Search

CN-122024065-A - Unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation

CN122024065ACN 122024065 ACN122024065 ACN 122024065ACN-122024065-A

Abstract

The invention discloses an unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation, which relates to the technical field of grassland ecological system monitoring, and comprises the steps of collecting standardized time-space sequence data through an unmanned aerial vehicle multispectral system; the method comprises the steps of processing data to generate a surface coverage distribution map and a three-dimensional model, quantifying landscape patterns and soil erosion indexes in two-dimensional and three-dimensional spaces based on the model, coupling the quantified indexes with an ecological process model, evaluating influences on species communication, soil preservation and carbon storage functions, and finally carrying out trend analysis and visual expression. The method solves the technical problems that the traditional method is low in efficiency, insufficient in satellite remote sensing resolution, difficult to carry out three-dimensional stereo monitoring and incapable of quantitatively correlating the pattern change with the ecological function influence, thereby realizing efficient, accurate and three-dimensional quantitative grassland crushing dynamics and scientifically evaluating the ecological results.

Inventors

  • ZHANG QING
  • ZHAO YANYUN

Assignees

  • 内蒙古大学

Dates

Publication Date
20260512
Application Date
20260203

Claims (10)

  1. 1. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation is characterized by comprising the following steps of: Step one, acquiring space-time sequence data of a grassland monitoring area through an unmanned aerial vehicle multispectral monitoring system, wherein the space-time sequence data comprises the steps of arranging ground control points in the monitoring area and measuring coordinates of the ground control points; Selecting a grassland key growth stage for flying, and keeping the flying parameters consistent; correcting the acquired multispectral image, and constructing a three-dimensional model of a research area by utilizing the overlapped image to generate a digital surface model, a digital terrain model and a vegetation canopy height model; Calculating a vegetation index based on the corrected image, and performing surface coverage classification to generate a surface coverage distribution map; calculating landscape space pattern quantization indexes including plaque density, maximum plaque index, edge density, average plaque area and polymerization degree index on a two-dimensional plane based on the surface coverage distribution map; Carrying out grassland height layering on a three-dimensional space by utilizing the vegetation canopy height model, and calculating landscape pattern indexes of each level; monitoring the soil erosion process by comparing the canopy height models of different periods; step four, coupling the quantization index with an ecological process model, and evaluating the influence of grassland fragmentation on species communication, soil retaining function and carbon storage function; and fifthly, carrying out trend analysis on the multi-period indexes and generating a comprehensive atlas.
  2. 2. The method for monitoring the multi-spectrum of the unmanned aerial vehicle for dynamic analysis of the crushing of grasslands according to claim 1, wherein for the first step, the space-time sequence data of the grassland monitoring area is collected by the unmanned aerial vehicle multi-spectrum monitoring system, and the method further comprises: When unmanned aerial vehicle flight mission is executed, the ground on-site investigation is synchronously carried out, and vegetation community structure, growth condition and soil surface state information are measured and recorded in a pre-divided ground sampling land block.
  3. 3. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation according to claim 1, wherein the correction processing of the collected multispectral images is performed for the second step, including eliminating distortion caused by the multispectral image acquisition device itself, atmospheric conditions and topography fluctuation.
  4. 4. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation of claim 1, wherein for the third step, grassland height layering is performed by using a vegetation canopy height model in a three-dimensional space, specifically, a height threshold is set to divide a grassland area into a high grassland layer, a medium grassland layer and a low grassland layer.
  5. 5. The unmanned aerial vehicle multispectral monitoring method of dynamic analysis of grassland fragmentation according to claim 1, wherein for step four, evaluating the impact of grassland fragmentation on species communication, specifically comprises: the grassland patches are regarded as nodes in the ecological network, and a connection distance threshold is set according to the mobility of the target species; if the distance between the two grassland patches is smaller than a threshold value, establishing connection, and further constructing an ecological connectivity network; By analyzing the structural changes of the ecological connectivity network, the key pedal grassland patches and the potential ecological corridor are identified.
  6. 6. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation according to claim 1, wherein for step four, evaluating the impact of grassland fragmentation on soil holding function, specifically comprises: Combining the bare land information identified in the surface coverage distribution map with the terrain gradient information, and inputting the combined bare land information and the terrain gradient information into a soil erosion estimation model as parameters to quantitatively estimate the soil erosion amount on each image unit; and comparing and analyzing the total soil erosion amount change of the areas before and after the grassland is crushed.
  7. 7. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation according to claim 6, wherein the soil erosion estimation model comprehensively considers rainfall erosion force, soil corrosiveness, gradient slope length, coverage and management factors and soil and water conservation measure factors.
  8. 8. The method of unmanned aerial vehicle multispectral monitoring for dynamic analysis of grassland fragmentation according to claim 1, wherein for step four, evaluating the impact of grassland fragmentation on carbon storage functionality comprises: Establishing a relation among vegetation indexes, vegetation heights and actually measured biomass on the ground, calculating biomass and carbon reserves of grassland vegetation in a region, analyzing spatial distribution changes of the carbon reserves in different periods, and quantitatively evaluating the loss of carbon fixing capacity of an ecological system caused by crushing and degradation of grasslands.
  9. 9. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation according to claim 1, wherein aiming at the fifth step, trend analysis is performed on the multi-stage index, and the method specifically comprises the following steps: Judging the overall trend and the change rate of the crushing of the grasslands; core areas and critical periods of intense variation are located.
  10. 10. The unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation according to claim 1, wherein for the fifth step, the generated comprehensive atlas specifically comprises a grassland fragmentation space-time evolution diagram, an ecological connectivity change diagram, a soil erosion risk diagram and a carbon sink function damage diagram.

Description

Unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation Technical Field The invention relates to the technical field of grassland ecological monitoring, in particular to an unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation. Background The crushing of the grasslands is a key sign of the degradation of the grasslands, and is characterized in that the complete grasslands are divided into isolated and fine patches, which not only directly lead to the reduction of the grassland area, but also cause a series of deep ecological problems such as the loss of habitat of species, the reduction of ecological connectivity, the aggravation of water and soil loss, the decline of carbon sink function and the like. At present, the monitoring of grassland conditions mainly depends on traditional artificial ground investigation and satellite remote sensing technology. Although the traditional ground investigation method can acquire accurate ground data, the method is time-consuming, labor-consuming, high in cost and capable of covering limited points, quick and comprehensive assessment on the area scale is difficult to realize, and particularly, the implementation difficulty is higher for areas with complex terrains and rare track. Meanwhile, the method is difficult to capture continuous change of the grassland spatial pattern, and the dynamic process of crushing cannot be effectively quantified. While satellite remote sensing technology can provide large-scale observation data, the space resolution is often limited, and key details such as small-scale grassland plaque boundaries, tiny erosion ravines and the like are difficult to accurately identify, so that the satellite remote sensing technology is not sensitive enough to monitoring in the early stage of fragmentation. In addition, satellite observation period is fixed, is difficult to flexibly adjust the observation time according to the key weather period of the grasslands, is easily influenced by cloud cover, is difficult to acquire continuous and high-quality time sequence data, prevents deep analysis of the relevance of the grassland growth rhythm and the crushing process, and focuses on analysis of the two-dimensional plane information of the ground surface, however, grassland crushing is not only reflected on the change of the plane pattern, but also accompanied with the change of vertical structure information such as the three-dimensional form change of the ground surface caused by vegetation height reduction and soil erosion, and cannot accurately quantify the three-dimensional form of the ground surface, so that the acquired grassland data is not comprehensive enough. Disclosure of Invention The invention aims to provide an unmanned aerial vehicle multispectral monitoring method for grassland fragmentation dynamic analysis, multi-phase high-resolution multispectral images and ground verification data are acquired through standard unmanned aerial vehicle flight, an accurate ground surface coverage classification chart and a ground surface three-dimensional model are generated after processing, further, the scheme quantifies the landscape pattern index and soil erosion change of grasslands from two dimensions of a two-dimensional plane and a three-dimensional space, the influence of grassland fragmentation on key ecological functions such as species migration, water and soil conservation and carbon storage capacity is quantitatively evaluated through a coupling ecological process model, finally, the dynamic process of grassland fragmentation and the ecological consequences thereof are comprehensively revealed through trend analysis and visual expression of multi-phase data, unmanned aerial vehicle remote sensing, three-dimensional modeling and ecological assessment are integrated, the comprehensive monitoring of the grassland fragmentation from a macroscopic pattern to a microscopic mechanism and from a plane to a three-dimensional process is realized, the limitations of traditional ground investigation and satellite remote sensing are overcome, the degradation track of the grassland ecological system can be effectively and accurately revealed, the space pattern change and the specific ecological function damage are directly associated, the ecological decision making method is provided, and the technical problem of the ecological system is effectively solved. In order to achieve the above purpose, the present invention provides the following technical solutions: the unmanned aerial vehicle multispectral monitoring method for dynamic analysis of grassland fragmentation is characterized by comprising the following steps of: collecting space-time sequence data of a grassland monitoring area through an unmanned aerial vehicle multispectral monitoring system; processing and extracting the acquired data to generate a surface coverage distribution map and a three-dimensi