Investigation of pingos in Kuganguaq delta on Disko Island

The growth and collapse history of two pingos located approximately 18 m a.s.l, and 35 m a.s.l. in Kuganguag, Disko lsland, west-central Greenland were examined. The pingos of this area were formed on Tertiary basalt rocks. One of the pingos is located in the middle of an alluvial fan, the other is on a river bed. Both have already collapsed. The pingo on the river bed had already collapsed at least 3545±60 year BP (14C dating from base of the pond sediments in the pingo crater). Both pingos formed after the sea's retreat as permafrost developed in the newly exposed delta bottom.

Age determination of sediment cores from the Skagerrak

Paleoceanographic and paleoenvironmental interpretations based on foraminifera, sedimentary data, radiocarbon dates, and stable isotope measurements were derived from two sections in the Skagerrak: a 115-m-thick Holocene marine section drilled onshore at Skagen near the northernmost tip of Jutland, Denmark, and a 9-m piston core from the Skagerrak, north of Skagen. The foraminiferal data show that arctic-subarctic environments in the deep Skagerrak-Kattegat area were succeeded by boreal conditions at 9.6 ka. This was a result of northward migration of the Atlantic polar front and inflow of warm Atlantic water into the area through the Norwegian Channel. A gradual warming of the water masses after 9.6 ka is indicated by the data. Rare foraminifera and high sedimentation rates are found between approximately 8.6 ka and 7.6 ka at both core locations. The modern foraminiferal assemblages of the area were fully established at 7.6 ka indicating that the modern circulation pattern in the Skagerrak-Kattegat after the opening of the English Channel and the Danish Straits was not established before this date. At 5.5 ka a sudden change to coarser sediments (higher-energy environments) and the appearance of the foraminifer Eoeponidella laesoeensis is recorded in the Skagen core. This indicates a rapid change in the hydrography reflecting altered meteorological and hydrographic conditions in the Skagerrak-Kattegat, including a strengthening of the Jutland Current and increased inflow of North Sea water into the Kattegat. The event is interpreted as a response to cooling at the end of the Holocene climatic optimum in late Atlantic time and possibly reflects a rapid cooling event of North Atlantic surface water masses.

Concentrations of As and other chemical elements in bottom sediments of the Baltic Sea

As is less toxic than Hg, Cd, Pb, Se, Zn, and Cu. The As clarke for clays and shales is 10 ppm. Our samples of bottom sediments from Kurshskii Bay were determined to contain from 15 to 26 ppm As and up to 34 ppm As in the vicinity of the Neman River mouth. Elevated As concentrations (50-114 ppm) were detected in four columns of subsurface bottom sediments (at depths of 10-65 cm) from the Vistula Lagoon. Elevated As concentrations (50-180 ppm) were also found in a few surface samples of sand from the Gdansk Deep near oil platform D-6. These sediments are either partly contaminated with anthropogenic As or contain Fe sulfides and glauconite, which can concentrate As and contain its elevated concentrations. The As concentration in columns of bottom sediments from the Gulf of Finland were at the natural background level (throughout the columns) typical of the area (9-34 ppm). We repeatedly detected very high As concentrations (up to 227 ppm As) in politic ooze from Bornholm Deep, in the vicinity of the sunken vessel with chemical weapons. The sources of elevated As concentrations in the Baltic Sea are the following: (1) chemical weapon (CW) material buried in the floor of the Baltic Sea; (2) As-bearing pesticides, agricultural mineral fertilizers, and burned coal and other fuels; (3) kerogen-bearing Ordovician rocks exposed on the bottom; and (4) As-rich Fe sulfides brought to the area together with construction sand and gravel. This mixture was used in paper production and for the construction of hydraulic engineering facilities in the Vistula Lagoon in the early 20th century and later caused the so-called lagoon disease.

Pollen record and dating of a sediment profile from Lake Voulkaria, Greece

A palynological investigation of a Holocene profile from Lake Voulkaria, western Greece, was carried out as a contribution to the environmental history of the coastal area of northwestern Acarnania and the Classical city of Palairos. It shows that deciduous oaks dominated the natural vegetation of the area throughout the Holocene. Until ca. 7000 B.C. Pistacia occurred abundantly, while other evergreen woody taxa were rare. At ca. 6300 B.C. an expansion of Carpinus orientalis/Ostrya can be observed. Around ca. 5300 B.C. spreading of Erica indicates a change to a drier climate and/or first human impact. Since ca. 3500 B.C. an increase of evergreen shrubs now clearly indicates land-use. The foundation of the Classical city of Palairos led to a temporary expansion of Phillyrea maquis. Within this period, molluscs of brackish water indicate the use of the lake as a harbour after the construction of a connection to the sea. The deciduous Quercus woodland recovered when human impact decreased in the area, and lasted until modern times.

Biomass of mesozooplankton in the western part of the Black Sea in 1992 during the Cooperative Marine Science Program for the Black Sea (CoMSBlack)

The "CoMSBlack92" dataset is based on samples collected in the summer of 1992 along the Bulgarian coast including coastal and open sea areas. The whole dataset is composed of 79 samples (28 stations) with data of zooplankton species composition, abundance and biomass. Sampling for zooplankton was performed from bottom up to the surface at standard depths depending on water column stratification and the thermocline depth. Zooplankton samples were collected with vertical closing Juday net,diameter - 36cm, mesh size 150 ?m. Tows were performed from surface down to bottom meters depths in discrete layers. Samples were preserved by a 4% formaldehyde sea water buffered solution. Sampling volume was estimated by multiplying the mouth area with the wire length. Sampling volume was estimated by multiplying the mouth area with the wire length. The collected material was analysed using the method of Domov (1959). Samples were brought to volume of 25-30 ml depending upon zooplankton density and mixed intensively until all organisms were distributed randomly in the sample volume. After that 5 ml of sample was taken and poured in the counting chamber which is a rectangle form for taxomomic identification and count. Large (> 1 mm body length) and not abundant species were calculated in whole sample. Counting and measuring of organisms were made in the Dimov chamber under the stereomicroscope to the lowest taxon possible. Taxonomic identification was done at the Institute of Oceanology by Asen Konsulov using the relevant taxonomic literature (Mordukhay-Boltovskoy, F.D. (Ed.). 1968, 1969,1972 ). The biomass was estimated as wet weight by Petipa, 1959 (based on species specific wet weight). Wet weight values were transformed to dry weight using the equation DW=0.16*WW as suggested by Vinogradov & Shushkina, 1987. Copepods and Cladoceras were identified and enumerated; the other mesozooplankters were identified and enumerated at higher taxonomic level (commonly named as mesozooplankton groups). Large (> 1 mm body length) and not abundant species were calculated in whole sample. The biomass was estimated as wet weight by Petipa, 1959 ussing standard average weight of each species in mg/m**3.

Mineral composition of sediments from ODP Leg 190 sites and DSDP Hole 31-297

This data report documents the acquisition of two new sets of normalization factors for semiquantitative X-ray diffraction analyses. One set of factors is for bulk sediment powders, and the other applies to oriented aggregates of clay-sized fractions (<2 µm). We analyzed mixtures of standard minerals with known weight percentages of each component and solved for the normalization factors using matrix singular value decomposition. The components in bulk powders include total clay minerals (a mixture of smectite, illite, and chlorite), quartz, plagioclase, and calcite. For clay-sized fractions, the minerals are smectite, illite, chlorite, and quartz. We tested the utility of the method by analyzing natural mudstone specimens from Site 297 of the Deep Sea Drilling Project, which is located in the Shikoku Basin south of Site 1177 of the Ocean Drilling Program (Ashizuri transect).

Biomass of mesozooplankton in the western part of the Black Sea in summer 1991 during the NATO Programme Hydroblack

The "Hydroblack91" dataset is based on samples collected in the summer of 1991 and covers part of North-Western in front of Romanian coast and Western Black Sea (Bulgarian coasts) (between 43°30' - 42°10' N latitude and 28°40'- 31°45' E longitude). Mesozooplankton sampling was undertaken at 20 stations. The whole dataset is composed of 72 samples with data of zooplankton species composition, abundance and biomass. Samples were collected in discrete layers 0-10, 0-20, 0-50, 10-25, 25-50, 50-100 and from bottom up to the surface at depths depending on water column stratification and the thermocline depth. Zooplankton samples were collected with vertical closing Juday net,diameter - 36cm, mesh size 150 µm. Tows were performed from surface down to bottom meters depths in discrete layers. Samples were preserved by a 4% formaldehyde sea water buffered solution. Sampling volume was estimated by multiplying the mouth area with the wire length. Mesozooplankton abundance: The collected materia was analysed using the method of Domov (1959). Samples were brought to volume of 25-30 ml depending upon zooplankton density and mixed intensively until all organisms were distributed randomly in the sample volume. After that 5 ml of sample was taken and poured in the counting chamber which is a rectangle form for taxomomic identification and count. Large (> 1 mm body length) and not abundant species were calculated in whole sample. Counting and measuring of organisms were made in the Dimov chamber under the stereomicroscope to the lowest taxon possible. Taxonomic identification was done at the Institute of Oceanology by Asen Konsulov using the relevant taxonomic literature (Mordukhay-Boltovskoy, F.D. (Ed.). 1968, 1969,1972). The biomass was estimated as wet weight by Petipa, 1959 (based on species specific wet weight). Wet weight values were transformed to dry weight using the equation DW=0.16*WW as suggested by Vinogradov & Shushkina, 1987. Taxon-specific abundance: The collected material was analysed using the method of Domov (1959). Samples were brought to volume of 25-30 ml depending upon zooplankton density and mixed intensively until all organisms were distributed randomly in the sample volume. After that 5 ml of sample was taken and poured in the counting chamber which is a rectangle form for taxomomic identification and count. Copepods and Cladoceras were identified and enumerated; the other mesozooplankters were identified and enumerated at higher taxonomic level (commonly named as mesozooplankton groups). Large (> 1 mm body length) and not abundant species were calculated in whole sample. Counting and measuring of organisms were made in the Dimov chamber under the stereomicroscope to the lowest taxon possible. Taxonomic identification was done at the Institute of Oceanology by Asen Konsulov using the relevant taxonomic literature (Mordukhay-Boltovskoy, F.D. (Ed.). 1968, 1969,1972). The biomass was estimated as wet weight by Petipa, 1959 ussing standard average weight of each species in mg/m3. WW were converted to DW by equation DW=0.16*WW (Vinogradov ME, Sushkina EA, 1987).

Radionuclide concentrations, ratios, and derived ages of massive sulfides from ODP Leg 139 and Leg 169 sites

Massive sulfide samples from the Bent Hill area were analyzed for 230Th/234U and 231Pa/235U disequilibria. Apparent ages calculated from these ratios are between 8.2 and >300 ka. Concordant ages were found for only three samples that originate near the surface from the clastic sulfide zone and suggest "true" ages of between 8.5 and 16.0 ka (mean of 230Th and 231Pa ages). The uranium vs. depth distribution in the Bent Hill Massive Sulfide deposit suggests an open system for uranium for the deeper part of the deposit, which was probably caused by extensive recrystallization processes inhibiting true age determinations.

Palynofacies analysis of sediments in the eastern Equatorial Atlantic

Analyses of the palynofacies and sporomorph thermal alteration indices (TAI) of sediments from Ocean Drilling Program (ODP) Sites 959 to 962 in the Cote d'Ivoire-Ghana Transform Margin, West Africa were undertaken to (1) determine the source and depositional conditions of the organic matter in the sediments, (2) refine a paleobathymetric curve derived from other data for Site 959, which drilled the most continuous sedimentary sequence from Pleistocene to Albian and (3) interpret the paleothermal history of the area. Twelve types of dispersed organic matter were identified: amorphous organic matter (AOM), marine palynomorphs, algae, resins, black debris, yellow-brown fragments, black-brown fragments, cuticles, plant tissue, wood, sporomorphs and fungi, The relative abundances of these organic matter components at each site were analyzed using cluster analysis, resulting in the identification of seven palynofacies assemblages at Site 959, five each at sites 960 and 961, and four at Site 962. Amorphous organic matter (which is chiefly marine derived), black debris and wood have played the most significant role in defining palynofacies assemblages. The palynofacies assemblages show some correlation with lithologic units, sediment sources and depositional environments. Previous palynofacies studies in passive margins have demonstrated that changes in the ratio of AOM to terrestrial organic matter are related primarily to proximal-distal positions of depositional environments relative to the shoreline. However, this assumption does not always hold true for a transform margin where tectonic factors play an important role in the organic matter distribution, at least in the early stages of evolution. Lithofacies, CCD paleodepths for the North Atlantic, trace fossil association, benthic foraminifera and palynofacies data were the criteria used for reconstructing a paleobathymetric curve for Site 959. A cyclicity in the organic matter distribution of the Upper Miocene to Lower Pliocene pelagic sediments could be related to fluctuations in productivity of biosiliceous and calcareous organisms, and sedimentation rates. A drastic increase in the amount of AOM and a decrease in black debris and wood in the carbonate and clastic rocks (Lithologic Unit IV) overlying the tectonized Albian sediments (Lithologic Unit V) at Sites 959 and 960 coincide with the presence of an unconformity. Qualitative color analysis of palynomorphs was undertaken for all sites, although the main focus was on Site 959 where detailed organic geochemical data were available. At Site 959, TAI values indicate an immature stage of organic maturation (<2) down to the black claystones of Lithologic Unit III at about 918.47 mbsf. Below this, samples show an increase with depth to a moderately mature stage (>2 except for the claystone samples between 1012.52 and 1036.5 mbsf, and one limestone sample at 1043.4 mbsf), reaching peak levels of 2.58 to 3.0 in the tectonized Albian sediments below the unconformity. These TAI values show a positive correlation with the Tma x values derived from Rock-Eval pyrolysis data. The highest values recorded in the basal tectonized units at all the sites (Sites 960-962 have mean values between 2.25 and 3.13) may be related to high heat flow during the intracontinental to syntransform basin stage in the region.

Abundance of mesozooplankton during the long-tern mesozooplankton analysis in Sevastopol Bay from 1976 to 2003

The present dataset includes results of analysis of 227 zooplankton samples taken in and off the Sevastopol Bay in the Black Sea in 1976, 1979-1980, 1989-1990, 1995-1996 and 2002-2003. Exact coordinates for stations 1, 4, 5 and 6 are unknown and were calculated using Google-earth program. Data on Ctenophora Mnemiopsis leidyi and Beroe ovata are not included. Juday net: Vertical tows of a Juday net, with mouth area 0.1 m**2, mesh size 150µm. Tows were performed at layers. Towing speed: about 0.5 m/s. Samples were preserved by a 4% formaldehyde sea water buffered solution. Sampling volume was estimated by multiplying the mouth area with the wire length. The collected material was analysed using the method of portions (Yashnov, 1939). Samples were brought to volume of 50 - 100 ml depending upon zooplankton density and mixed intensively until all organisms were distributed randomly in the sample volume. After that 1 ml of sample was taken by calibrated Stempel-pipette. This operation was produced twice. If divergence between two examined subsamples was more than 30% one more subsample was examined. Large (> 1 mm body length) and not abundant species were calculated in 1/2, 1/4, 1/8, 1/16 or 1/32 part of sample. Counting and measuring of organisms were made in the Bogorov chamber under the stereomicroscope to the lowest taxon possible. Number of organisms per sample was calculated as simple average of two subsamples meanings multiplied on subsample volume. Total abundance of mesozooplankton was calculated as sum of taxon-specific abundances and total abundance of Copepods was calculated as sum of copepods taxon-specific abundances.

Pollen analysis of a peat profile from the Rieseberger Moor, Lower Saxony, Germany

The Rieseberger Moor is a fen, 145 hectares in size, situated about 20 km east of Brunswick (Braunschweig), Lower Saxony, Germany. Peat was dug in the fen - with changing intensity - since the mid-18th century until around AD 1955. According to Schneekloth & Schneider (1971) the remaining peat (fen and wood peat) is predominantly 1.5 to 2 m thick (maximum 2.7 m). Part of the fen - now a nature reserve (NSG BR 005) - is wooded (Betula, Salix, Alnus). For more information on the Rieseberger Moor see http://de.wikipedia.org/wiki/Rieseberger_Moor. Willi Selle was the first to publish pollen diagrams from this site (Selle 1935, profiles Rieseberger Torfmoor I and II). This report deals with a 2.2 m long profile from the wooded south-eastern part of the fen consisting of strongly decomposed fen peat taken A.D. 1965 and studied by pollen analysis in the same year. The peat below 1.45 m contained silt and clay, samples 1.48 and 1.58 m even fine sand. These samples had to be treated with HF (hydrofluoric acid) in addition to the treatment with hot caustic potash solution. The coring ended in sandy material. The new pollen data reflect the early part of the known postglacial development of the vegetation of this area: the change from a birch dominated forest to a pine forest and the later spreading of Corylus and of the thermophilous deciduous tree genera Quercus, Ulmus, Tilia and Fraxinus followed by the expansion of Alnus. The new data are in agreement with Selle's results, except for Alnus, which in Selle's pollen diagram II shows high values (up to 42% of the arboreal pollen sum) even in samples deposited before Corylus and Quercus started to spread. On contrary the new pollen diagram shows that alder pollen - although present in all samples - is frequent in the three youngest pollen spectra only. A period with dominating Alnus as seen in the uppermost part of Selle's pollen diagrams is missing. The latter is most likely the result of peat cutting at the later coring site, whereas the early, unusually high alder values of Selle's pollen study are probably caused by contamination of the pollen samples with younger peat. Selle took peat samples usually with a "Torfbohrer" (= Hiller sampler). This side-filling type of sampler with an inner chamber and an outer loose jacket offers - if not handled with appropriate care - ample opportunities to contaminate older peat with carried off younger material. Pollen grains of Fagus (2 % of the arboreal pollen sum) were found in two samples only, namely in the uppermost samples of the new profile (0.18 m) and of Selle's profile I (0.25 m). If this pollen is autochthonous, with other words: if this surface-near peat was not disturbed by human activities, the Fagus pollen indicates an Early Subboreal age of this part of the profile. The accumulation of the Rieseberg peat started during the Preboreal. Increased values of Corylus, Quercus and Ulmus indicate that sample 0.78 m of the new profile is the oldest Boreal sample. The high Alnus values prove the Atlantic age of the younger peat. Whether Early Subboreal peat exists at the site is questionable, but evidently none of the three profiles reaches to Late Subboreal time, when Fagus spread in the region. Did peat-growth end during the Subboreal? Did younger peat exist, but got lost by peat cutting or has younger peat simply not yet been found in the Rieseberg fen? These questions cannot be answered with this study. The temporary decline of the curve of Pinus for the benefit of Betula during the Preboreal, unusual for this period, is contemporaneous with the deposition of sand (Rieseberger Moor II, 1.33 - 1,41 m; samples 1.48 and 1.58 m of the new profile) and must be considered a local phenomenon. Literature: Schneekloth, Heinrich & Schneider, Siegfried (1971). Die Moore in Niedersachsen. 2. Teil. Bereich des Blattes Braunschweig der Geologischen Karte der Bundesrepublik Deutschland (1:200000). - Schriften der wirtschaftswissenschaftlichen Gesellschaft zum Studium Niedersachsens e.V. Reihe A I., Band 96, Heft 2, 83 Seiten, Göttingen. Selle, Willi (1935) Das Torfmoor bei Rieseberg. - Jahresbericht des Vereins für Naturwissenschaft zu Braunschweig, 23, 46-58, Braunschweig.

Registry of all samples from the Tara Oceans Expedition (2009-2013)

The Tara Oceans Expedition (2009-2013) sampled the world oceans on board a 36 m long schooner, collecting environmental data and organisms from viruses to planktonic metazoans for later analyses using modern sequencing and state-of-the-art imaging technologies. Tara Oceans Data are particularly suited to study the genetic, morphological and functional diversity of plankton. Data sets in this collection provide methodological and environmental context to all samples collected during the Tara Oceans Expedition (2009-2013).

Use of camera trapping in determining Iberian lynx population parameters: The use area and its limitations

Below are the results of the survey of the Iberian lynx obtained with camera-trapping between 2000 and 2007 in Sierra Morena. Two very important aspects of camera-trapping concerning its efficiency are also analyzed. The first is the evolution along years according to the camera-trapping type used of two efficiency indicators. The results obtained demonstrate that the most efficient lure is rabbit, though it is the less proven (92 trap-nights), followed by camera-trapping in the most frequent marking places (latrines). And, we propose as a novel the concept of use area as a spatial reference unit for the camera-trapping monitoring of non radio-marked animals is proposed, and its validity discussed.

Implicaciones de la accesibilidad configuracional en la movilidad peatonal. El caso de Madrid = The implications of configurational accessibility in pedestrian mobility. The case of Madrid

- Resumen La hipótesis que anima esta tesis doctoral es que algunas de las características del entorno urbano, en particular las que describen la accesibilidad de su red de espacio público, podrían estar relacionadas con la proporción de viajes a pie o reparto modal, que tiene cada zona o barrio de Madrid. Uno de los puntos de partida de dicha hipótesis que el entorno urbano tiene una mayor influencia sobre los viaje a pie que en sobre otros modos de transporte, por ejemplo que en los viajes de bicicleta o en transporte público; y es que parece razonable suponer que estos últimos van a estar más condicionadas por ejemplo por la disponibilidad de vías ciclistas, en el primer caso, o por la existencia de un servicio fiable y de calidad, en el segundo. Otra de las motivaciones del trabajo es que la investigación en este campo de la accesibilidad del espacio público, en concreto la denominada “Space Syntax”, ha probado en repetidas ocasiones la influencia de la red de espacio público en cómo se distribuye la intensidad del tráfico peatonal por la trama urbana, pero no se han encontrado referencias de la influencia de dicho elemento sobre el reparto modal. De acuerdo con la hipótesis y con otros trabajos anteriores se propone una metodología basada en el análisis empírico y cuantitativo. Su objetivo es comprobar si la red de espacio público, independientemente de otras variables como los usos del suelo, incluso de las variables de ajenas entorno no construido, como las socioeconómicas, está o no relacionada estadísticamente con la proporción de peatones viajes en las zonas urbanas. Las técnicas estadísticas se utilizan para comprobar sistemáticamente la asociación de las variables del entorno urbano, denominadas variables independientes, con el porcentaje de viajes a pie, la variable dependiente. En términos generales, la metodología es similar a la usada en otros trabajos en este campo como los de CERVERÓ y KOCKLEMAN (1997), CERVERÓ y DUNCAN (2003), o para los que se utilizan principalmente en la revisión general de TRB (2005) o, más recientemente, en ZEGRAS (2006) o CHATMAN (2009). Otras opciones metodológicas, como los métodos de preferencias declaradas (ver LOUVIERE, HENSHER y SWAIT, 2000) o el análisis basado en agentes (PENN & TURNER, 2004) fueron descartados, debido a una serie de razones, demasiado extensas para ser descritas aquí. El caso de estudio utilizado es la zona metropolitana de Madrid, abarcándola hasta la M-50, es decir en su mayor parte, con un tamaño aproximado de 31x34 Km y una población de 4.132.820 habitantes (aproximadamente el 80% de la población de la región). Las principales fuentes de datos son la Encuesta Domiciliaria de Movilidad de 2004 (EDM04), del Consorcio Regional de Transportes de Madrid que es la última disponible (muestra: > 35.000 familias,> 95.000 personas), y un modelo espacial del área metropolitana, integrando el modelo para calcular los índices de Space Syntax y un Sistema de Información Geográfica (SIG). La unidad de análisis, en este caso las unidades espaciales, son las zonas de transporte (con una población media de 7.063 personas) y los barrios (con una población media de 26.466 personas). Las variables del entorno urbano son claramente el centro del estudio. Un total de 20 índices (de 21) se seleccionan de entre los más relevantes encontrados en la revisión de la producción científica en este campo siendo que, al mismo tiempo, fueran accesibles. Nueve de ellos se utilizan para describir las características de los usos del suelo, mientras que otros once se usan para describir la red de espacios públicos. Estos últimos incluyen las variables de accesibilidad configuracional, que son, como se desprende de su título, el centro del estudio propuesto. La accesibilidad configuracional es un tipo especial de accesibilidad que se basa en la configuración de la trama urbana, según esta fue definida por HILLIER (1996), el autor de referencia dentro de esta línea de investigación de Space Syntax. Además se incluyen otras variables de la red de espacio público más habituales en los estudios de movilidad, y que aquí se denominan características geométricas de los elementos de la red, tales como su longitud, tipo de intersección, conectividad, etc. Por último se incluye además una variable socioeconómica, es decir ajena al entorno urbano, para evaluar la influencia de los factores externos, pues son varios los que pueden tener un impacto en la decisión de caminar (edad, género, nivel de estudios, ingresos, tasa de motorización, etc.). La asociación entre las variables se han establecido usando análisis de correlación (bivariante) y modelos de análisis multivariante. Las primeras se calculan entre por pares entre cada una de las 21 variables independientes y la dependiente, el porcentaje de viajes a pie. En cuanto a los segundos, se han realizado tres tipos de estudios: modelo multivariante general lineal, modelo multivariante general curvilíneo y análisis discriminante. Todos ellos son capaces de generar modelos de asociación entre diversas variables, pudiéndose de esta manera evaluar con bastante precisión en qué medida cada modelo reproduce el comportamiento de la variable dependiente, y además, el peso o influencia de cada variable en el modelo respecto a las otras. Los resultados fundamentales del estudio se expresan en dos modelos finales alternativos, que demuestran tener una significativa asociación con el porcentaje de viajes a pie (R2 = 0,6789, p <0,0001), al explicar las dos terceras partes de su variabilidad. En ellos, y en general en todo el estudio realizado, se da una influencia constante de tres índices en particular, que quedan como los principales. Dos de ellos, de acuerdo con muchos de los estudios previos, corresponden a la densidad y la mezcla de usos del suelo. Pero lo más novedoso de los resultados obtenidos es que el tercero es una medida de la accesibilidad de la red de espacio público, algo de lo que no había referencias hasta ahora. Pero, ¿cuál es la definición precisa y el peso relativo de cada uno en el modelo, es decir, en la variable independiente? El de mayor peso en la mayor parte de los análisis realizados es el índice de densidad total (n º residentes + n º puestos de trabajo + n º alumnos / Ha). Es decir, una densidad no sólo de población, sino que incluye algunas de las actividades más importantes que pueden darse una zona para generar movilidad a pie. El segundo que mayor peso adquiere, llegando a ser el primero en alguno de los análisis estadísticos efecturados, es el índice de accesibuilidad configuracional denominado integración de radio 5. Se trata de una medida de la accesibilidad de la zona, de su centralidad, a la escala de, más un menor, un distrito o comarca. En cuanto al tercero, obtiene una importancia bastante menor que los anteriores, y es que representa la mezcla de usos. En concreto es una medida del equilibrio entre los comercios especializados de venta al por menor y el número de residentes (n º de tiendas especializadas en alimentación, bebidas y tabaco / n º de habitantes). Por lo tanto, estos resultados confirman buena parte de los de estudios anteriores, especialmente los relativas a los usos del suelo, pero al mismo tiempo, apuntan a que la red de espacio público podría tener una influir mayor de la comprobada hasta ahora en la proporción de peatones sobre el resto de modos de transportes. Las razones de por qué esto puede ser así, se discuten ampliamente en las conclusiones. Finalmente se puede precisar que dicha conclusión principal se refiere a viajes de una sola etapa (no multimodales) que se dan en los barrios y zonas del área metropolitana de Madrid. Por supuesto, esta conclusión tiene en la actualidad, una validez limitada, ya que es el resultado de un solo caso — Abstract The research hypothesis for this Ph.D. Thesis is that some characteristics of the built environment, particularly those describing the accessibility of the public space network, could be associated with the proportion of pedestrians in all trips (modal split), found in the different parts of a city. The underlying idea is that walking trips are more sensitive to built environment than those by other transport modes, such as for example those by bicycle or by public transport, which could be more conditioned by, e.g. infrastructure availability or service frequency and quality. On the other hand, it has to be noted that the previously research on this field, in particular within Space Syntax’s where this study can be referred, have tested similar hypothesis using pedestrian volumes as the dependent variable, but never against modal split. According to such hypothesis, research methodology is based primarily on empirical quantitative analysis, and it is meant to be able to assess whether public space network, no matter other built environment and non-built environment variables, could have a relationship with the proportion of pedestrian trips in urban areas. Statistical techniques are used to check the association of independent variables with the percentage of walking in all trips, the dependent one. Broadly speaking this methodology is similar to that of previous studies in the field such as CERVERO&KOCKLEMAN (1997), CERVERO & DUNCAN (2003), or to those used mainly in the general review of T.R.B. (2005) or, more recently in ZEGRAS (2006) or CHATMAN (2009). Other methodological options such as stated choice methods (see LOUVIERE, HENSHER & SWAIT, 2000) or agent based analysis (PENN & TURNER, 2004), were discarded, due to a number of reasons, too long to be described here. The case study is not the entire Madrid’s metropolitan area, but almost (4.132.820 inhabitants, about 80% of region´s population). Main data sources are the Regional Mobility Home Based Survey 2004 (EDM04), which is the last available (sample: >35.000 families, > 95.000 individuals), and a spatial model of the metropolitan area, developed using Space Syntax and G.I.S. techniques. The analysis unit, in this case spatial units, are both transport zones (mean population = 7.063) and neighborhoods (mean population = 26.466). The variables of the built environment are clearly the core of the study. A total of 20 (out of 21) are selected from among those found in the literature while, at the same time, being accessible. Nine out of them are used to describe land use characteristics while another eleven describe the network of public spaces. Latter ones include configurational accessibility or Space Syntax variables. This is a particular sort of accessibility related with the concept of configuration, by HILLIER (1996), one of the main authors of Space Syntax, But it also include more customary variables used in mobility research to describe the urban design or spatial structure (here public space network), which here are called geometric characteristics of the such as its length, type of intersection, conectivity, density, etc. Finally a single socioeconomic variable was included in order to assess the influence non built environment factors that also may have an impact on walking (age, income, motorization rate, etc.). The association among variables is worked out using bi-variate correlation analysis and multivariate-analysis. Correlations are calculated among the 21 independent variables and the dependent one, the percentage of walking trips. Then, three types of multi-variate studies are run: general linear, curvilinear and discriminant multi-variate analysis. The latter are fully capable of generating complex association models among several variables, assessing quite precisely to what extent each model reproduces the behavior of the dependent variable, and also the weight or influence of each variable in the model. This study’s results show a consistent influence of three particular indexes in the two final alternative models of the multi-variate study (best, R2=0,6789, p<0,0000). Not surprisingly, two of them correspond to density and mix of land uses. But perhaps more interesting is that the third one is a measure of the accessibility of the public space network, a variable less important in the literature up to now. Additional precisions about them and their relative weight could also be of some interest. The density index is not only about population but includes most important activities in an area (nº residents + nº jobs+ nº students/Ha). The configurational index (radius 5 integration) is a measure of the accessibility of the area, i.e. centrality, at the scale of, more a less, a district. Regarding the mix of land uses index, this one is a measure of the balance between retail, in fact local basic retail, and the number of residents (nº of convenience shops / nº of residents). Referring to their weights, configurational index (radius 5 integration) gets the higher standardized coefficient of the final equation. However, in the final equations, there are a higher number of indexes coming from the density or land use mix categories than from public space network enter. Therefore, these findings seem to support part of the field’s knowledge, especially those concerning land uses, but at the same time they seem to bring in the idea that the configuration of the urban grid could have an influence in the proportion of walkers (as a part of total trips on any transport mode) that do single journey trips in the neighborhoods of Madrid, Spain. Of course this conclusion has, at present, a limited validity since it’s the result of a single case. The reasons of why this can be so, are discussed in the last part of the thesis.