The project deals with obtaining relevant environmental information for monitoring global change in protected areas from terrestrial photography (images obtained with a conventional digital camera) coupled to conventional monitoring stations as a viable, complementary and low-cost alternative to the acquisition of remote information. This information allows a deeper understanding of ecohydrological processes in areas of environmental value and the establishment of effective adaptation actions, as well as monitoring their impact. Its evaluation was programmed in two pilot cases: Sierra Nevada National and Natural Park (PNNSN, Granada and Almería) and Sierra de Cardeña y Montoro Natural Park (PNSCM, Córdoba), representative of Mediterranean high mountain and dehesa enclaves, respectively, both with high environmental value and unique in terms of biodiversity associated with their climatic/hydrological conditions.

In the PNNSN, the presence of snow determines the spatial and temporal distribution of its ecosystems and the uniqueness of its biodiversity. In this case, there is a snow monitoring point in the surroundings of the Poqueira Refuge, at more than 2500 m a.s.l., which has a digital camera coupled to a complete meteorological station developed and operated by the Working Group of this Project (included in the REDIAM of the Consejería de Medio Ambiente de la Junta de Andalucía), but its data have only been used to obtain information on the evolution of the fraction of snow cover over time. In this work, new relevant variables for climate change impact indicators were studied at this control point and from the images collected: a) temperature at which precipitation occurs in the form of snow; b) occurrence of precipitation in the form of hail; c) occurrence of rain on snow, conditions that accelerate snow melt and are associated with torrential events; d) daily melt threshold with impact on the flow recorded downstream in the Poqueira riverbed.

In the PNSCM, the dehesa constitutes a natural ecosystem with human intervention in balance with its environmental and social values. In this case, a digital camera was installed at a control point in the Park where the IFAPA has a tower for monitoring energy flows and a meteorological station, instrumentation and available data that have been made available for this Project. The ecohydrological variables to be monitored were: a) occurrence of fog and/or dew; b) soil moisture status from colour change; c) start of runoff generation during rain showers; d) evolution of grass cover growth; e) start and end of flowering in trees; f) start and/or progress of erosive phenomena associated with rainfall and runoff.

The information obtained was contrasted with the meteorological data collected in both areas, as well as with the spectral information obtained by the sensors on board the Landsat 7 and 8 (PNNSN) and Sentinel-2 (PNSCM) satellites, in order to explore the validity of the methodology in the new variables to be analysed in this Project.

The Project also included scientific dissemination activities at academic, technical, administrative/management and social levels.

The overall aim of the project is to evaluate the potential of terrestrial photography (photography obtained with a conventional digital camera) as a standard system for environmental monitoring of ecohydrological variables for monitoring global change and adaptation actions in dehesa and high mountains, which complements and enhances the information recorded by conventional measuring stations, and makes it possible to set up quasi-real-time forecasting systems at a very low cost.

To this end, the following specific objectives are carried out:

• To obtain variables of environmental interest from digital information associated with terrestrial images with sufficient approximation for the level of monitoring and adaptation to climate change.
• To evaluate the potential of this technology in the ecohydrological monitoring of Mediterranean high mountain environments: Pilot Case 1. Sierra Nevada National and Natural Park (Granada and Almeria).
• To evaluate the potential of this technology in the ecohydrological monitoring of dehesa environments: Pilot Case 2. Sierra de Cardeña y Montoro Natural Park (Córdoba).
• Design and disseminate a control and early warning system associated with this information for the early detection of threshold states for the management of the Parks.

PILOT CASE 1: NATIONAL AND NATURAL PARK OF SIERRA NEVADA

Sierra Nevada is a mountain massif located in the southeast of the Iberian Peninsula. Parallel to the Mediterranean coast, the area is characterised by its steep altitudinal gradient. Snow appears at altitudes above 2000 m.a.s.l. from November to May, with a very variable spatial distribution from year to year. The area is a privileged enclave from the point of view of biodiversity, with a large number of endemic flora and fauna. For this reason, the highest 169,239 ha of the massif have been protected since 1989 by the legal status of Natural Park. In addition, a total of 86,208 ha were declared a National Park in 1999.

The control point used in the project is located at Refugio Poqueria, at an altitude of over 2,500 m above sea level. Since 2005 it has had a complete meteorological station developed by the River Dynamics and Hydrology group and since the summer of 2009 an automatic digital photography camera.

The use of terrestrial photography (more information http://www.uco.es/dfh/redcamaras/) coupled with the information obtained from the meteorological station, makes it possible to obtain variables of ecohydrological interest with sufficient approximation for the level of control and management. The ecohydrological variables obtained for this case study are defined below.

Temperature at which precipitation in the form of snow occurs

The objective of this indicator is to use terrestrial photography to monitor the threshold temperature at which precipitation occurs in the form of snow, also analysing the variables obtained from the meteorological station in the area. With the processed meteorological data and images, the photographs are visually analysed to identify whether it precipitates as snow or rain. In parallel, the precipitation and temperature data obtained from the weather station are analysed. Finally, the results obtained from both databases are compared and hypotheses are defined to obtain the threshold temperature. This indicator is analysed on a temporal scale: hourly (at the time the images were taken), seasonal (autumn, winter and spring) and annual.

The results show that the use of terrestrial photography has a high capacity to analyse the threshold temperature at which snow precipitation occurs. After analysis at different time scales, less error was obtained when defining a threshold temperature per season of the year and for each hydrological year of the period analysed. However, it should not be forgotten that there are limitations in establishing threshold values such as: lack of night-time images, small image extent and the need for weather stations close to the digital camera installation.

Occurrence of rain on snow (ROS).

In this case, the aim is to study the specific phenomena of rain on snow (ROS) that can be observed on a detailed scale from ground photography. For this indicator, a total of 1,800 terrestrial photographs were analysed between the hydrological years 2009 and 2016. The analysis was carried out on an annual scale in the snowy season (November-May). In addition, the behaviour in the 5 hours of the day in which the images are taken (8:00 a.m.// 10:00 a.m.// 12:00 p.m.// 14:00 p.m.// 16:00 p.m.) was analysed.

The 2012/2013 hydrological year is the one with the highest number of OSR days, with a total of 60 days. With respect to the hours analysed, 8:00 a.m. is the hour with the least number of OSR days, even in years in which no rain-on-snow events were observed, such as 2011. This fact may be due to the fact that 8:00 a.m. is the time when temperatures are lowest, so precipitation is more likely to be in the form of snow and not rain.

There is a proportional relationship between the probability of ROS and the probability of precipitation in the year in question, although this relationship is more pronounced if the events analysed occur during periods of low temperatures.

Impact of the daily melt on the flow recorded downstream in the Poqueria river channel.

This indicator aims to analyse whether there is an impact of the melt observed in the area of the terrestrial photograph on the flow at the point of closure of the basin. For this purpose, two variables are used: snow thickness, obtained by means of the milestones installed within the photographed area; flow at the point of closure, obtained by means of hydrological simulation using the WiMMed model.

The variables are analysed on a daily scale, in the period from November 2009 to August 2016 (7 hydrological years). A total of 2407 photos were analysed and processed to obtain the thickness in each of them. The hydrological years are analysed independently of each other. The figure shows the evolution of flow and thickness for the example year 2011-2012 and how a correlation of 0.6 is obtained in a lag period of 18 days.

The results concluded that there is a more significant impact of snowmelt on flow more directly in dry years than in wet years. However, it should not be forgotten that the area covered by terrestrial photography is 0.001% of the total catchment area. Therefore, a larger area analysed by photography or more monitoring sites installed would help to improve this relationship and thus be able to anticipate flow pulses due to snow melt periods.

YEAR 2011-2012

Beginning and end of the flowering period of the piorno tree

The piorno is a high mountain shrub species with a wide presence in the Sierra Nevada. They normally form in areas between 1000 and 2700 m.a.s.l. Their flowering is a function of local temperatures and snowfall, therefore the analysis of their ecohydrological response is considered an indirect way of validating the conclusions of hydroclimatic studies. The objective of this indicator was to analyse terrestrial photography as a system for monitoring the temporal evolution of phenological variables in relation to the climatic regime.

The selected images are cropped in the area to be analysed, so that the rest of the pixels do not influence the colouring of the scrubland area. The information of the pixels (RGB) is obtained from each of the cropped images, to see where the change in colouring occurs and to be able to distinguish the dates corresponding to the phenological stages.

The phenological stages are unequivocally differentiated by studying the RGB. In relation to the dates obtained, there is not enough data to be able to obtain a trend either in the delay or advance of the flowering of the piorno. Therefore, further analysis in future decades could lead to more certain patterns than those analysed in the 8 years of the study.

Conclusions regarding the methodology developed in the project:

• The results obtained confirm that the use of cameras for the study of ecohydrological states is presented as a technique that complements observations by means of meteorological stations, remote sensors and field campaigns.
• Its potential is based on obtaining high quality results at a time scale adapted to the problem in question and with a low investment in costs.

Conclusions regarding pilot case 1 (PNNSN):

• The use of digital camera coupled to meteorological station has proved to be an effective methodology in a mountain climate context influenced by Mediterranean characteristics, as is the case of Sierra Nevada.
• Temperature is a very influential variable in precipitation phenomena in the form of snow. Its threshold varies depending on the hourly and seasonal scale at which it is analysed. It varies between 0.9ºC-0.3ºC for autumn, -1.4ºC and 1ºC for winter and -1.5ºC and 0.8ºC for spring.
• The study of phenomena associated with hail is not possible to observe by terrestrial photography due to the clouding of the lens.
  Rain-on-snow events are directly related to the probability of precipitation. The probability of these types of events can vary between 20% for dry years and 30% in wet years. Their relationship with temperature is defined at temperatures below -2ºC, for which the probability of a rain-on-snow event exceeds 51%.
• Photo-observed melt has an impact on the flow at the catchment closure point 18 days after the observed melt. This is only observed in dry years. No relationship is found in wet years.
• The RGB technique is consolidated as effective for the study of phenology in the area.

PILOT CASE 2: SIERRA CARDEÑA Y MONTORO NATURAL PARK

The Sierra de Cardeña y Montoro Natural Park is made up, throughout its 38,436 ha, of dehesa, pine forest and Mediterranean woodland. This variety, together with the fact that it borders the Sierra de Andújar Natural Park, makes it a natural refuge for Iberian fauna, where both game species and protected species (lynx, wolf) coexist, along with a multitude of migratory birds, reptiles, amphibians and arthropods. In addition, the dehesa areas, whose pastures contain up to 70 different species per square metre, create a unique space in terms of biodiversity, which is why it was declared a Natural Park in 1989.

The area selected for the case study is an area of dehesa, consisting mainly of holm oaks and natural pasture. It is located at 734 m.a.s.l. within the Martín Gonzalo stream basin and has since 2012 with various instrumentation (implemented by the Institute for Agricultural and Fisheries Research and Training of Andalusia, IFAPA) and since December 2017 within the framework of this project, with a digital photography camera (more information http://www.uco.es/dfh/redcamaras/).

The ecohydrological variables selected to evaluate the potential of terrestrial photography as a methodology for monitoring in dehesa environments are defined below.

Occurrence of fog

In order to estimate a threshold value for the occurrence of fog, the days of occurrence of this phenomenon were visually analysed by means of photography and the meteorological variable data were studied and found to be related to relative humidity (RH). On all the days when fog occurred, the RH was higher than 54%, so this is considered to be the RH threshold for fog to occur.

Soil moisture status from vegetation colour change

Soil moisture is directly related to the state of the vegetation and this relationship depends on the type of vegetation, with slower-growing canopies (trees) being more resistant to change than other faster-growing ones (scrub or grass) (Gómez-Giráldez et al., 2014). For this reason, and taking into account the study period used, the vegetation chosen as an indicator was grass cover. The following figure clearly shows the change in grass colour. However, it is difficult to distinguish this change on consecutive days, so it is necessary to quantify the degree of greenness or vigour. The Green Chromatic Coordinate (GCC), a non-linear transformation of the digital levels of the green colour of the photo, was used to appreciate this change on consecutive days. This index was calculated as the mean value of the pixels of a homogeneous region of interest (ROI) of grass.

The figure shows the results obtained for GCC (top) and mean daily rainfall and temperature (bottom) in the study period. The dependence of the grass condition on precipitation and temperature can be seen, with the maximum value of GCC occurring after the highest concentration of rainfall (March) and decreasing from this maximum value as the average temperature increases. As a threshold value, the average temperature of 15 degrees is considered, after which the GCC value starts to decrease more drastically.

Growth of grass cover

To estimate the evolution of the height of the grass cover, a 1 metre high pole was placed in the chamber area. A daily photograph was analysed and the height of the grass was estimated. The days when the grass stops growing in height (early May) coincide with the time when grass vigour starts to decline, with average temperatures exceeding 15°C and rainfall has started to cease.

Beginning and end of flowering of trees

For the determination of flowering by photography, four different stages have been separated:

• Initial stage (1)
• Initial flowering stage (2i)
• Maximum flowering stage (2m)
• Withering stage (3)

The determination was carried out visually, choosing the corresponding photos at midday and on days without rain in order to have the best possible view of the tree canopy.

The dates obtained for each of the stages are also related to rainfall and temperatures, with stage 2i occurring at the end of the heaviest rains and stage 2m just before the beginning of wilting coinciding with average daily temperatures above 15 ºC.

Conclusions regarding the methodology developed in the project:

• The results obtained confirm that the use of cameras for the study of ecohydrological states is presented as a technique that complements observations by means of meteorological stations, remote sensors and field campaigns.
• Its potential is based on obtaining high quality results at a time scale adapted to the problem in question and with a low investment in costs.

Conclusions regarding pilot case 2 (PNSCM):

The digital terrestrial camera has been used to identify ecohydrological variables in the dehesa ecosystem:

• 38 days with the presence of fog have been observed, in that the daily relative humidity threshold was 54%.
• The Green Chromatic Coordinate (GCC) index was correctly estimated from digital photography to quantify grass colour and vigour. This index showed a high correlation (0.75) with the NDVI vegetation index obtained from Sentinel-2 satellite images. The GCC values show an increase with rainfall and a strong decrease coinciding with mean daily temperatures above 15ºC.
• The height of the grass was estimated from digital photography. The maximum height obtained in the area was 45 cm and its growth ceases when the average daily temperatures start to be higher than 15ºC.
• It was possible to observe the different stages of flowering of the holm oak from terrestrial photography. Flowering begins on 8 April 2018, peaks on 5 May 2018 and withers on 20 May 2018.
• Due to the characteristics of the terrain (low slope and high vegetation cover), it was not possible to distinguish runoff phenomena and, therefore, neither erosion.

The team involved in the project is part of the Research groups at the Andalusian Institute for Earth System Research (IISTA), seat of University of Córdoba.

 

María José Polo Gómez – Responsible Researcher

María Cristina Aguilar Porro – Researcher

Carmen Galán Soldevilla – Researcher

José Luis Quero Pérez – Researcher

María José Pérez Palazón – Researcher

Alicia Jurado López – Researcher

Different forms of dissemination have been carried out in the social media throughout the project. Thus, the results have been exposed in the different media in which there has been availability of action. The most important ones:

 

SEMINARS Y WORKSHOPS

• I Congreso Internacional de Montañas – CIMAS celebrado el 5 y 11 de marzo en Granada. By Invitation Fundación Biodiversidad.

• Defence of Final Master’s Degree Projects in Environmental Hydraulics 2017-2018. Three students of the master’s degree have developed their Final Projects within the framework of the project. Defence held on 21 September 2018 in Cordoba.

FAIRS AND DISSEMINATION CONFERENCES

• Ingenios en Ruta 2018. The Scientific Culture Unit of the UCO designs this programme in which scientific researchers from the institution itself visit primary and secondary schools and institutes. The Group participated in the 2018 edition with its workshop «Water in a changing environment» with a visit to three centres located in Belalcázar, Pedro Abad and Villanueva de Córdoba, all in the province of Córdoba on 6, 13 and 16 March of this year. The heads of the Scientific Culture Unit of the UCO, recorded one of the sessions, which can be followed at (https://www.youtube.com/watch?v=-rpL5C8YfUc)

• Researcher’s Night –  “Ingenios en Ruta”.Within the European Marie Sklodowska Curie programme, the Descubre Foundation programmes this activity in the same format as the one described above. In this case, students of 2nd EPO and 1st and 2nd Bachillerato from three schools in the province were visited in September 2017 by members of the DFH with the workshop «the way of water» in order to explain the scientific work of the group and the main advances obtained in the projects

• Researcher’s Night “Feria de los Ingenios”. The European Marie Sklodowska Curie Programme is planning to carry out this activity in 350 different cities across Europe. The aim is to bring science closer to the general public and thus have a wider media reach. On the evening of 29 September 2017, the DFH showed the public the monitoring network belonging to the group, what it is and how it works.

SOCIAL NETWORKS

Twitter @dfh_uco

Facebook @gdfhuco

Through the group’s profile, all the news related to the project and the data used for its methodology are disseminated. The profile has a total of 136 followers and an average reach of 185 people per publication.

CONGRESS CONTRIBUTIONS

• Torralbo, P., Herrero, J., Gilabert, A., Gallego-Galán, C., Pérez-Palazón, M.J., Quero, J.L., Galán, C. y Polo, M.J. 2018.Terrestrial photography for monitoring eco-hydrological variables inMediterranean mountain areas. EuropeanGeosciencesUnion General Assembly 2018, Viena, Austria. Fecha de celebración: 10/04/2017
• Pérez-Palazón, M.J, Pimentel, R. y Polo, M.J. Torrentiality and aridity in Sierra Nevada (Spain): observed trends andprojected climate scenarios. EuropeanGeosciencesUnion General Assembly 2018, Viena, Austria. Fecha de celebración: 10/04/2017 
• Pérez-Palazón, M.J, Gilabert, A, Barberá, J., Torralbo, P. y Polo M.J. Coupling time-lapse photography to weather stations for monitoring snow processes in Mediterranean mountain regions: a case study in SierraNevada (Spain).Remote Sensing and Hydrology Symposium (RSHS18). Córdoba, España. Fecha de celebración: 10/05/2018

• Jurado López, A. y Polo, M.J.Clima e hidrología de montaña. Adaptación del lenguaje a los destinatarios. VI Congreso de comunicación social de la ciencia. Córdoba, España. Fecha de celebración 24/11/2017

 

DIGITAL CAMERA STATION TRACKING DESIGN

Within the framework of the project, the creation of a website is carried out. (http://www.uco.es/dfh/redcamaras/) which hosts the monitoring network managed and studied by the DFH. It serves as a verification and control of the observed states. The website has an interactive map and includes information on the entire network of cameras and real-time images of the Refugio. Poqueira.