Fractal And Geostatistical Researches On Spatial Patterns Of Macrophytes In Lakes

The spatial patterns of macrophytes on different levels were studied by usingFractal Geometry and Geostatistics. Those levels included leave, branch, individual,population, community, a bay, the whole lake (Baoan), lake districts, and region. Thefractal character of different lakes' shorelines was studied on certain principles ofLandscape Ecology. A primary estimate on the services value of Baoan Lakeecosystem was made. The main contents of research included: the spatial pattern ofmodule and individual of some dominant macrophytes in the Baoan Lake, fractalcharacter of horizontal pattern of Nelumbo nucifera Association in the lake,macrophytes community pattern in a bay (Huangfengkou region) of Baoan Lake, theaquatic vegetation spatial pattern and its dynamics of Baoan Lake in 1992 and 1996, thefractal character of some lakes' shorelines in Hubei Province, and the services value ofthe Baoan Lake ecosystem. The research dealt with spatial patterns of macrophytes atdifferent scales, which enables the maximum extraction of ecological information, andprovides theoretical foundation for the optimal utilization and protection ofmacrophytes resources.The main results are listed as follows:1. The fractal dimension (FD) of Potamogeton malaianus leaves was between 1.322and 1.520, with an average of 1.441, and a standard deviation of 0.0547. The FDof Potamogeton crispus leaves was between 1.381 and 1.565, with an average of1.472, and a standard deviation of 0.0489. The FD of Najas marina leaves wasbetween 1.209 and 1.406, with an average of 1.325, and a standard deviation of0.0547. The FD of Hydrilla verticillata leaves was between 1.001 and 1.217, withan average of 1.103, and a standard deviation of 0.0456. The sequence of averageFD of leaves was: Potamogeton crispus >Potamogeton malaianus> Najas marina>Hydrilla verticillata, which indicated the difference in the ability of spatialoccupying and utilizing among different species' leaves.2. The FD of Hydrilla verticillata branches was between 1.308 and 1.408, with anaverage of 1.359, and a standard deviation of 0.0409. The FD of Ceratophyllumoryzetorum branches was between 1.594 and 1.678, with an average of 1.631, and astandard deviation of 0.0359. The FD of Najas marina branches was between1.464 and 1.535, with an average of 1.490, and a standard deviation of 0.0265.The sequence of average FD of branches was: Ceratophyllum oryzetorum> Najasmarina> Hydrilla verticillata, which indicated the difference in the ability of spatialoccupying and utilizing among different species' branches.3. The FD of Vallisneria spiralis individual pattern was between 1.249 and 1.430,with an average of 0.0619, and a standard deviation of 0.0619. The FD ofCeratophyllum oryzetorum individual pattern was between 1.185 and 1.491, withan average of 1.339, and a standard deviation of 0.1009. The FD of Myriophyllumspicatum individual pattern was between 1.175 and 1.396, with an average of 1.290,and a standard deviation of 0.0757. The sequence of average FD of branches was:Vallisneria spiralis>Ceratophyllum oryzetorum> Myriophyllum spicatum, whichindicated the difference in the ability of spatial occupying and utilizing amongdifferent species' individuals.4. The fractal character of horizontal pattern of Nelumbo nucifera Association showedthe difference between two macrophytes' distribution in the association. Nelumbonucifera was the dominant species in August. The box-counting dimension (BD)of distribution of Nelumbo nucifera was 1.92, with an information dimension (ID)of 1.88;the BD of distribution of Trapa insica was 1.04, and an ID of 1.11. Thisresult showed that Nelumbo nucifera uniformly distributed at the given scales from0.5 to 8 m. The variety of Nelumbo nucifera 's FD in the sequent samples wasdiscussed.5. The semivariogram of the variables on lengthwise transect showed that: from 0 to980 m, most variables demonstrated a significant correlation, with the exception ofMyriophyllum spicatum and Hydrilla verticillata. From 0 to 400 m, every variablehad a fine linear relationship with separation distance. From 400 to 980 m, thecorrelation coefficient of some variable increased, and others decreased, whichimplied that there existed a divergence in these variables at this scale. The turningpoints of Nymphoides peltata and Trapa bispinosa semivariogram were distinct, thedistance of turning points occurring at 300 and 460 m respectively.6. The semivariogram of the variabls on transverse transect showed that: from 0 to660 m, Ceratophyllum oryzetorum, Myriophyllum spicatum, Nymphoides peltataand the community as a whole had a fine linear fit. From 380 to 660 m,Ceratophyllum oryzetorum, Potamogeton maackianus and Nymphoides peltata hada fine linear fit, while Myriophyllum spicatum and the community as a wholedemonstrated a decreased correlation coefficient. Potamogeton maackianus had aturning point at 220 m, and the community had a turning point around 520 m.7. The transect investigation revealed that: every macrophyte species and the wholemacrophyte community had a fractal dimension between 1.5 and 2, which indicatedthat certain spatial heterogeneity existed at different scales. By comparison of thetwo transects' fractal dimension, spatial heterogeneity of the lengthwise transectwas higher than transverse transect, implying that certain environmentalheterogeneity existed in the two transects. The transect investigation alsoindicated that when we carried out an submersed macrophyte communityinvestigation in the Baoan Lake, the scale 520 m should be adopted, while 380 mshould be considered as a single species investigation scale.8. In the littoral macrophytes community, proportion of spatial structure of Vallisneriaspiralis and Myriophyllum spicatum were high, which indicated that theirdistribution changed little at microscale (less than 5 m). Their isotropic modelswere exponential, which indicated their contagious distribution. The ranges oftheir models were 48.6 and 34.2m respectively. In their range, they had spatialautocorrelation. Beyond the ranges, their autocorrelation decreased. Isotropicmodels of Ceratophyllum oryzetorum and the community were linear, theirproportions of spatial structure were low, which indicated their great change atmicroscale and their moderate contagious distribution.9. The quadrat investigation indicated that in the littoral macrophytes community, thedistribution of Vallisneria spiralis in all-direction had no significant difference, butthe highest regression coefficient was in the direction of NE-SW, so its dominantpattern was in the direction of NE-SW. The dominant pattern of Myriophyllumspicatum and Ceratophyllum oryzetorum was in the direction of N-S. Thecommunity indicated a dominant pattern in the direction of NW-SE.10. By carrying out transect investigation in the Baoan Lake from June to October in1996, the following results were obtained: 1) Distribution area of emergentmacrophytes was low, and that of submersed macrophytes was high. 2) There werefew dominant species of submerged plant, including Myriophyllum spicatum,Vallisneria spiralis, Ceratophyllum oryzetorum and Potamogeton maackianus. 3)Comparative biomass of Potamogeton maackianus was higher than any otherspecies and reached 70.18%. The other three species (Vallisneria spiralis,Ceratophyllum oryzetorum and Myriophyllum spicatum) constituted 6.012, 12.18and 10.934%, respectively. 4) Compared with the composition of submersedmacrophytes in 1992, despite of the high comparative biomass of Potamogetonmaackianus that year (58.472%), the 1996 record (70.18%) was even higher. Andthe other three species (Vallisneria spiralis, Ceratophyllum oryzetorum andMyriophyllum spicatum) constituted 11.912, 8.714 and 14.506%, respectively. Ina word, the composition of aquatic vegetation was drab, the biodiversity ofsubmersed macrophytes was low, and the competition among dominant species wastense in the Baoan Lake.11. By comparison of BD of patterns corresponding to different biomass ranges ofPotamogeton maackianus in the Baoan Lake from June to October in 1992 and1996, some results are summarised as follows: 1) the growth of Potamogetonmaackianus in June of 1992 was inferior to that in the same month of 1996, and itsdistribution area in June of 1992 was smaller than that in June of 1996. 2) In Julyof 1992, it grew better than it did in the same month of 1996;a high biomass(higher than 7000g/m2) was well maintained. A biomass pattern (lower than5000g/m2) was well maintained in July of 1996. 3) In next two month of 1992and 1996, the BD of their biomass pattern exhibited no substantial difference. 4)Potamogeton maackianus in October of 1992 grew better than it did in the samemonth of 1996, its distribution area was larger than that in 1996, and the area ofhigh biomass was also larger than that in 1996. 5) By analzing of biomass patternin Baoan Lake in 1992, the area of low biomass was stable from June to October,and the area of high biomass was greatly developed. But in 1996, the distributionarea of Potamogeton maackianus was steadily reduced.12. By comparison of BD of patterns corresponding to different macrophytes biomassranges in the Baoan Lake from June to October in 1992 and 1996, the basicdynamics of macrophytes biomass was similar as that of Potamogeton maackianus.From June to October in 1992, the area of low biomass was stable, and the area ofhigh biomass greatly developed. But in 1996, the distribution area of macrophyteswas steadily reduced.13. The average of shoreline development indexes (SDI) of selected 31 lakes was 4.50,with a standard deviation of 2.088;the Shangjin Lake had the highest SDI of 11.33,and the Wangmu Lake had the smallest SDI of 1.68. The average of shorelinebox-dimension (SBD) of the 31 lakes was 1.219, with a standard deviation of 0.078;the Shangjin Lake had the highest SBD of 1.419, and the Dachang Lake had thesmallest SDB of 1.096.14. By using the equation of perimeter-area and its transform, the 31 lakes had a wholeSBD of 1.3998 and 1.6956, respectively. The value of 1.3998 was close to themean of SBD (1.219). Two regression coefficients (R2) of two linear relationshipswere higher than 0.80, which indicated that the shoreline of these lakes exhibited aself-similarity.15. Based on Korcak's Law and the area distribution of lakes in Hubei Province, theSBD of lakes in Hubei Province in 1980s was 1.52. The result showed that thelakes in Hubei Province had a self-similarity and complex shorelines.16. The stepwise regression of macrophytes biomass in some lakes of Hubei Provincevs four morphological indexes (SDI, SFD, Area, Perimeter) was: Biomass =45625.0 + 2746.0·SDI -46277.2·SFD (R2 = 0.6024), in which the SDI is shorelinedevelopment index and SFD is shoreline fractal dimension. From the equationabove, the SDI and SFD are the influencing factors of the biomass.17. The Baoan Lake ecosystem services and its indirect economic values wereestimated on the basis of ecological function analysis and some economic methods.The study showed that the indirect economic values (in RMB) of organic matterproduction, CO2 fixation, O2 release, nutrient recycling, water conservancy andwater supply and SO2 degradation were 1.51×107, 4.77×107, 3.63×107, 4.45×105, 2.30×108, 5.15×105Yuan/a, respectively. And the sum is 3.30×108Yuan/a.From the incomplete estimation above, we have known that the Baoan Lake hasgreat ecological and economic values.