Font Size: a A A

Spatial-temporal Evolution Processes And Driving Forces Of Karst Rocky Desertification In The Typical Karst Valley Area

Posted on:2019-12-14Degree:MasterType:Thesis
Country:ChinaCandidate:F ChenFull Text:PDF
GTID:2381330566468494Subject:Physical geography
Abstract/Summary:PDF Full Text Request
Karst rocky desertification is the most serious ecological disaster in southwestern China.It has established a highly efficientandaccurateinterpretationmethodforrock desertification and quantitatively evaluates its evolutionary history and driving forces.It is very important and urgently needed for ecological restoration and sustainable development of the region.This paper addresses the deficiencies of past interpretation methods for rocky desertification.Using 1990,1995,2000,2004,2011TM?Thematic Mapper?remote sensing images and 2016 OLI?Operational Land Imager?remote sensing images as data sources,based on lithologic background removal of non-karmic regions and land use types for extraction of water bodies,construction land,etc.Rocky desertification may occur in remote sensing imagery pre-processing and classification using CART?a Classification And Regression Tree?decision tree,and a method of rock desertification interpretation with high accuracy and high efficiency is established.For the first time,the geo-detector was introduced into the study of the driving force of rocky desertification.Quantitative analysis of factors such as lithology,slope,permanent population density andaverageannualprecipitationcontributedtothe desertification drive.The results show that:?1?This method avoids The traditional methods of“rocky desertification lands often occur in non-karst,paddy fields,rivers and lakes,and urban road construction land”often occur at the same time;it is possible to reduce the time-consuming error of understanding the translation time and labor costs,and to reduce human error.The possibilities are economical,fast,and efficient.?2?The rocky desertification in the study area changes with the evolution rate of 14.18km2·a-1during the study period,showing the evolutionary process of“first deterioration and improvement”,of which 2004 is an important transition year;The main method of change.?3?The greatest impacts on the formation and evolution of rocky desertification are lithology?0.35?and permanent population density?0.30?,and the driving contribution rates are 30.17%and 25.86%,respectively.It shows that the lithologic basement is the largest natural background for the formation and evolution of rocky desertification,and the density of the resident population is the most human-driven.In the futurescenario simulationof karst rocky desertification,this paper used the Cellular Automata?CA?-Markov model to simulate the spatio-temporal changes of rocky desertification in the typical karst troughs in southwest China in 2011-2021.The results show that:?1?Using the cellular automata model to simulate the spatial distribution of rocky desertification,the Kappa verification accuracy of all types of rocky desertification is above 0.5,which meets the theoretical requirements and can accurately reflect the distribution of rocky desertification in the future.?2?Combining geo-detectors with CA-Markov models can more accurately predict the future evolution trend of rocky desertification,reduce the arbitrariness of human subjective selection factors,and reduce the possible misjudgment in predicting rocky desertification scenarios.It's possible to ensure prediction accuracy.?3?With the overall improvement of the rocky desertification,there are still a few cases where rocky desertification increases and worsens.In the process of restoration and management,the changes in the ecological environment in these deteriorated areas must be taken into account.In summary,the karst rock desertification interpretation method based on the lithologic background and land use map and then using the CART decision tree classification method can provide a method reference for peers in the study of the spatial andtemporalevolutionofrockydesertification.The spatio-temporal evolution data of rocky desertification in the typical karst trough area from 1990 to 2016 and the spatial and temporal evolution data of rock desertification in 2021predictedbyCA-Markovcanprovidethelatestrock desertificationdatareferenceforpeerresearchers.Quantitative analysis of rocky desertification based on geo-detectors can better explore the mechanism of occurrence and development of rocky desertification,and provide basis for better formulation of rock desertification control measures in the vast areas affected by rocky desertification disasters.
Keywords/Search Tags:Karst, rocky desertification, Spatial and temporal evolution, Geodtector, Driving force, Cellular Automata, Markov
PDF Full Text Request
Related items