排序方式: 共有42条查询结果,搜索用时 46 毫秒
1.
2.
Evaluation of current statistical approaches for predictive geomorphological mapping 总被引:5,自引:3,他引:5
Predictive models are increasingly used in geomorphology, but systematic evaluations of novel statistical techniques are still limited. The aim of this study was to compare the accuracy of generalized linear models (GLM), generalized additive models (GAM), classification tree analysis (CTA), neural networks (ANN) and multiple adaptive regression splines (MARS) in predictive geomorphological modelling. Five different distribution models both for non-sorted and sorted patterned ground were constructed on the basis of four terrain parameters and four soil variables. To evaluate the models, the original data set of 9997 squares of 1 ha in size was randomly divided into model training (70%, n=6998) and model evaluation sets (30%, n=2999).In general, active sorted patterned ground is clearly defined in upper fell areas with high slope angle and till soils. Active non-sorted patterned ground is more common in valleys with higher soil moisture and fine-scale concave topography. The predictive performance of each model was evaluated using the area under the receiver operating characteristic curve (AUC) and the Kappa value. The relatively high discrimination capacity of all models, AUC=0.85–0.88 and Kappa=0.49–0.56, implies that the model's predictions provide an acceptable index of sorted and non-sorted patterned ground occurrence. The best performance for model calibration data for both data sets was achieved by the CTA. However, when the predictive mapping ability was explored through the evaluation data set, the model accuracies of CTA decreased clearly compared to the other modelling techniques. For model evaluation data MARS performed marginally best.Our results show that the digital elevation model and soil data can be used to predict relatively robustly the activity of patterned ground in fine scale in a subarctic landscape. This indicates that predictive geomorphological modelling has the advantage of providing relevant and useful information on earth surface processes over extensive areas, such data being unavailable through more conventional survey methods. 相似文献
3.
Liisa Nevalainen Marttiina V. Rantala Tomi P. Luoto Milla Rautio Antti E. K. Ojala 《Boreas: An International Journal of Quaternary Research》2015,44(2):401-412
Long‐term fluctuations in lake‐water optical properties were examined using a Holocene sediment sequence and multi‐proxy palaeolimnological approach in Lake Einstaken, Nordaustlandet, Svalbard. UV‐absorbance of sedimentary cladoceran remains provided information on underwater UV exposure and changes in lake‐catchment coupling processes were inferred from sediment geochemistry. In addition, aquatic community succession was used as an indicator for lake‐water bio‐optical properties and a Holocene record of sun activity (sunspots) was utilized to evaluate long‐term solar forcing. The results indicated that the UV‐absorbance of cladoceran remains was highest (i.e. maximum UV‐induced pigmentation) for a short period during the early Holocene and for several millennia during the mid‐Holocene. Sun activity was high during these time intervals, probably impacting the UV intensities, but it is probable that the amount of UV‐attenuating compounds (e.g. dissolved organic carbon (DOC)) also significantly affected the underwater UV environment and were low during high UV exposure. Benthic autotrophic communities also responded to the millennial changes in lake‐water optical properties. UV‐resistant Nostoc cyanobacterial colonies were established during the mid‐Holocene, indicative of high underwater UV intensities, and Fontinalis mosses thrived during the early Holocene, indicating a highly transparent water column. The results further suggested that underwater UV exposure decreased during the late Holocene, which is probably attributable to increased DOC and decreased solar forcing. Owing to the location of Lake Einstaken and its catchment in the periglacial barren landscape of the polar desert, the fluctuations of bio‐optical lake‐water properties were apparently forced by postglacial environmental processes and Holocene climate development. These factors controlled sea shoreline proximity, water discharge, ice‐cover duration and littoral‐benthic primary production and further affected the underwater UV environment. Although the role of solar forcing cannot be underestimated, the current record emphasizes the role of climate‐mediated lake‐catchment interactions in impacting bio‐optical properties and UV exposure of high arctic aquatic systems. 相似文献
4.
Stefan Engels Angela E. Self Tomi P. Luoto Stephen J. Brooks Karin F. Helmens 《Journal of Paleolimnology》2014,51(4):529-547
Multiple regional chironomid–climate calibration datasets are available to reconstruct quantitatively July air temperatures from fossil chironomid assemblages. We examined the relationship between July air temperature and the 40 most common chironomid taxa in three independent Eurasian calibration (training) sets. The estimated temperature optimum of each chironomid taxon is systematically lower (by ~1–2 °C) in a Norwegian calibration set compared to Finnish and Russian calibration sets. This result might partly be explained by the fact that the Norwegian calibration set extends further at the cold end of the temperature gradient. A difference in continentality between the Russian sites and the European sites might also contribute to this pattern. The number of taxa that show a statistically significant unimodal response to temperature is higher in the Norwegian calibration set (34 out of 40 taxa) compared to the modern Finnish (11 of 37 taxa; 3 common taxa absent) and the Russian calibration set (20 of 40 taxa), probably due to the longer temperature gradient incorporated in the Norwegian calibration set. We applied all three calibration sets to fossil chironomid assemblages from the high-latitude study site of Sokli (northeast Finland), a site with a unique series of lacustrine deposits covering (amongst others) the Holocene, part of early MIS 3 (at ~53 ka) and MIS 5d–c (at ~110–95 ka) and with independent proxy-records for comparison. In the early Holocene and during MIS 5c, the chironomid-based temperature inferences from all three inference models had similar values. Temperature reconstructions based on the Norwegian calibration set are 2–4 °C lower for the late Holocene, early MIS 3 and MIS 5d than the inferred temperatures based on the other calibration sets. Although the lakes included in the Finnish calibration set are located closest to the site of Sokli, evaluation tests and a comparison with independent proxy data suggests that the Norwegian calibration set provides the most suitable analogues for reconstruction purposes for most of the fossil assemblages. Our results imply that when choosing a calibration set for quantitative climate reconstructions on glacial timescales, regional proximity of the fossil site may not be a sufficient basis, and the length of the temperature gradient of the calibration dataset and factors such as the continentality gradient covered by the calibration set must also be considered. 相似文献
5.
Extreme temperatures are key drivers controlling both biotic and abiotic processes, and may be strongly modified by topography and land cover. We modelled mean and extreme temperatures in northern Fennoscandia by combining digital elevation and land cover data with climate observations from northern Finland, Norway and Sweden. Multivariate partitioning technique was utilized to investigate the relative importance of environmental variables for the variation of the three temperature parameters: mean annual absolute minima and maxima, and mean annual temperature. Generalized additive modeling showed good performance, explaining 84–95 % of the temperature variation. The inclusion of remotely sensed variables improved significantly the modelling of thermal extremes in this system. The water cover variables and topography were the most important drivers of minimum temperatures, whereas elevation was the most important factor controlling maximum temperatures. The spatial variability of mean temperatures was clearly driven by geographical location and the effects of topography. Partitioning technique gave novel insights into temperature-environment relationship at the meso-scale and thus proved to be useful tool for the study of the extreme temperatures in the high-latitude setting. 相似文献
6.
Tomi P. Luoto Marjut Kaukolehto Jan Weckström Atte Korhola Minna Väliranta 《Quaternary Research》2014
Paleoclimate reconstructions based on biological proxies present methodological challenges, especially during non-analog conditions, such as the early Holocene. Here, two chironomid-based training sets from Finland were amalgamated to create a more accurate transfer function of summer air temperature. The aim was to reconstruct Holocene paleoclimate in northernmost Lapland, in an area that has been either too warm or too cold for reliable reconstructions using the original calibration models. The results showed that the combined calibration model had improved performance statistics. The temperature trends inferred from the downcore chironomid record using the original and combined models were very similar. However, there were major changes in their absolute values with the combined model showing greatly improved accuracy. The chironomid-based temperature reconstruction showed significant correlation with the previous pollen-based reconstructions from northwestern Finnish Lapland. However, differences were observed in the temperature trends of the early Holocene, when the chironomid-inferred temperatures rapidly increased, but the pollen-based reconstructions lagged behind suggesting that a cool climate continued for much longer. However, similar to the chironomid record, new plant macrofossil evidence from northwestern Finland also showed warmer-than-present early Holocene temperatures. Therefore, we conclude that the early Holocene was probably warm in northern Lapland. 相似文献
7.
Soil moisture has a pronounced effect on earth surface processes. Global soil moisture is strongly driven by climate, whereas at finer scales, the role of non‐climatic drivers becomes more important. We provide insights into the significance of soil and land surface properties in landscape‐scale soil moisture variation by utilizing high‐resolution light detection and ranging (LiDAR) data and extensive field investigations. The data consist of 1200 study plots located in a high‐latitude landscape of mountain tundra in north‐western Finland. We measured the plots three times during growing season 2016 with a hand‐held time‐domain reflectometry sensor. To model soil moisture and its temporal variation, we used four statistical modelling methods: generalized linear models, generalized additive models, boosted regression trees, and random forests. The model fit of the soil moisture models were R2 = 0.60 and root mean square error (RMSE) 8.04 VWC% on average, while the temporal variation models showed a lower fit of R2 = 0.25 and RMSE 13.11 CV%. The predictive performances for the former were R2 = 0.47 and RMSE 9.34 VWC%, and for the latter R2 = 0.01 and RMSE 15.29 CV%. Results were similar across the modelling methods, demonstrating a consistent pattern. Soil moisture and its temporal variation showed strong heterogeneity over short distances; therefore, soil moisture modelling benefits from high‐resolution predictors, such as LiDAR based variables. In the soil moisture models, the strongest predictor was SAGA (System for Automated Geoscientific Analyses) wetness index (SWI), based on a 1 m2 digital terrain model derived from LiDAR data, which outperformed soil predictors. Thus, our study supports the use of LiDAR based SWI in explaining fine‐scale soil moisture variation. In the temporal variation models, the strongest predictor was the field‐quantified organic layer depth variable. Our results show that spatial soil moisture predictions can be based on soil and land surface properties, yet the temporal models require further investigation. Copyright © 2017 John Wiley & Sons, Ltd. 相似文献
8.
Determinants of sediment properties and organic matter in beach and dune environments based on boosted regression trees
下载免费PDF全文
![点击此处可从《地球表面变化过程与地形》网站下载免费的PDF全文](/ch/ext_images/free.gif)
Grain size properties and the variation of organic matter in coastal beach and dune environments are assumed to be controlled by the intensity of aeolian processes, time and the sediment source. However, assumptions are based on relatively limited empirical studies. In this study, we examined which environmental variables are the main predictors of multiple topsoil properties. To achieve this, we analysed an extensive dataset systematically collected across all beach zones and a large geographical area at the Finnish Baltic Sea coast characterized by post‐glacial land uplift. We included a comprehensive set of predictors in the analysis and applied boosted regression trees, a modern modelling technique particularly suited for analysis without prior assumptions of the data model. The results suggest that mean grain size and sorting are mainly determined by northing and fetch. Northing, disturbance and fetch predicted the variation of soil organic matter while litter cover was strongly related to disturbance. Based on the analyses, we were able to identify the main drivers of multiple topsoil properties on land uplift beaches. Parent material is suggested to determine sediment textural properties, which largely masks the effects of transient processes. Mean grain size and sorting are highly interdependent: grains become finer and sorting improves with increasing shore exposure. The intensity of momentary geomorphic processes controls the accumulation of litter whereas the slower accumulation of organic matter in the soil is influenced also by the static exposure setting. Skewness and kurtosis of the grain size distribution are mainly influenced by unmeasured processes, potentially relating to the geomorphological origin of the sediment. Copyright © 2014 John Wiley & Sons, Ltd. 相似文献
9.
Tomi P. Luoto Liisa Nevalainen Tommi Kauppila Mira Tammelin Kaarina Sarmaja-Korjonen 《Quaternary Research》2012,78(2):248-255
A sediment core from Lake Arapisto, Finland, was examined for fossil diatom assemblages to reconstruct changes in Holocene nutrient availability. Our aim was to investigate the long-term relationship between lake trophic status and climate by comparing the diatom-based phosphorus reconstruction with paleoclimatic proxies. Our results showed that the cold early Holocene was characterized by elevated nutrient conditions concurrent with newly exposed fertile ground. As the climate rapidly warmed and ice sheet further retreated, the catchment vegetation developed, which resulted in decreased nutrient flux into the lake. The Holocene Thermal Maximum (HTM), between ~ 8000 and 4000 cal yr BP, was characterized by oligotrophic conditions, which may have been caused by low effective precipitation and stable watershed vegetation. After the HTM, the lake became more productive. There was no particular increase in the trophic state that could be connected to more recent human influence. Although lake productivity has been shown to be affected by temperature, our record indicated that the nutrient dynamics were driven by complex interactions between changes in temperature, precipitation, catchment, and in-lake processes. Understanding of long-term nutrient dynamics and the associated processes can help in resolving relationships between lake productivity and climate during past and present climate changes. 相似文献
10.
Numerical studies of earth surface processes in relation to their environment are one of the central topics in physical geography. However, collinearity between explanatory variables and spatial autocorrelation can hinder the detection of key environmental correlates underlying response–explanatory variables' relationships identified by traditional regression methods. Moreover, conclusions about the potential importance of environmental variables have generally made on analysis conducted only at one spatial scale (resolution). In this study, a variation partitioning method provided a framework to obtain new insights into the relative roles of different factors determining patterned ground activity at multiple spatial resolutions. The variation in the distribution of the sorted and non sorted patterned ground was decomposed into independent and joint effects of relief, soil and spatial variables (geographical location) based on a multi resolution system of 1 ha, 25 ha and 100 ha cells covering in total 100 km2 of a subarctic landscape in northern Finland.The independent effects of relief and soil variables captured the largest fraction of the variation in the non-sorted patterned ground distribution, while relief had a major contribution for sorted patterned ground activity. The independent effect of spatial autocorrelation on sorted patterned ground was higher than that on non-sorted patterned ground. However, a considerable amount of variation in the distribution of both patterned ground types was accounted for by the joint effects of explanatory variables and may thus be causally related to two or all three groups of variables. Our analyses produced often contradicting results at different resolutions. Consequently, this has substantial implications for the study of geomorphological systems, since the choice of resolution can have a major effect on the inferences of analyses. Our results draw attention to the roles of resolution and spatial autocorrelation in the study of geomorphological systems. 相似文献