| Agriculture provides the largest share of food supplies and other critical ecosystem services needed not only for humans but also for other biota.Rice(Oryza sativa)is the most consumed staple food in the world,especially in Asian countries where at least 90% of the rice is cultivated and consumed.For instance,in China,rice has been reported as the primary exposure source to metals and metalloids.Rice is also known to be particularly sensitive for mercury(Hg)biomagnification because of essential water requirement for its growth and development.In contrast,the flow of metals and metalloids through other agricultural ecosystems has been less studied.Sugarcane(Saccharum officinarum)is another prominent agronomic crop cultivated in southern China.Since sugarcane is grown in areas with water scarcity where irrigation is minimal or limited,sugarcane ecosystems are an ideal comparison to rice to understand differences in metal and metalloid flow or flux in various agronomic crops.Moreover,investigating the contamination sources of multiple metals/metalloids between rice paddies and sugarcane fields,and identifying the quantities/proportions of each contamination source(i.e.source apportionment)is a vital process in addressing the metal/metalloid contamination.To that end,multiple metals and metalloids,including arsenic(As),cadmium(Cd),chromium(Cr),zinc(Zn),copper(Cu),lead(Pb),manganese(Mn),and total mercury(THg)were examined at a lead-Zinc(PbZn)mine tailings site,and at two reference sites in southwestern Guangxi,southern China.The first main objective of this dissertation was to compare metal and metalloid contamination and patterns of biomagnification between the two dominant agroecosystems in southern China,i.e.rice paddies and sugarcane fields,which are also common agronomic crops in other tropical countries.Along with this objective,three fundamental research questions were asked: a)For what metals and metalloids was there contamination at the mine site? b)How were metal and metalloid residues different between rice paddy and sugarcane agroecosystems? c)In what metals/metalloids and agroecosystems was there indication of biomagnification patterns? To address this objective with its associated research questions,samples were collected in both rice paddy and sugarcane agricultural ecosystems at the Pb-Zn mine tailings site and the two reference sites.The collected samples were broadly spread across the trophic levels in food web,comprising the soil(300 samples),primary producers(i.e.rice grains [37 samples] and leaves [129 samples],sugarcane leaves [194 samples]),herbivorous invertebrates(i.e.grasshoppers [320individuals] and crickets [188 individuals]),invertebrate insectivores(i.e.spiders [522 individuals]),and vertebrate insectivores(such as frogs [161individuals]).To answer the first research question,eight different ANOVA(analysis of variances)tests(herein,one for each metal and metalloid)were carried out independently for soil and for each collected biological species in every agroecosystem,comprising sampling site(three levels: the mine site,the close reference site,and the far reference site)as an explanatory variable.ANOVA was then followed by Tukey Honest Significant Difference(HSD)multiple comparisons.The results showed that Pb and Zn had larger contamination levels,as would be expected at a Pb-Zn mine tailings site,each of which had elevated residues at the mine site compared to the two reference sites in both agricultural ecosystems in soil and in many living organisms.Other metals also had some evidences of higher contamination at the mine site,including cases of Cd,Cr,and THg.The higher residues of some of these secondary metals may be ascribed to the existence of mine tailings.To answer the second research question,the metal and metalloid residues were compared between the two agroecosystems at a site for soil,as well as for the living species that these two agroecosystems had in common,with two-sample t-tests for every metal and metalloid.The findings revealed that,of the eight studied metals and metalloids,seven(excluding Mn)had elevated residues in rice paddy soil compared to sugarcane soil in at least two of the three study sites.Likewise,five metals,including Cu,Cr,Cd,Pb,as well as,Zn were higher in species sampled in rice paddy than those in sugarcane agroecosystems in at least two species,or two study sites.These findings signify that rice paddy agroecosystems are particularly sensitive to contamination of multiple metals and metalloids potentially because of irrigation or centrality of water for their growth and development,as well as excessive use of chemical fertilizers,particularly nitrogenous(N)fertilizers.To answer the third research question,bioaccumulation factors(BAFs)were firstly computed for every animal and plant species,as the residues of metals and metalloids in tissues of animal and plant species over the residues of metals and metalloids in the surrounding medium/environment,which was the average residues in soil at every sampling area.Then BAFs were compared among biological species,with ANOVA tests for every metal and metalloid in each agroecosystem at every site,followed by multiple comparisons,through the Tukey HSD technique.The results revealed that biomagnification patterns were more broadly distributed than just THg,which is clearly recognized to biomagnify in food chains.The strongest patterns of biomagnification were found for THg and Zn,while intermediary patterns of biomagnification were as well observed for Cd,Cr,Cu,as well as Pb;biodilution patterns were noticeable for As and Mn.When quantitatively grading biomagnification patterns through post-hoc multiple comparisons among biological species’ BAFs,rice paddy agroecosystems had more pronounced biomagnification patterns(3 strong relationships [i.e.all insectivorous species had larger than or as large BAFs as all other living species],7 medium relationships [i.e.one of the insectivorous species had the largest BAFs,but the other insectivore had an intermediate BAFs,similar to the herbivorous species],6 weak relationships[i.e.one of insectivorous species had the largest BAFs,while the other insectivore had lower BAFs,similar to the primary producers],and 3 no relationships [i.e.no statistically significant variations among biological species in BAFs])compared to sugarcane agroecosystems(1 strong,2 medium,8 weak,9 no relationships;P-value of a repeated measures t-test = 0.0023)across 6 metals(Cr,Cd,Cu,Hg,Pb,and Zn).This could be because there were larger residues of multiple metals there;biomagnification patterns would have not been noticeable if there were very small metal residues in the environment.The second main objective of this dissertation was to determine the possible pollution sources of various metals and metalloids,and quantify the amount or proportions that every possible source contributes,comparing between rice paddy,as well as sugarcane agricultural soil of southwestern Guangxi,southern China.To address this objective,methods similar to the first main objective were used to collect and analyze metals and metalloids in soil samples.Then,the Positive Matrix Factorization(PMF)receptor model,was used to apportion the sources of metals and metalloids and their contribution rates.The PMF receptor model results demonstrated that at the mine site in rice paddy agroecosystems,mining activities seem to have contributed a greater proportion(60.7%)of metals/metalloids,followed by irrigation water(25.8%),as well as agrochemical application(13.5%).At rice paddies in the close reference site,agrochemical application appeared to have the greatest contribution(42.8%)of metals/metalloids,followed by irrigation water(22.7%),natural sources(17.4%),as well as mining activities(17.2%).At rice paddies in the far reference site,agrochemical application seemed to have the most contribution(40.6%)of metals/metalloids,followed by irrigation water(32.5%),as well as natural sources(26.9%).In contrast,for sugarcane ecosystems,at the mine site,as well as close reference sites in sugarcane agricultural ecosystems,agrochemical application appeared to have the greatest contribution(50.1% and 57.4%,respectively)of metals/metalloids,followed by mining activities(49.9% and 42.6%,respectively).At the far reference site in sugarcane agricultural ecosystems,agrochemical application appeared to be the predominant source factor(51.2%)of metals/metalloids,followed by natural sources(48.8%).In summary,rice paddy agroecosystems had higher contamination levels of various metals in soil and in many living organisms compared to sugarcane agroecosystems.Biomagnification patterns of metals were more pronounced in rice paddy than in sugarcane agroecosystems.Furthermore,the PMF receptor model suggested that the best solution was four or three source factors of metals/metalloids for rice paddy agroecosystems at a particular site,and four source factors being identified at all three sites.In comparison,in sugarcane agroecosystems,only two source factors of metals/metalloids were observed at a particular site,with three source factors determined across all three sites.These findings concurred with the expectation that sources of metal/metalloid contamination may be more complicated in rice paddy agricultural ecosystems receiving irrigation water compared to sugarcane agricultural ecosystems that are not irrigated.Therefore,the results in this dissertation expand our knowledge on the complexity of agroecosystems and on the levels and pathways of metals/metalloids in agroecosystems.In addition,these results have important practical implications for prevention and control of metal/metalloid pollution in agricultural ecosystems. |
PDF Full Text Request