Tresystem. With es of a tree pensupport and upname as the ft, which gripsto the entire foroots of trees doating foundatund basement, The supportinwhich is true borks on similarthe root netwondation systemAyam” foundacomponents:  aferring the loaucts, if cost minturing plants, dation” foundaof 22 mm extmm in length, wFor illustratioof models weher with closem” system, whlose-end founded in a glass  cs remoulded  adetermined basgn Concept kar Foundatioystem mentionees of various ptheir natural snetrate and gripphold the mainfoundation sys the soft soil oundation systdescribed earlition has the we, where the totng soil layer wbut the form ofr principles, sprk. This gives m. ation system, tha light platformad. The platfonimization is aor on a largeration model wternal diametewere spaced eqon purposes, there prepared  fe-end ones.  Nhere the pipes dation model wchamber of 30and used as thsed on BS 1377on” drew fromned above. Theproportions ansurvival instinp the soil as then trunk and uppstem is essentiaand spreads  ttem. The root ier, which coneight of the strtal load imposewill then be ‘fof load is changpreading and dthe second dehe self-weightm and polyvinorm can be  ma top priority. r scale, producwas made of a per and 15 mmqually at 34 mhe arrangemenfor each set  oNote that the  cwere filled wiwith 34 mm sp00 mm x 300he soil bed in 7 (1990), and tm both the floae word “akar” nd sizes stand tnct of reachingey grow and eper ground parally a firm basthe load over network concenstitutes the firructure balanced by the strucooled’ into belged), and theredistributing addfinition of the t of the “Akar nyl chloride  (made of foam The PVC pipeced from recypiece of 100 mm internal diammm or 51 mm nt of pipes for of pipes (i.e.  eclose-end pipeith in situ soil.pacing. 0 mm x 300 mthe chamber the results are ating foundatiocomes from  ttall with stabilg water sourcexpand, eventurts of the tree. se supported bthe subsoil,  tept resonates wrst definition oed by the remocture is less or ieving that theefore will not sditional structu‘floating’ conFoundation” wPVC) pipes thfor instance, es can be cut-oycled plastic  amm x 100 mm meter. 3 sets centre-to-centthe 34 mm spequal pipe lenges were left  e. Figure 2 showmm. An organat its natural given in Tableon principles athe local Malality provided bes undergroundually developinby a network othus giving thwith that of thof the ‘floatingoval of soil anthe same as the sustained loasettle or subsidural load on thncept adopted iwas reduced bhat function  aor any localloffs retrieved aas an additionax 10 mm thicof PVC pipetre and glued tacing system igths), one witempty inside  aws the plan annic clay samplwater contene 1.  as ay by d, ng of he he g’ nd he ad de. he in by as ly at al ck s, to is th as nd le nt. The soil sample was first remoulded in a separate container, then covered and left overnight prior to being used to form the soil bed in the test chamber (Figure 3). The purpose of this procedure was to allow uniform redistribution of moisture within the soil mass after the disturbance it underwent during sampling and transportation. On the following day, the soil sample was transferred to the glass chamber to form the soil bed by lightly compacting 4 layers of wet soil weighing 6 kg each. The compaction was necessary to avoid entrapment of large quantities of air within the soil voids, which can cause errors in the settlement measurements during load tests. The compacted height of each layer was carefully monitored to ensure uniformity in the soil bed overall density (Figure 4). The final thickness of the soil bed was fixed at 200 mm, measured from the base of the chamber. For the load test, dead weights were placed on top of the miniature foundation model to exert vertical stresses in the sequence of 0.25, 0.50, 0.75, 1.00 and 1.25 kPa (Figure 5). The vertical displacement under each load was recorded at prescribed time intervals of 5, 10, 15, 30, 60, 120, 240, 480 seconds, and finally terminated after a time lapse of 16 minutes (i.e. 960 seconds). A control test was also included in the test programme (i.e. sample Control), with only the plywood platform placed on the soil bed and loaded accordingly.   4. Results: Analysis and Discussions 4.1 Rate of Settlement Judging from the gradients of the settlement-time plots of Figures 6 and 7, settlement rates did not vary significantly between the open-end and close-end systems. This could be attributed to the intended ‘floating’ mechanism of both systems, where the applied stress was evenly spread over the contact surface between foundation and soil. By taking a closer look at the manner in which the plots unfolded, most settlements took place within the first 8 minutes. As these settlements were maintained for a short time lapse of just over quarter of an hour, prevalence of the undrained loading conditions were only to be expected. The pipes were likely to be pressed into positions of stability in bearing the load and transferring it to the surrounding grounds. For the pipes spaced at 34 mm  centre-to-centre apart, the  close-end system appeared to be more effective in settlement reduction compared to the open-end system  (Figures 6a and 6b). However this deficiency of the open-end system was compensated with greater pipe lengths, as shown by the equal settlements of L75_34 and L75_51 in the same figures. The settlement reduction seemed slightly diminished with increased spacing of the pipes for the close-end system, but was quickly recovered with increased pipe lengths (Figure 7a). As for the open-end system, having the pipes further apart was clearly detrimental to this effect, causing up to almost 15 % of additional settlement (Figure 7b).  “阿卡尔基础”软土施工沉降控制英文文献和中文翻译(3):http://www.751com.cn/fanyi/lunwen_54419.html