A Land Surface Model For The Forest On Complex Topography

The significance of researching land surface is it focused on all kinds of physicalprocesses how we sustain our living between land surface and atmosphere. Different surfacehas different processes of heat and mass (for example, water, CO2) exchange whether smalldistrict scale or the big global scale. Otherwise, with the development of human society,human activities intensified the climate change. Therefore, modeling is an effective way thatcan reproduce and predict the human activities’ impact on the meteorological environment,atmospheric environment and climate change. At present, research of flat terrain model hasbeen basically matured, but, the land surface model for the forests in mountainous area is lesscommon, thus we need to build a new land surface model for the forest on the complextopography.The land surface model has experienced three stages development. In1960s, in the basisof atmospheric general circulation model (AGCM) for numerical weather prediction, the firstgeneration land surface model, Bucket Model was created. It is the earliest parameterizationscheme that was made by Manebe in1969. One can note that the bucket model does not takeinto account vegetation or groundwater at all.The soil-vegetation-atmosphere transfer model (SVAT), introduced by Deardorff (1978),which was seemed as a foundation, the second generation land surface model was built. Thesecond generation land surface model，compared with the first generation, it put theconsideration for heat, water, momentum exchange and vegetation into consideration, as wellas the snow cover situation, but it also exists a lot of problems. In1980s, Dickinson made theBiosphere Atmosphere Transfer Scheme (BATS). It has three soil layers and one vegetationlayer. Vegetation in BATS is assumed to be a flat porous and uniform layer. BATS does nottake groundwater into account, but it does vegetation.Another practical land surface model is Simple Biosphere Model(SiB), which was madeby Sellers at the same stage. SiB has three soil layers and two vegetation layers. In SiB’s twovegetation layers, the top layer consists of trees or shrubs while the ground layer is for grasses.The snow depth prediction is very crude and no treatment of snow temperature. SiB can beoperated in offline and coupled mode.The third generation land surface model, it not only put the vegetation’s photosynthesisinto consideration, but also the carbon circulation process. It also refers to the relationshipbetween insolation and photosynthesis, then, the strength of photosynthesis will affect thecarbon circulation. Masterpieces of third generation land surface model are LSM, SiB2andAVIM. In order to evaluate the function of the artificial satellite which was utilized to observethe ecosystem in the Daihachiga River of the Jintsu River basin system, the detailedobservation of this ecosystem function is performed. As one of the centralized station of thisriver basin, Takayama coniferous forest site was established with the artificial plantation ofJapan cedar in eastern Takayama, otherwise, the forest observation using an observationtower that is performed (Saito et al.2010) there.This TKC site (36degrees08minutes23seconds north latitude,137degrees22minutes15seconds east longitude) locates approximately14km away from the east to the ce nter ofTakayama city, about800meters height above sea level, observing site is a downward-facingslope(334°)in north-northwest direction with an21degrees’ average inclination. Evergreenconiferous forest in this TKC site is an artificial plantation o f Japan cedar with a height of20meters. In the north side slope of the observation tower, there is a place of residence(200meters towards north). Figure6shows the location of the observation tower. In addition,Figure7shows the arrangement of measuring instruments.A practical land surface model should include at least five physical or biologicalprocesses：(1) Kinetic thermodynamic process of soil: it defines the process of water and heattransport between soil and atmosphere.(2) Physiological process of plant: it defines the biophysical process of planttranspiration(3) Process of turbulent transport: it defines the role of turbulent transport in land surfacemodel, atmosphere and boundary layer resistance of momentum, heat mass transport.(4) Water balance process: it identifies the distribution of precipitation and can be usedto estimate the soil moisture, runoff, evaporation, transpiration.(5) Radiation balance process: it identifies the available energy of each vegetation layerand soil surface, also can be used to estimate the sensible heat and latent heat.Besides these five aspects, recently, more and more researchers put the carboncirculation into consideration.(6) Carbon circulation process, the vegetation will affect climate and CO2concentration,so it should be mastered well. It can show the carbon source, carbon sink, carbon absorptionby vegetation and soil, and it also refers to vegetation physiological process.In this model, the soil is divided into4layers, the upper two layers correspond to thoserespectively showing diurnal and seasonal variations of temperature, and the root system ofvegetation is evenly distributed in these two layers. Temperature and moisture in the deepestlayer are left to keep their climatic means values. This assumption is much satisfactory when the simulation periods are not too extended. So according to the different situation, we usedifferent physical equations to calculate.This new land surface model aims to improve the structural design and calculationmethods, thus compared with the past land surface model, there are some innovation points inthis model. Firstly, it divided the vegetation into a more detailed classification according tothe height of the vegetation. This will improve the accuracy of model simulation. Secondly,the basis of this model is set up in a slope with a more profound practical significance, and thesoil is also divided more finely, the physical equations and empirical analytical method arecombined to solve the soil temperature and soil moisture in each soil layer, this will not onlyensure the calculation accuracy, but also saves the computation time of each layer. Thirdly, inthis land surface model, I referenced to the results of Mizuno predecessor—Presumption ofinsolation distribution in complex topography, it will make this model more complete,different shapes of complex terrain corresponding to the different distribution of insolation,and this will affect a lot of parameters’ value. Fourthly, a more detailed consideration ofprecipitation distribution is enumerated in this model, it divides precipitation into some partsand calculate them respectively. Fifthly, it calculates the different turbulence flux caused bywind with the different vegetation density. Finally, the interactions between the terrestrialbiosphere and climate in atmospheric model are represented by soil-vegetation-atmosphere-transfer scheme (SVAT). Although SVAT schemes that include partially dynamic vegetationmodules are emerging, most SVAT schemes do not take into account the dynamic aspects ofvegetation that are necessary for modeling climate.Actual land surface processes will involve complex atmospheric physical processes,biological processes, and hydrological processes, and so on. So it is a new multidisc iplinaryfield of research includes meteorology, ecology, hydrology and agroforestry actually. Landsurface model is an evolving process, therefore there are still some points need improvements,such as how to do a best soil stratification, how to simulate the most accurate hydraulicprocess, how to make a best model structure, how to calculate different weather phenomena.Some land surface model that best fit the actual environment will be got if more in-depthresearch in these areas continues.