Feasibility Study Of Using Fine Sand As The Substrate Of Extremely Shallow Subsurface Constructed Wetland

The substrate is the largest part in construction investment of subsurface flowconstructed wetlands (SSFCW)，it accounts for about40to50%of the total investment.To significantly reduce the construction cost，a pilot scale study was employed toinvestigate the feasibility of using fine sand as the substrate of extremely shallowSSFCW, in which the influent was the effluent from a primary sedimentation tank in amunicipal wastewater treatment plant in Xi’an. This study investigated the relationshipamong the different seasons, different fillers depth, different wetland plants andprocessing efficiency,and investigated climatic conditions and On-site maintenance andmanagement impacted on the treatment effect. The substrate depth of experimentalgroup was0.1m， Festuca arundinacea, Lolium perenne, Cynodon dactylon,Phragmites australis and Typha orientalis Presl were planted.The control group wassubstrate0.1m deep SSFCW without packing plant control and substrate0.6m deepSSFCW，and Cynodon dactylon and Typha orientalis Presl were planted. During theseedling stage, to avoid congestion, add wheat straw into fine sand. The theory HRT ofall wetlands is6d，running water depth were0.08m and0.58m，the water loading ofexperimental group wetlands is0.03m3/d, intermittent water three times a day，and thewater loading of0.6m deep SSFCW is0.22m3/d，intermittent water eight times a day.The system running for six months.The main results and conclusions are shown asfollows:(1) In non-winter, the average removal efficiencies for SCOD, NH4+-N, TN and TPwere higher than65.2%,93.7%,82.8%and95.0%,compared with the non-winterefficiency than in winter, although significant differences and fall, but the removal isstill good, showing extremely shallow SSFCW good treatment effect. This phenomenonis consistent with gravel and coarse sand as the substrate of extremely shallow SSFCW. Though0.1m deep and0.6m deep wetlands design theory HRT is same, in the runningprocess0.6m deep wetlands treatment efficiency decreased,0.1m deep wetlandstreatment efficiency is significantly higher than0.6m deep wetlands.(2) The depth of substrate was closely related with the hydraulic loading rate andhydraulic retention time, and directly affects the removal efficiency of pollutants inCWs. Studies have shown that： the DO of0.1m deep SSFCW effluent wassignificantly higher than0.6m deep SSFCW, effectively solves the traditional largedepth of SSFCW and lack of DO in the anaerobic area is too large, enhance theefficiency of the system to remove pollutants. Even0.1m deep control group withoutthe plant effluent DO was also significantly higher stably than the influent DO,suggesting that even substrate porosity decreased because of the thickness of the finesand is only0.1m, reaeration role continued strong.(3) In winter Festuca arundinacea and Lolium perenne still showed strong vitality,and multiple tiller, this can reduce the hassle for species, saving the cost of managementand maintenance.The0.1m deep wetland planting Festuca arundinacea and Loliumperenne effluent DO0.1m was significantly higher than other0.1m deep wetlands (P<0.05), TN, TP and NH4+-N average removal rate reached78.7%,85.0%and88.9%,and stably higher than other0.1m deep wetlands, so Festuca arundinacea and Loliumperenne are more suitable as wetland plants in cold northern areas.(4) Wetland plants plays a crucial role in removing pollutants, each0.1m deepwetland effluent pollutant removal efficiency was significantly better than the0.1mdeep control group. After system running about three months the substrate void ratio of0.1m deep wetlands increased by7%-25%, the0.1m deep control group decreased by20%.(5) During the seedling stage, the addition of wheat straw in the substrate resultedin the increasing of SCOD and turbidity of effluent, while it was not significantinfluence by76d (P>0.05). The wheat straw adding to fine sand can significantlyimprove the efficiency of denitrification (P=0.012<0.05).