14 sep 30 sep 16 oct 1 nov 17 nov 3 dic 19 dic

26 jun 12 jul 28 jul 13 ago 29 ago 14 sep 30 sep 16 oct 1 nov 17 nov 3 dic 19 dic

10 Jun26 Jun12 Jul28 Jul13 Aug29 Aug14 Sep

30 Sep16 Oct1 Nov 17 Nov3 Dec 19 Dec

**b) DMIN**

37°5'0"N

37°0'0"N

37°10'0"N

**Legend ROBLEDALES p\_value** 0.012500 - 0.050000 0.050001 - 0.100000 0.100001 - 0.150000 0.150001 - 0.300000 0.300001 - 0.991700

37°5'0"N

37°0'0"N

Fig. 7. Maps of the EVI attributes and trends for Sierra Nevada Oak woods generated by the Monparq application. a) DMAX and b) DMIN: Dates when the Maximum and Minimum EVI values are reached, indicators of phenology. c) Sen's slope of the 2001-2009 EVI\_mean

3°25'0"W 3°20'0"W 3°15'0"W

**d) Mann-Kendall p-value of the EVI\_mean trend**

0.012500 - 0.050000 0.050001 - 0.100000 0.100001 - 0.150000 0.150001 - 0.300000 0.300001 - 0.991700

3°25'0"W 3°20'0"W 3°15'0"W

´ <sup>036</sup> 1.5 km

´ <sup>036</sup> 1.5 km

Fig. 6. Maps of the EVI attributes for Sierra Nevada Oak woods generated by the Monparq application. EVI\_mean: EVI annual mean, an estimator of annual primary production; EVI\_sCV: EVI seasonal Coefficient of Variation, a descriptor of seasonality; MAX and MIN: Maximum and Minimum EVI annual values, indicators of the maximum and minimum photosynthetic activity.

37°10'0"N

**Legend Coeficiente de Variación CV** 0.068700 - 0.102100 0.102101 - 0.131200 0.131201 - 0.156100 0.156101 - 0.181400 0.181401 - 0.214200 0.214201 - 0.250500 0.250501 - 0.286900 0.286901 - 0.331600 0.331601 - 0.384900 0.384901 - 0.451500 0.451501 - 0.547600

0.068700 - 0.102100 0.102101 - 0.131200 0.131201 - 0.156100 0.156101 - 0.181400 0.181401 - 0.214200 0.214201 - 0.250500 0.250501 - 0.286900 0.286901 - 0.331600 0.331601 - 0.384900 0.384901 - 0.451500 0.451501 - 0.547600 **b) EVI\_sCV**

37°5'0"N

37°0'0"N

37°10'0"N

37°5'0"N

37°0'0"N

Fig. 6. Maps of the EVI attributes for Sierra Nevada Oak woods generated by the Monparq application. EVI\_mean: EVI annual mean, an estimator of annual primary production; EVI\_sCV: EVI seasonal Coefficient of Variation, a descriptor of seasonality; MAX and MIN: Maximum and Minimum EVI annual values, indicators of the maximum and minimum

3°25'0"W 3°20'0"W 3°15'0"W

**d) MIN**

3°25'0"W 3°20'0"W 3°15'0"W

´ <sup>036</sup> 1.5 km

´ <sup>036</sup> 1.5 km

3°25'0"W 3°20'0"W 3°15'0"W

**c) MAX**

3°25'0"W 3°20'0"W 3°15'0"W

´ <sup>036</sup> 1.5 km

´ <sup>036</sup> 1.5 km

37°10'0"N

**a) EVI\_mean**

37°5'0"N

37°0'0"N

37°10'0"N

37°5'0"N

37°0'0"N

photosynthetic activity.

Fig. 7. Maps of the EVI attributes and trends for Sierra Nevada Oak woods generated by the Monparq application. a) DMAX and b) DMIN: Dates when the Maximum and Minimum EVI values are reached, indicators of phenology. c) Sen's slope of the 2001-2009 EVI\_mean trend d) Mann-Kendall p-value of the 2001-2009 EVI\_mean trend.

Satellite-Based Monitoring of Ecosystem Functioning in Protected Areas:

images were freely provided by the MODIS Land website.

**7. References** 

1697-2473

pp. 13-29, ISSN 1385-0237

770, ISSN 1365-2486

8220

1009 (Electronic), 0364-152X (Linking)

No.2, pp. 200-212, ISSN 1466-8238

Esfagnos, ISBN 978-84-932095-1-3,

*Sensors*, Vol.10, No.2, pp. 1291-1314, ISSN 1424-8220

structure. *Bird Study*, Vol.53, No.1, pp. 12-19, ISSN 0006-3657

Recent Trends in the Oak Forests (*Quercus pyrenaica* Willd.) of Sierra Nevada (Spain) 371

de Ciencia e Innovación (Proyecto CGL2010-22314, subprograma BOS, Plan Nacional I+D+I 2010). D. Alcaraz-Segura was partially covered by the Inter-American Institute for Global Change Research (IAI, CRN II 2031 and 2094) under the US National Science Foundation (Grant GEO-0452325). The trend test was run using the MATLAB code "Seasonal Kendall Test with Slope for Serial Dependent Data" provided by Jeff Burkey through the MATLAB Central file exchange (http://www.mathworks.com, accessed September 2009). Satellite

Alcaraz-Segura, D.; Baldi, G.; Durante, P. & Garbulsky, M. (2008a). Análisis de la dinámica

Alcaraz-Segura, D.; Cabello, J.; Paruelo, J.M. & Delibes, M. (2008b). Trends in the surface

Alcaraz-Segura, D.; Cabello, J. & Paruelo, J. (2009a). Baseline characterization of major

Alcaraz-Segura, D.; Cabello, J.; Paruelo, J.M. & Delibes, M. (2009b). Use of descriptors of

Alcaraz-Segura, D.; Chuvieco, E.; Epstein, H.E.; Kasischke, E.S. & Trishchenko, A. (2010a).

Alcaraz-Segura, D.; Liras, E.; Tabik, S.; Paruelo, J. & Cabello, J. (2010b). Evaluating the

Alcaraz-Segura, D.; Paruelo, J. & Cabello, J. (2006). Identification of current ecosystem

Arriero, E.; Sanz, J.J. & Romero-Pujante, M. (2006). Habitat structure in Mediterranean

Baldi, G.; Nosetto, M.D.; Aragón, M.R.; Aversa, F.; Paruelo, J.M. & Jobbagy, E.G. (2008).

Blanca, G. (2001). Flora, In: *Parque Nacional de Sierra Nevada*, Calvo-Flores D, (Ed.), 106-148,

*Applied Vegetation Science*, Vol.11, pp. 431-440, ISSN 1402-2001

temporal del NDVI en áreas protegidas: tres casos de estudio a distintas escalas espaciales, temporales y de gestión. *Ecosistemas*, Vol.17, No.3, pp. 108-117, ISSN

vegetation dynamics of the National Parks of Spain as observed by satellite sensors.

Iberian vegetation types based on the NDVI dynamics. *Plant Ecology*, Vol.202, No.1,

ecosystem functioning for monitoring a national park network: A remote sensing approach. *Environmental Management*, Vol.43, No.1, January, pp. 38-48, ISSN 1432-

Debating the greening vs. browning of the North American boreal forest: differences between satellite datasets. *Global Change Biology*, Vol.16, No.2, pp. 760-

Consistency of the 1982–1999 NDVI Trends in the Iberian Peninsula across Four Time-series Derived from the AVHRR Sensor: LTDR, GIMMS, FASIR, and PAL-II.

functional types in the Iberian Peninsula. *Global Ecology and Biogeography*, Vol.15,

deciduous oak forests in relation to reproductive success in the Blue Tit Parus caeruleus: Capsule Effects operate during laying and incubation and with less success in breeding territories characterized by a young and immature vegetation

Long-term satellite NDVI datasets: Evaluating their ability to detect ecosystem functional changes in South America. *Sensors*, Vol.8, pp. 5397-5425, ISSN 1424-

Fig. 8. Landscape picture showing the start of the growing season (13th April 2011) in the northernmost *Quercus pyrenaica* oak wood of Sierra Nevada National Park (Spain), the oak wood of the Alhama River at Dehesa del Camarate. The picture shows how the green sprouts of the oak trees are starting to come out while the leaves of the undergrowth shrubs are well developed.

To spread the use of our monitoring approach and to make possible for managers the exploitation of such information, we have developed a software tool named "Monparq Monitoring System for Parks" that allows a non-advance user to assess the differences between locations, to explore the different environmental controls across the northern and southern slopes, and to evaluate the inter-annual trends in ecosystem functioning. This tool provides managers with valuable information to assess management effectiveness in an adaptive management strategy. It will help managers answering questions like, what ecosystems are undergoing major changes?, or how do management actions affect ecosystem functioning stability?

## **6. Acknowledgments**

Thanks to L. Sevilla, who helped processing the datasets, to F.J. Bonet and B. Benito from the Sierra Nevada Global Change Observatory for providing climate data, to the Park managers and J. del Río from the Andalusian Environmental Agency for their guidance and valuable information in the field, and to F. Maestre for the English revision. Financial support was given by FEDER Funds, Junta de Andalucía (GLOCHARID and SEGALERT P09–RNM-5048 projects), Organismo Autónomo de Parques Nacionales (Proyecto 066/2007), and Ministerio de Ciencia e Innovación (Proyecto CGL2010-22314, subprograma BOS, Plan Nacional I+D+I 2010). D. Alcaraz-Segura was partially covered by the Inter-American Institute for Global Change Research (IAI, CRN II 2031 and 2094) under the US National Science Foundation (Grant GEO-0452325). The trend test was run using the MATLAB code "Seasonal Kendall Test with Slope for Serial Dependent Data" provided by Jeff Burkey through the MATLAB Central file exchange (http://www.mathworks.com, accessed September 2009). Satellite images were freely provided by the MODIS Land website.
