OPERATING INSTITUTE: BGU.
MAIN PURPOSE: Hydro-geo-ecology, ecosystem services, biodiversity, remote sensing, afforestation.
ECOSYSTEM TYPE: Semiarid forest, shrubland, arable land.
EXPERIMENTAL TREATMENTS: Soil management, irrigation, fertilisation, grazing.
LOCALISATION: 31.196629726829233 34.90997314453125
FACILITIES: The Northern Negev LTER network in addition to 4 field sites is composed of the following facilities: (1) A hydrological laboratory that provides permanent and portable rain simulators, laser particle size analysers and hydrological instrumentation that monitor in situ rainfall, runoff, soil moisture, and sedimentation; (2) A field laboratory that enables investigation of the tradeoffs among water use, carbon sequestration, energy fluxes, and radiation budgets. It also includes a sonic anemometer, as well as instruments for measuring meteorological and soil conditions. The field lab is connected with a stable isotope lab that includes a Gas Source – Isotope Ratio Mass Spectrometer; (3) A remote sensing lab that provides image processing facilities and hyperspectral spectrometers.
Currrently, there are two prevailing methods of afforestation in the semi-arid areas of the northern Negev Desert: (1) Rain Fed Afforestation (RFA) where direct rainfall is the main water resource and planted trees replace the natural growing shrubs; (2) Runoff Harvesting Afforestation (RHA) that spatially integrates natural and human-made systems in order to harvest runoff water from the natural system and concentrate it into water-enriched patches where trees are planted. The study aims to compare the two afforestation systems and also between them and the adjacent natural areas.
● Cierniewski, J., Karnieli, A., Kuśnierek, K., Goldberg, A., and Herrmann, I. 2013. Approximating the average daily surface albedo with respect to soil roughness and latitude. International Journal of Remote Sensing.
● Maestre T. F., Quero, L. J., Gotelli, J. N., (...) Wang, D., Zaady, E. 2012. Plant Species Richness and Ecosystem Multifunctionality in Global Drylands. Science 335: 214–218.
● Sher, Y., Zaady, E., Ronen, Z., Nejidat, A. 2012. Nitrate accumulation in soils of a semi-arid ecosystem following a drought-induced shrub death. European Journal of Soil Biology 53: 86–93.
TA PROJECTS: Land use impact on soil quality using soil arthropods in the Negev (QBSN)
TA User (visit): Cristina Menta, Dept. of Life Sciences, Parma University, Italy (March, 2015 – 06 days)
Project Description: We propose an efficient and low-cost biological index of soil quality: QBS-ar index (acronym of soil biological quality, in Italian “Qualità Biologica del Suolo”). This index is a biological method that joins the biodiversity of soil microarthropods community with the degree of soil animal vulnerability. It is based on the following concept: the higher is the soil quality, the higher is the number of microarthropod groups morphologically well adapted to soil. This protocol, through the study of the soil microarthropods, provides information on the soil biological quality, which is an indicator of land degradation. QBS-ar has been developed by an Italian team (Parisi et al, 2005) more than ten years ago, and it has been applied by Italian and European specialists in several ecosystems, agricultural lands, grasslands, urban soils, woods at different level of naturality, and degraded soils (Galli et al., 2014; Hartley et al. 2008; Menta et al., 2008, 2011, 2014). It has been adopted by WP2 of ExpeER as the protocol to study soil fauna.
The study is intended for the assessment of the soil mesofauna present in a studied site using a simple index which doesn’t involve species identification skills. Soil samples will be collected in 5 sites of the Negev LTER, soil microarthropods will be extracted by a dynamic extraction method (Berlese-Tüllgren funnel) in laboratory and successively identified at the level of Recognizable Taxonomic Unit (RTU). Each RTU receives a score ranging from 1 to 20, according to its adaptation degree to soil, following a score grid. The final index sums up these scores to the QBS-ar index of soil health. This study will compare 5 sites of the Negev LTER and evaluate the effects of the land use on the soil quality. We will also compare these results to the scores obtained in several sites already studied of the colder Europe. We will also develop a modified version of the QBS-ar index that include the abundance data, similarly to Rubenstein (2012) weighted index.
Mapping forest background reflectance in an arid region using multi-angle remote sensing data
TA User (visit): Jan Pisek, Tartu Observatory, Toravere, Estonia (March, 2015 – 05 days)
Project Description: Since ground layer (understory or background) has an essential contribution to the whole-stand reflectance signal in many forests, its reflectance spectra are urgently needed in various forest reflectance modelling efforts. However, systematic reflectance data covering different site types are almost missing. Measurement of background reflectance is a real challenge because of extremely high variability of irradiance at the forest floor, weak signal in some parts of the spectrum and its variable nature. Forest background might consist of several sub-layers (tree regeneration, shrub, grasses or dwarf shrub, mosses or lichens, litter, bare soil), it has spatially-temporally variable species composition and ground coverage. Additional problems are introduced by patchiness of ground vegetation, ground surface roughness and understory-overstory relations. Due to this variability, remote sensing might be the only technology to provide consistent data at the required spatially extensive scales. The objective of this project is to collect in situ measured background reflectance data in Yatir Forest, Israel. The obtained in situ measurements will be used for the validation and comparison with the background reflectance retrievals over the same area (Yatir Forest) using MODIS BRDF data and methodology outlined in Pisek et al. (2012, 2015). The methodology was originally developed for the forest background signal retrieval in a boreal region. Here its performance will be tested in arid forest conditions, which is a necessary step before conducting global-scale mapping over forested areas. The results can be also used as an input for improved modeling of local carbon and energy fluxes.
Prior Spectral Knowledge for Ecosystem Service Monitoring and Understanding (PriorSpecK4ECOS)
TA User (visit): Andrea Baraldi , Università degli Studi di Napoli Federico II, Italy (February / March, 2015 – 90 days)
Project Description: Although rarely acknowledged by the remote sensing (RS) community, prior knowledge-based preliminary classification (pre-classification) of multi-spectral (MS) imagery, suitable for driven-by-knowledge continuous color space discretization (quantization, partitioning), has a long history. Equivalent to color naming in a natural language, it is the deductive counterpart of popular inductive data learning algorithms for vector data quantization, like the k-means algorithm. In recent years, the presentation of the prior knowledge-based Satellite Image Automatic Mapper (SIAM) and RGB-SIAM color quantizers in operating mode has brought new developments in the design and implementation of hybrid inference-based (combined deductive/top-down and inductive/bottom-up) remote sensing (RS) image understanding systems (RS-IUSs). The SIAM/ RGB-SIAM expert systems are capable of generating, alternately automatically and in near real-time, multi-level pre-classification maps and multi-scale segmentation maps of a multi-source (e.g., spaceborne/airborne) multi-resolution MS image, either radiometrically calibrated (into top-of-atmosphere reflectance, surface reflectance or surface albedo values) or radiometrically uncalibrated, respectively. The goal of the present research and technological development project proposal is to employ the novel SIAM/RGB-SIAM pre-attentive vision technology to improve, in terms of quantitative metrological/ statistically-based quality indexes of operativeness (QIOs, encompassing degree of automation, accuracy, efficiency, robustness, scalability, timeliness and costs), any pre-existing ecological processing model whose input land surface variables, either categorical (e.g., land cover (LC) classes) or continuous (e.g., leaf area index), are estimated from RS imagery. To reach this overarching goal, specific research topics are gathered into three main categories (I) to (III), namely: (I) Sensory data. Automatic driven-by-knowledge stratified data pre-preprocessing, e.g., MS image mosaicking, co-registration, compositing, topographic correction, surface albedo estimation, etc. (II) Categorical variables. Automatic multi-sensor multi-resolution LC/LC change (LCC) detection in single-date/multi-temporal MS/hyperspectral imagery, e.g., rain fed afforestation, runoff harvesting afforestation, managed/unmanaged grazing, etc. (III) Continuous variables. Spectral prior knowledge-based stratification of statistical/semi-empirical ecosystem processing models, e.g., development and validation of innovative all-channel combined categorical/continuous spectral variables (e.g., greenness index).
Ecosystem service and dis-service analysis of rain fed and runoff harvesting afforestation (ECOFOR).
TA User (visit): Jan Dick, Centre for Ecology & Hydrology (CEH), Edinburgh, UK. (May, 2014 - 8 days)
Project Description: The utility of ecosystem services as a concept for landscape planning is well established. However there is a need for research on indicators to assess the services both positive and negative (so called dis-services). A range of indicators have been proposed e.g. MA, TEEB and CICIES. These are high level classifications which provide an over view of the type of services which should be included and differ primarily on their treatment of the biological compenent of the service (biodiversity, primary productivity etc).
In this study we will directly compare two published lists one based on the MA and the other TEEB in a place-based assessment.
In this project we will work closely with in-country collaborators to compare two lists of ecosystem services (modified -Dick et al 2011 and Kandziora 2013) for two specific land uses in the Negev ExpeER site, Israel. The ecosystem services and dis-services of the rain fed afforestation and runoff harvesting afforestation areas of the site will be analysed. This work builds on the detailed measurements of ecosystem function collected by the site which are necessary to be able to utilise the Kandziora list of ecosystem service indicators. Additional services and dis-services will be added as appropriate.
Effect of the Clean Development Mechanism Scheme on Dryland Afforestation and Farming (ECDMSDAF).
TA User (visit): Henri Rueff, School of Geography, Oxford University, UK (October, 2013 – 10 days).
Project Description: It is widely acknowledged that tree plantations in deserts have potential for becoming important carbon sinks (Sedjo 1999, Gruenzweig et al. 2003, Lal 2004, Olschewski 2005, Ornstein 2009, UNCCD 2005). With a CO2-enriching atmosphere, dryland trees improve their water-use-efficiency and hence become more resistant to drought and can be planted deeper into deserts. The UNFCCC reported that the Clean Development Mechanism (CDM) represented, as of June 2012, a $215.4 billion total investment while the Kyoto protocol had been extended until 2020. Rainfed agriculture at the edge of deserts faces high yield uncertainties. Desert farmers may benefit from renouncing wheat and plant rainfed trees instead to cushion risk and protect land from degradation, while securing a steady income from trading emission reductions through the CDM. It is currently unknown whether desert farmers in non-annex I countries may actually benefit from foregoing wheat for trees while capturing the high yield uncertainty of the former, characterizing desert agriculture. This project aims at answering this question by taking advantage of the knowledge acquired by scientists and foresters working in the Yatir forest, which is part of the Negev LTER. Yatir forest is unique in that it was planted in the early 1960s in the driest possible conditions for a large-scale afforestation project. Scientists and foresters working there have accumulated substantial experience in desert forest management and data. Nearby experimental farms have also measured wheat yields under various climatic conditions and can make this data available. The project shall compare, on an infinite time horizon, the net present value (NPV) of engaging in a small-scale plantation of Aleppo Pine seeking a reward from emission reductions and forest thinning by products, to the NPV of a wheat farm, which yields are predicted on an iterative simulation to capture weather uncertainties. Results shall have scientific relevance by providing a more accurate understanding of the worth of tree planting in desert under the CDM scheme when erratic weather and yields are fully integrated in the predictions. Findings will also have important policy relevance in understanding the minimum price of emission reductions needed for farmers to engage in the plantation of trees, and identify how transaction costs can be reduced. To carry out this research, Dr Henri Rueff shall visit Dr Moshe Schwartz at the Ben Gurion University of the Negev during 10 days, to acquire and analyse remote sensing data, interview foresters and scientists active in Yatir on the forest management, acquire forest yield data, acquire agricultural data, and interview dryland farming scientists and farmers.
Surface fluxes in Semi AriD Environment (SSADE).
TA User (visit): Eric Ceschia, CESBIO, France (June, 2013 – 12 days).
Project Description: The Center for the Study of the Biosphere from Space (CESBIO) is a laboratory that aims to develop knowledge on continental biosphere dynamics and functioning at various temporal and spatial scales. It is recognised worldwide for his expertise in remote sensing and surface flux measurements. CESBIO has recently developed new methods for 1) calculating GHGbudgets for crops by combining in situ GHG flux measurements and a Life Cycle Analysis approach and 2) for partitioning Evapotranspiration measured on the field into its components Transpiration and Evaporation. CESBIO is also developing models for estimating regional water and C fluxes by combining the modeling approach with remote sensing data.The Remote Sensing Laboratory (RSL) is currently part of the Department of Solar Energy and Environmental Physics, Jacob Blaustein Institute for Desert Research, Ben-Gurion University of the Negev. The laboratory aims at two main goals: 1) Advance scientific theories, methodologies, and applications in remote sensing, image processing, and geographic information system (GIS) management of the earth's resources. 2) Generate a scientific environment leading to the development of scientific exchange beneficial to the various disciplines of the institute. Satellite, aerial imagery, and ground data are applied to environmental problems with special emphasis on arid and semi-arid regions. This laboratory is in charge of the Negev Highly Instrumented Experimental (HIES) and Observational Site (HIOS). Our two laboratories already have strong collaboration in the field of earth observation by means of remote sensing since Gérard Dedieu (CESBIO) and Arnon Karnieli (Ben Gurion University) are co-Principal Investigators of the Venµs mission. The aim of this demand is to develop further some collabrorations and exchange of expertise between the two laboratories. In particular, we aim at increasing our collaboration concerning the experimental activities. As the CESBIO experimental sites are part of the ICOS and GHG Europe networks, our first goal is to characterize the Negev site and help the Israeli team to match the standards of the ICOS sites. Also those two sites will be used for validation of the Venµs remote sensing products and standardized equipment would help to validate the remote sensing products in contrasted climates. In a second time, we will help the Israeli team to adapt their methods for processing micometeorological data to the new ICOS standards and for partitioning the ETR fluxes as well as calculating the GHG budgets at their sites. Experimental data and information concerning field operations at the crop sites will be collected in order to finalise the calculations Those results could then be integrated in a synthesis paper concerning the GHG budgets for crops within the frame of the GHGEurope project. This paper should include the effect of albedo changes on the radiatife forcing of the crop sites. Discussion will be held concerning this methodology. Finally, in the perspective of the Sentinel 2 and Venµs satellite missions (launched in 2014) we will discuss and plan a common approach for estimating regional ETR and biomass production by using the SAFYE model (Duchemin et al. 2008) developed at CESBIO.
Gaseous nitrogen loss from biological soil crusts in desert environments (NgasLoss).
TA User (visit): Bettina Weber, University of Kaiserslautern, Germany (May, 2013 – 10 days).
Project Description: Many deserts are characterized by high landscape diversity, i.e. marked patchiness in the distribution of vegetation. In the northern Negev Desert of Israel, at Sayeret Shaked Park LTER site, there are two main types of patches; 1) shrub patches consisting of herbs and shrubs on slightly raised mounds; and 2) biological soil crust patches consisting of algae, cyanobacteria, lichens and mosses that are characterized by the presence of a relatively impermeable soil crust. Plant growth in the shrub patches is observed to be strongly limited by nitrogen (N), while N fixing cyanobacteria are thought to contribute significant amounts of N to the biological soil crust patches. Runoff following rainfall events is expected to transfer N from the crust covered biological soil crust patches to the shrub patches.
As a result of severe droughts during the last years, the landscape within the LTER site is changing mainly due to the death of a significant fraction of the dominant perennial shrubs. For more than a year we have followed the biophysical developmental stages of BSC invading the sites of the dead shrubs. The proposed project will be conducted within this experimental setup.
Denitrification, an anaerobic respiration that reduces N oxides (NO3-, NO2-) to N gases (N2O, N2), is an important process, depleting biologically available N from drylands.
Our hypotheses are: 1) that biological soil crusts, covering the soil surface in these areas, play a crucial role in N2O emission as they decrease the water infiltration rate and seal the soil surface; and 2) that N2O emission differs between the different developmental stages of biological soil crusts, i.e. initial cyanobacteria-dominated versus well-developed lichen- and bryophyte-dominated crusts. We will test our hypotheses both in the field and in the laboratory.
Our goal is to confirm that BSC structures, which occupy arid landscapes following climatic change-induced vegetation reduction, may contribute significantly to the emission of greenhouse gases (nitrous oxide).
Scintillometry under advective conditions in a semi-arid ecosystem (DesertScint).
TA User (visit): Hendrik de Bruin, Wageningen University, The Netherlands (May, 2013).
Project Description: Sustainable management of water resources in arid and semi-arid ecosystems requires accurate information on water utilization over a range of space and time scales. Water use in desert regions is highly variable in space and time. In effort to address this spatial heterogeneity, models yielding regional evapotranspiration (ET) maps were developed utilizing remote sensing imagery. Reliable validation of such models requires in situ measurement of the energy balance components at a spatial scale comparable to the satellites pixel size. Recently scintillometry was demonstrated to allow the measurement of sensible heat fluxes at footprint dimensions from 500 to 10,000 m, an area comparable with several pixels of a satellite image (tens of Landsat thermal pixels or a few MODIS thermal pixels). Under unstable conditions and for Bowen ratios larger than ~0.75, the method was proved to be reliable over homogeneous fields. Most of the experiments leading to this conclusion have been conducted in semi-arid areas. The need to expand agricultural land has brought many to develop agriculture in areas not naturally self-sustaining such activities. This has resulted in irrigated agricultural fields spread out in many semi-arid ecosystems. These irrigated fields yield high ET values, forming a sharp contrast with the dry surroundings. When the fields are an order of magnitude of several 10s of dunam local advection is likely to occur. The objective of this proposal is to study the ability of a displaced-beam laser scintillometer (DBLS) to measure mechanically generated turbulence under local advective conditions. The Negev LTER is located in a semi-arid region, where small irrigated agricultural fields are surrounded by dry natural ecosystems, creating frequent advective conditions. These conditions are ideal for testing the above defined research objective.
See his feedback
Exploration of relationship between vegetation-soil parameters and hyperspectral data (EVAHYPE).
TA User (visit): Christoph Salbach, Helmholtz Centre Potsdam GFZ, Germany (January, 2013).
Project Description: The objective of the proposed research is to link vegetation and soil variables with hyperspectral data. More specifically, to explore the most sensitive and relevant spectral bands, between the blue (400 nm) and the short wave infrared (2,500 nm), that are related to each of the biophysical variables. In order to fulfill these objectives two statistical methods will be employed: (1) Principle Components Regression (PCR) and Partial Least-Squares Regression (PLSR) methods that are used for soil reflectance spectroscopy, agricultural, and industrial applications, as well as for mass spectrometry metabolic data; and (2) Machine learning methods, such as Random Forest (RF), that pose attractive features for spectroscopic analysis. Parameters from these two methods will be compared for their ability to choose the most important bands out of a high-resolution spectra, robustness against noisy training data, and model validation assessments. The visit to the NEGEV site will enable comparisons and standardization of hyperspectral measurements and allow us to run the statistical models on the obtained spectral measurements. To realize the above objectives, a period of 3 months is planned.
See his feedback
Phenological and primary production studies by remote sensing (PHESENSENSING).
TA User (visit): Cornelia Glaesser, Martin Luther University, Institute of Geosciences, GERMANY (November, 2012).
Project Description: The current TA proposed project takes advantage of high-resolution (6.5 m) and high-temporal (a few days revisit time) Rapid-Eye data, which will be available during the 2012-2013 growing season over the NEGEV LTER site (Israel), in order to explore methods for spaceborne assessing of phenological scheme.
The NEGEV LTER is located in a semi-arid region. Natural vegetation in these regions is characterized by three ground features, in addition to bare surfaces – biological soil crusts, annuals, and perennials. It is hypothesized that these three elements have distinguishable phenological cycles that can be detected by spectral ground measurements and by calculating the weighted normalized difference vegetation index (NDVI). The latter is the product of the derived NDVI of each ground feature and its respective areal cover. Each phenological cycle has the same basic elements – oscillation from null (or low) to full photosynthetic status and back to a stage of senescence. However, they vary in phase. The biological soil crusts show the earliest and highest weighted NDVI peak during the rainy season, and their weighted NDVI signal lasts longer than that of the annuals. The annuals are dominant in late winter and early spring while the perennials predominate in late spring and during the summer.
For this project, RapidEye images will be acquired in Germany. A 10-day visit to the NEGEV site will be devoted to initialize the field spectral measurements that will be continued by Prof. Karnieli throughout the season. Spectral and image data will be mutual processed.
See her feedback
Remote Sensing of Forest Canopy Cover in the Negev LTER with RapidEye (RFCCRapidEye).
TA User (visit): Ibrahim Ozdemir, Suleyman Demirel University, Isparta, TURKEY (November, 2012).
Project Description: Forest canopy cover, also called crown cover, is described as the proportion of land surface covered by the vertical projection of tree crowns in a definite area. Maps of forest canopy cover are required for a wide range of ecological applications, including assessment of the status of suitable wildlife habitat, determination of fuel and potential fire risks, watershed management, erosion control, monitoring disturbance factors (thinning, illegal logging, insect damages etc.), detection of carbon pools, as well as timber management purposes. Measurement of forest canopy cover based on field measurements and air photographs is too laborious and time consuming to be used in extensive forest inventories. Satellite remote sensing seems to provide a good alternative to these traditional methods to map and monitor canopy cover and its changes over wide geographic regions. The potential use of new satellite systems should be investigated for estimating forest canopy cover. The RapidEye satellite image, with a 5 m spatial resolution and an additional red-edge band, is a cost efficient means for this purpose. Furthermore, in the literature, there is no work employing satellite images with 5 m resolution for estimating forest canopy cover. Therefore, our research goal is to develop and test a remote sensing platform for the prediction of forest canopy cover using vegetation indices derived from RapidEye satellite imagery. The research site will be the Yatir forest, part of the Negev LTER, located within a mature pine forest (Pinus halepensis, Mill) planted in a semi-arid zone. Multispectral remote sensing data from RapidEye sensor will be acquired. This recently launched satellite records radiance in five broad bands corresponding to the blue, green, red, red-edge and near-infrared (NIR) part of the electromagnetic spectrum. A set of 60 plots with the size of 2500 m2 will be designed, both in the aerial photos and in the satellite data. Forest canopy cover will be determined manually on the aerial photographs and then the corresponding vegetation indices will be calculated for each sample plot. We will assess the relationships between the vegetation indices (Normalized vegetation index, NDVI; soil adjusted vegetation index, SAVI; linear four point interpolation approach, REIP) and the forest canopy cover by simple and multiple linear regression analysis. 40 sample plots will be used for model fitting, and the remainder (n = 20) will be used for model validation. The expected research results are; 1) We will determine the vegetation indices that are the most correlated with forest canopy cover. 2) We will develop an approach to reduce the effect of understory vegetation on spectral values of the satellite data in calculating the vegetation indices. 3) We will generate a model for predicting and mapping the canopy cover, which might be used to monitor the changes on the forest canopy cover.
See his feedback
Ground and space measurements and modeling of soil albedo (GaSMoSAN).
TA User (visit): Jerzy Cierniewski, Adam Mickiwicz University, Poznan, POLAND (August, 2012).
Project Description: Energy exchange at the ecosystem scale is primarily influenced by surface albedo. Therefore, albedo can have substantial effects on climate. Strong links were evidenced between surface albedo and biodiversity, phenology, carbon sequestration, and more. The aim of the project is to explore the diurnal variation in broadband blue-sky albedo of bare soils in semi-arid ecosystems (Negev-LTER). Data sets, measured from ground level under clear sky conditions at minute intervals, enable us to find quantitative relations between data and the solar zenith angle for different desert surfaces with respect to their roughness. Our experience in measuring and modelling of soil surface bidirectional reflectance and albedo, which has been accomplished so far in the Negev, allows us to suppose that these relations can be clearly specific and different for smooth and rough surfaces. The relations will enable us to predict instantaneous albedo values of the surfaces at a chosen latitude during any day of the year. We expect that results of this study will allow us to predict the optimal local solar time when instantaneous soil albedo of those surfaces reaches their average diurnal value. The optimal time for this average albedo observation with an accepted error will be considered in context of its achievement by satellite radiometers on sun-synchronous orbits. The average diurnal surface albedo seems to be helpful as a basis for studies of environmental processes associated with the energy transfer between soil, vegetation and atmosphere. Soil albedo will be measured by albedometers within a spectral range of 335-2800 nm using data loggers, and the soil surface roughness will be recorded with stereo-photographs technology.
Output: Effects of Soil Surface Irregularities on the Diurnal Variation of Soil Broadband Blue-Sky Albedo (Article in press). Cierniewski, J., Karnieli, A., Kazmierowski, C., Krolewicz, S., Piekarczyk, J., Lewinska, K., Goldberg, A., Wesolowski, R., Orzechowski, M. - Article in Press
See his feedback