Soil chemical and microbiological properties affected by land use, management, and time since deforestations and crop establishment: the case of wheat (not irrigated) and cotton (irrigated) fields in Filyria, Kilkis, Greece

Abstract

Deforestation is a common practice in many countries worldwide since manyyears, in order to gain land for agricultural purposes, and is one of the main reasonsfor the global net release of CO2 from soil to atmosphere. The conversion fromnatural vegetation to cropland often leads to a depletion of the soil organic carbon(SOC) stock due to reduced input of biomass and enhanced decomposition afterphysical disturbance. The effect of forest clearing for other land uses is often a lowlevel of soil microbial population.In a part of Northern Greece, Filyria, which belongs to prefecture of Kilkis, threedeforestations took place in 1933, 1971 and 1980, in order to give land for agricultureto refugees and residents with no property. The natural vegetation consists mainly ofQuercus pubescens stands. Each family was given 2-9 hectares of deforested land.The agricultural fields are cultivated with wheat, cotton and tobacco, and cherry treesare established in several plots. The soils of the study area have developed fromlimestone, and are classified as Xeralfs according to Soil Taxonomy. The plots of thestudy are cultivated with cotton and wheat. Wheat fields are not irrigated and theirmanagement is considered as reduced tillage, whereas cotton plots are irrigated bysprinklers, considered as conventional tillage and are altered with wheat every 2 years(two consecutive years of cotton crops, one year of wheat). The big amount of timetranscurred since the first deforestation, the availability of three different deforestationdates and the remaining undisturbed forest, offer an interesting opportunity ofstudying these changes, along with assessing factors such as land use (forestry vsagriculture) and land management types (wheat fields vs cotton fields) correlatingwith the amount of time since deforestation (26, 35 and 73 years).Soil microbial and biochemical status in both natural and agricultural ecosystemshave been used as bioindicators of soil stress or benefits from reclamation efforts. Inaddition to chemical and physical properties, soil microbial properties may indicatesuitable management and restoration practices towards sustainability. Sincemicroorganisms respond rapidly to changing environmental conditions they areconsidered as sensitive indicators of soil health and could be therefore used for soilstatus monitoring. Organic carbon and nitrogen in the soils, have different behaviourin the different soil fractions, where clay, mainly, and silt seem to preserve better thecontents of the two elements. The aims of this thesis are to determine the changes interms of soil microbiological properties that resulted from 3 stages of deforestation(1933, 1971, 1980), by measuring and assessing microbial biomass carbon (MBC),SOC, the MBC/SOC ratio, soil respiration (SR), the metabolic quotient (qCO2), andthe relationship between SOC, N and the soil fractions. Land use (forestry vsagriculture), type of crop management (wheat vs cotton) and the amount of time sincedeforestation (26, 35 and 73 years since deforestations) are the factors discussed.Three sampling categories referring to deforestation year (1933, 1971, and 1980)in the cultivated fields and one category referring to the natural remaining forest wereconsidered. For each deforestation year category, two crop subcategories wereconsidered (wheat and cotton), and from each crop subcategory, 12 plots weresampled. In adjacent areas, twelve plots of remaining undisturbed forest (Quercuspubescens) were selected. A composite sample was taken in each plot (72 cultivatedplots and 12 forest plots). The composite sample was obtained by mixing fifteenrandom subsamples from the 0-15 cm mineral layer. Soil samples were collected atthe end of November 2005, after harvest.Microbiological and chemical properties at whole soil samples, and chemicalproperties at soil particle-size fractions were studied. Soil organic carbon wascalculated from the total carbon measurement subtracting C from carbonates. SoilMicrobial Carbon was determined by the fumigation-incubation method and posteriorC in extracts was determined by wet oxidation with dichromate. The microbialquotient (MBC/SOC) represented the fraction of MBC with respect to the SOC.Potential Soil Respiration was determined in closed jars and under laboratorycontrolledconditions. The metabolic quotient (qCO2) represents the potential soilrespiration per unit microbial biomass, and was calculated as SR/MBC.Particle-size fractionation was carried out through ultra sonification. Particulateorganic matter (POM) was separated in the coarse sand fraction, and the fractions offine sand (FS), silt and clay were analyzed for C (as previously SOC), N (as totalnitrogen at a LECO C/N/H analyzer) and the ratio C/N.The results showed significant differences for the values of SOC, MBC,MBC/SOC, SR, qCO2, between forest and agricultural plots. Higher values for forestswere detected for SOC, MBC and SR (these three properties are highly and positivelycorrelated to each other). This was expected according to the findings of manyresearchers, as tillage promotes soil organic matter (SOM) decomposition, decreasingthe sources of energy and the available substrates for the microorganisms, leading todecreases in MBC, and SR which is highly dependent by the availability of SOM.Factors that were significant for these three properties were changes in land use andthe amount of years since deforestation. For SOC and MBC, management as a factorwas also significant. Forest values for the two microbiological ratios (MBC/SOC andqCO2) had significantly lower values, as reported by several researchers forMBC/SOC and by many researchers for qCO2. These lower values for the forest soilsof the ratios are indicators of less stressed conditions of the microbial community, dueto better preserve of SOC and higher accumulation of microbial biomass. For theseratios, land use was always a significant factor. For the MBC/SOC ratio the amount ofyears since deforestation was also significant, whereas for qCO2 the type of crop(management factor) was significant. The newest deforestation (1980) holded thebiggest values for every property, except for the ratios. The amount of time sincedeforestation was adversely associated with SOC contents. Continuous soil tillageinduces C loss and releases large amounts of CO2 to the atmosphere. After manyyears systems reach or are near an equilibrium, leading to lesser values of qCO2.Wheat plots had always significantly higher values than cotton plots, when referringto SOC and MBC, whereas for qCO2 the reverse trend was observed. The reducedtillage practices (including N fertilization which accumulates higher N contents inwheat fields compared to cotton fields) of the wheat crops have improved conditionsof the soil to protect SOC, compared to the conventional tillage practices of the cottonplots. For the other two properties management was not a significant factor. Thequantity and the amount of years since land use change are more important in theSOM breakdown functions than the type and intensity of cultivation. The interactionbetween the amount of years and management was not significant for any property.Nitrogen and the C to N ratio in the whole soil, and particulate organic matter(POM) followed the same trends as SOC. Always significantly higher in forest(referring to land use), in the most recent deforestation (1980) (referring to the amountof years since deforestation), and in the wheat than in cotton plots (but notsignificantly, and not for the C/N ratio, referring to management as a factor). These 3factors were significant for N, but only the first two (land use and the amount of yearssince deforestation) for the C/N ratio and POM. The amounts of N and POM werehighly correlated with the contents of OM of a soil, and to each other. The ratio inforests refers to more fresh (less humified) material hence in a higher value. In theforest soils most of SOC and N were present in the fine sand fraction, which isexpected (fresh organic material), with significantly higher values than silt and clay(except for the N contents in forest fine sand and clay where there are no significantdifferences). The reverse trend was observed for the agricultural soils, where clayholded the (significantly) greater values, due to protection of SOM. The C/N ratioholds its (significantly) greatest values for forest in the fine sand fraction, whereas forthe agricultural plots in the silt fraction. Land use was a significant factor for all theproperties in the studied fractions. Forest soils holded the greater values for all theproperties in all the fractions compared to the agricultural plots, and except the Ncontents in clay, the differences were significant. Cultivation results in the breakingup of the aggregates, leading to protected C losses from silt and clay. Soil organicmatter losses lead to N losses. The great losses of POM also lead to N losses, as thesetwo properties were significantly correlated.The amount of years since deforestation affected significantly C and the ratio, butnot N. Nitrogen was mostly dependent on management practices (N fertilization). Themost recent deforestation (1980) holded the greater values for all the properties in allthe fractions (except for clay, where for C and N the 1st deforestation (1933) hadsimilar values, and for the ratio there are no significant differences in values betweenthe silt and clay fractions. Clay´s capacity to preserve carbon seems to diminish asyears go by). For all the deforestation categories, clay had significantly higher valuesreferring to C and N, whereas for the ratio the significantly higher values were foundin the silt fraction. The stabilization of organic matter in the clay particles results in asteady content throughout the years, so we observed similar values for the C/N ratioin this fraction.Management affected the properties in all the fractions, not always significantly,though overall it was a significant factor. The biggest values of SOC and N(significant for almost all the comparisons) were always found in clay, but for the C/Nratio in silt. The SOC contents in the clay fraction of the fields had no significantdifferences regardless management type, due to the stable behaviour of this fraction.Clay in the wheat crops showed always significantly higher values for N, but not inthe other fractions. Enhanced nitrogen fertilization of the wheat fields resulted inaccumulation of N on the clay particles of those fields (also resulting in lower C/Nvalues). Referring to the fine sand fraction SOM withheld in FS was quickly lost,hence management did not affect the values significantly.The interaction between time and management was not a significant factor.Deforestation and further establishment of agricultural use leaded to decreases inSOC and levels of microbiological properties, suggesting a deterioration of soilquality. Despite the fact that a relatively large amount of time had elapsed sincedeforestation and subsequent crop establishment in the three compared deforestationstages (1933, 1971 and 1980), the amounts of SOC, MBC and SR were significantlyhigher in soils deforested in the most recent stage (1980). The type of crop affectedthe studied soil properties. Cotton plots (considered as conventional management) hadsignificantly lower SOC and MBC values and microbiological efficiency levels thanthe wheat plots (considered as reduced management).Land use changes had significant effects in SOC and N contained in the soilfractions. The amount of POM, and C and N in the fine sand fraction presented themost distinctive differences (losses, due to loss of fresh organic material) amongforest (significantly higher values) and agricultural plots. The amounts of SOC and Nin the clay fraction should not be considered as a sensitive indicator in order tomonitor changes in land use regimes, as the contents remained rather stable dependingon the nature of land use.The amount of time since deforestation played a higher role in the order:sand>silt>clay and was adversely associated with the values of SOC and N in theparticles, except for clay where after 25 years of cultivation the capacity of thisfraction to preserve SOM seems to have reached an equilibrium. Soil organic matterbound in the clay particles are well protected, hence values in clay were always higherin this particle-size fraction referring to agricultural fields.Irrespective of the management that follows, the most significant factor was theland use change for the majority of the studied properties in the soil particle fractions.

Η εκχέρσωση δασικών εδαφών (αποδάσωση) υπήρξε μια πρακτική στην Ελλάδα (καθώς και στον υπόλοιπο κόσμο) τον περασμένο αιώνα προκειμένου να προκύψει γη κατάλληλη για γεωργική χρήση. Στη Φιλυριά Κιλκίς, 3 αποδασώσεις (1933, 1971, 1980) έλαβαν χώρα σε δάσος Quercus pubescens και προέκυψαν χωράφια 20-90 στρεμμάτων, καλλιεργούμενα κυρίως με σιτάρι, βαμβάκι και κερασιές. Εξετάστηκε η περίπτωση χωραφιών σιταριού (μη αρδευόμενα) και βαμβακιού (αρδευόμενα) αναφορικά με τις χημικές και μικροβιολογικές ιδιότητες των εδαφών. Οι παράμετροι εξέτασης αφορούν την αλλαγή χρήσης γης (δάσος - αγροτική γη), τον τρόπο διαχείρισης της αγροτικής γης (αρδευόμενα - μη αρδευόμενα), και την επίδραση του χρόνου που πέρασε από την περίοδο (1933, 1971, 1980) της αποδάσωσης έως την λήψη και ανάλυση των εδαφικών δειγμάτων.Οι ιδιότητες που εξετάστηκαν είναι η βιομάζα, η αναπνοή του εδάφους, ο οργανικός άνθρακας, ο λόγος βιομάζας προς οργανικό άνθρακα, ο λόγος αναπνοής προς βιομάζα του εδάφους, το άζωτο του εδάφους και ο λόγος άνθρακα προς άζωτο.Υπήρξαν σημαντικές διαφορές μεταξύ ιδιοτήτων των δασικών και γεωργικών εδαφών για όλες τις εξεταζόμενες ιδιότητες. Τα δασικά εδάφη (μάρτυρας) παρουσίαζαν μεγαλύτερες τιμές για όλες τις ιδιότητες εκτός από τους λόγους βιομάζας προς οργανικό άνθρακα και αναπνοής προς βιομάζα του εδάφους, όπως αναμενόταν σύμφωνα με την υπάρχουσα βιβλιογραφία, καθώς η αγροτική διαχείριση μειώνει τις ποσότητες οργανικού άνθρακα στο έδαφος με άμεση συνέπεια στις μικροβιολογικές ιδιότητες. Η νεότερη αποδάσωση (1980) είχε τιμές πιο κοντά σε αυτές των δασικών εδαφών, σε σχέση με τις 2 προηγούμενες (1933, 1971). Επομένως η χρήση γης (δάσος - αγροτική γη) και ο χρόνος από την αποδάσωση είναι στατιστικά σημαντικές παράμετροι. Η διαχείριση γης (αρδευόμενα ή μη αρδευόμενα εδάφη) δεν είναι στατιστικά σημαντικός παράγοντας για τις περισσότερες ιδιότητες, ωστόσο παρουσιάζει τιμές πιο κοντά σε αυτές των δασικών εδαφών για τα μη αρδευόμενα εδάφη.Η ανάλυση άνθρακα και αζώτου έγινε στα κλάσματα του εδάφους. Η άργιλος παρουσιάζει σημαντικά μεγαλύτερη σταθερότητα για τις εξεταζόμενες ιδιότητες (C,N, C/N). Η χρήση γης και η διαχείριση ως παράμετροι είναι στατιστικά σημαντικοί παράγοντες, ενώ ο χρόνος σε ελάχιστες περιπτώσεις και κυρίως στα μεγαλύτερα κλάσματα του εδάφους.Η διαφοροποίηση στη χρήση γης (δασικό έδαφος - γεωργικό έδαφος) παρουσίασε τις μεγαλύτερες και στατιστικά σημαντικότερες διαφορές από τις υπόλοιπες εξεταζόμενες παραμέτρους. Σε όλες τις περιπτώσεις η ποσότητα οργανικού άνθρακα και ποσότητας και ποιότητας βιομάζας στο έδαφος ήταν στατιστικά μεγαλύτερη στα δασικά εδάφη.