The Responses Of Belowground Net Primary Production And Fine Root Dynamics To Different Management Practices In Songnen Grassland

Grassland ecosystems with important belowground carbon stock are largely dispersed on the earth. Belowground carbon stock is the central part of carbon cycle and allocation of grassland ecosystems. Knowledge of the belowground carbon process （exceptionally root turnover process） of grasslands can help us to better assimilate the terrestrial ecosystems carbon cycle. However, because of its importance for millions of people, grasslands are being seriously disturbed by excessive reclamation, mowing, grazing, and farming management practices from human among the natural and anthropogenic factors. Grassland degradation through excessive reclamation has become a serious environmental and ecological problem in China and its restoration is an urgent issue. To this end, it is imperative to consider the response of belowground part to different ecological managements for a sustainable and efficient restoration from the entire plant outlook. The Songnen grassland, one of the main grasslands in China, located in Jilin province is a typical meadow steppe. The effects of different ecological managements, especially addition of litter to sowing area, tillage, clipping aboveground biomass on belowground net primary production （BNPP）, root fraction (f_BNPP), fine root dynamics （root production, root mortality, standing crop and root turnover） and fine root lifespan have not been reported yet. Therefore, study on the response BNPP,f_BNPP, fine root dynamics and root lifespan to different ecological managements may improve our conception of Songnen grassland belowground ecological process and results may contribute to fill the literature gap in this field. Besides, by combining both aboveground and belowground biomass data, we might establish an approach concerning the effects of different management practices on Songnen grassland, as well as provide suggestions on the rational exploitation and or the sufficient restoration.In order to study different grassland managements measures for BNPP,f_BNPP, root production, mortality, turnover and fine root lifespan, in 2012-2014 two main sites using different measures of ecological managements were established. Site 1:Litter combined with seeding site. Management consisted of adding different amounts of litter (0g m^-2,200 g m^-2, 400 g m^-2 and 600 g m^-2) to previously degraded abandoned land and randomly fenced to seeding and non seeding treatments. In each treatment, four sub-treatments were set: non seeding keeping litter （NSKL）, non seeding moving litter （NSML）, seeding keeping litter （SKL） and seeding moving litter （SML）. Site 2:A degraded land was fenced into seven different treatments including:（MT）, maize no tillage keep residues （MNTKR）, maize no tillage move residues （MNTMR）, artificial grassland non clipping litter （AGNCL）, artificial grassland clipping litter （AGCL）, natural grassland non clipping litter （NGNCL）, and natural grassland clipping litter （NGCL）.In addition, to understand the correlation between species distribution patterns and soil salinity gradient in this degraded abandoned land at the stage of natural recovery, three representative semi-vegetated patches were selected and analyzed for soil electrical conductivity （EC）, soil pH, soil organic matter （SOM） and soil nitrogen （N）. Based on the different appearance and distribution of species the patches were divided into a center section （CS）, middle section （MS） and outer section （OS） each.The ingrowth Donut method was performed to collect root samples for BNPP and fBNpp determination and the improved minirhizotron or window glass method was used to determine fine root dynamics and lifespan during the whole study periods.In the first site, BNPP ranged from 53-301 g m^-2, and was increased 34% by seeding mainly due to the increase of Leymus chinensis’production. There was a trend that BNPP in 2012 was higher than those in 2013 and 2014, respectively due to the relative adequate rainfall occurred. Moving annual litter had more positive effects on BNPP in the seeding treatment than the non seeding treatment, especially in 2012 and 2014. With seeding, moving annual litter averagely increased BNPP 18.19% more than keeping annual litter highlighting seeding and moving annual litter have adding effects on BNPP. Overall,f_BNPP varied from 0.28 to 0.42, increased as litter amount increased due to allocation of more biomass to belowground. The results involve that the combination of litter and seeding enhance BNPP and f_BNPP, which may increase global C sequestration.In site 2, BNPP varied from 220 g m^-2 to 1331 g m^-2. Overall, a clear tendency showed that MT ranked significantly the highest BNPP value and the lowest was found with AGNCL treatment. Combined MT, MNTKR, MNTMR treatments had positive effects on BNPP compared to that of AGCL, AGNCL, NGCL, NGNCL treatments. Tillage and clipping litter practices enhanced BNPP. A positive relationship was found between BNPP and SOM suggesting that SOM should be considered as an indicator for total carbon （C） sequestration in grassland,f_BNPP ranged within 0.25-0.54 and was remarkably higher in 2014 compared to 2012 and 2013. f_BNPP was better improved by tillage and clipping litter practices probably due to drought leading to the allocation of more biomass to belowground.Fine root dynamics （root production, mortality, standing crop and root turnover） occurred throughout the entire growing season, fluctuating with time at both 0-10 cm and 10-20 cm soil surfaces and was related to soil moisture content. The peak values were observed in Jnly and Sentemher The annual root turnover rate ranged in 1.9-2.9 vr^-1 in the unner 20 cm of soil and was significantly higher with AGCL and NGCL. Also, significant differences in root longevity existed among treatments with a median root lifespan for cohorts ranging from 364-785 days and 380-786 days, for 0-10 cm and 10-20 cm soil depth respectively and more expressed in clipping litter treatments.By investigating the correlation between species distribution patterns and soil salinity gradient, we found that in contrast to EC, SOM significantly increased from the center to the outer section. Aboveground biomass and species distribution patterns were strongly associated with SOM and contrasted to EC. In the center section, the annual species Kochia scoparia （L.） Schrad represented more than 90% of the total biomass, whereas the perennial species Leymus chinensis （Trin.） Tzvel and Phragmites australis （Cav.） Trin. ex Steud were the most dominant species in the outer section with a contribution of more than 90% to the total biomass. In the middle section, most of the species achieved mutual coexistence and the perennial species appeared to benefit from the interspecific relationships with their neighbors.In conclusion, the results underlined that the combination of litter and seeding, clipping litter, tillage ecological managements increased BNPP,f_BNPP, fine root dynamics and lifespan which may enhance global C sequestration. Therefore, they appear to be valuable in facilitating model projections for climate carbon feedbacks in the future. A sustainable and efficient grassland restoration imply of increasing both above and belowground production and the practices used in the present study should be considered by managers performing restoration and presents good alternative through designing protocols for increasing grassland production. Moreover, study for a restoration of this land should aim at increasing SOM contents, reducing salinity stress, and restoring climax species community, as there is a strong response of vegetation to the gradient of biochemical soil parameters. Therefore, it is recommended that future studies should incorporate techniques that can contribute to reducing the salinity and pH and increasing the SOM and N content of the soil to achieve a sustainable and effective restoration of abandoned old field Songnen grassland.