SCS模型在貴州省畢節(jié)市石橋小流域坡面產流模擬中的應用

張興奇, 徐鵬程, 顧璟冉

(南京大學 地理與海洋科學學院, 江蘇 南京210023)

SCS模型在貴州省畢節(jié)市石橋小流域坡面產流模擬中的應用

張興奇, 徐鵬程, 顧璟冉

(南京大學 地理與海洋科學學院, 江蘇 南京210023)

[目的] 研究喀斯特地區(qū)徑流形成規(guī)律，得到適合研究區(qū)的SCS(space cooperation system)模型參數取值。[方法] 利用野外徑流小區(qū)觀測獲得的徑流數據與降雨資料，對研究區(qū)的降雨進行雨型分類，并利用SCS模型進行坡面產流模擬。[結果] (1) 根據降雨隨歷時的分配可將研究區(qū)的降雨雨型分為4類，即前期型(Ⅰ類)、中期型(Ⅱ類)、后期型(Ⅲ類)以及均勻型(Ⅳ類)。(2) 基于SCS模型對研究區(qū)坡面徑流進行模擬時，初損系數λ取值多為0.01，0.05，遠小于λ=0.2的取值。(3) 隨著坡長的增加，前期型降雨、中間型降雨和后期型降雨的CN(curve number)值變幅較小，均勻型降雨的CN值呈現增加的趨勢。隨著坡度的增加，前期型降雨和中期型降雨的CN值呈現減少的趨勢，后期型降雨和均勻型降雨的CN值增幅不大。[結論] 基于SCS模型得到的模擬徑流量與實測徑流量之間的相關系數和模型效率系數都較高，具有較好的模擬效果。

降雨雨型; SCS模型; 徑流模擬; 貴州

文獻參數： 張興奇, 徐鵬程, 顧璟冉.SCS模型在貴州省畢節(jié)市石橋小流域坡面產流模擬中的應用[J].水土保持通報，2017,37(3)：321-328.DOI:10.13961/j.cnki.stbctb.2017.03.055； Zhang Xingqi， Xu Pengcheng, Gu Jingran. Application of SCS model to simulate runoff in slope field at Shiqiao small watershed in Bijie City of Guizhou Province[J]. Bulletin of Soil and Water Conservation, 2017,37(3)：321-328.DOI:10.13961/j.cnki.stbctb.2017.03.055

降雨產流是引起土壤侵蝕的主要原因之一,定量地計算降雨徑流量,對于認識土壤侵蝕發(fā)生規(guī)律有著重要意義。貴州喀斯特地貌發(fā)育，成土速度慢，土層淺薄，山高坡陡[1]，加之降雨的時空分布不均等特點，導致該區(qū)成為土壤侵蝕最嚴重的地區(qū)之一。因此在該地區(qū)采取有效的水土流失防治措施迫在眉睫，這就需要對喀斯特地區(qū)的徑流形成規(guī)律進行研究。

SCS模型最早是美國農業(yè)部水土保持局對流域上發(fā)生的水文過程進行模擬建立的數學經驗模型，具有結構簡單、參數少和廣泛使用等優(yōu)點。目前。國內外學者對SCS模型開展了廣泛的研究：Ponce等[2]論證得到λ具有區(qū)域性，其取值的合理性需進一步研究。周淑梅等[3]利用1987—2006年14場降雨事件確定了最適合黃土丘陵溝壑區(qū)的初損值為0.1。王紅雷等[4]研究表明坡度坡長對坡面產匯流有重要影響，坡度坡長是CN值的影響因子。王興鵬等[5]發(fā)現從理論上講，CN取值介于0～100，但在實際條件下，CN值在30～100變化。王英等[6]優(yōu)化出了黃土高原地區(qū)3種土地利用方式的λ值，但模型模擬效率還是不高，所以提出降雨強度修正函數，將降雨強度因子引入SCS模型中。

本文基于SCS 模型的基本原理和應用流程，利用貴州省畢節(jié)市試驗小區(qū)2012—2014年的51場降雨資料，通過對SCS模型進行參數修正(主要是初損系數、CN值和降雨的修正等)，計算在不同的參數水平下的降雨徑流量,將模擬值與實測值相比較,探討該方法的合理參數和模型的有效性。

1 研究區(qū)概況

1.1 研究區(qū)自然環(huán)境概況

研究區(qū)位于貴州省畢節(jié)市鴨池鎮(zhèn)石橋小流域，屬長江水系烏江流域上段六沖河的3級支流區(qū)，流域面積8.19 km2；海拔高度1 400～1 743 m，地勢起伏大，相對高差為343 m；屬亞熱帶濕潤季風氣候，年均降雨量863 mm，降雨主要分布在5—9月，年均氣溫14 ℃。小流域內碳酸鹽類石灰?guī)r廣泛分布，平均裸巖率達30%，植被覆蓋率低，是貴州省境內具有代表性的喀斯特石漠化地區(qū)。

1.2 徑流小區(qū)布設

徑流小區(qū)布設在石橋小流域，共布設了5個坡長小區(qū)、5個坡度小區(qū)和2個標準小區(qū)，小區(qū)基本情況見表1。

表1 貴州省畢節(jié)市小流域徑流小區(qū)基本情況

1.3 試驗方法

試驗數據來自12個徑流小區(qū)2012—2014年51次侵蝕性降雨野外觀測記錄以及室內分析結果。降雨產流產沙后第1時間實地觀測取樣，觀測集流池、分流池內的水深，用以計算坡面徑流深：

Q=103V/S

(1)

V=S1Q1+rS2Q2

(2)

式中：Q——坡面徑流深(mm)；V——坡面總產流體積(m3)；S——野外徑流小區(qū)面積(m2)；S1——集流池池底面積(m2)；S2——一級分流池池底面積(m2)；Q1——集流池水深(m)；Q2——一級分流池水深(m)；r——一級分流系數。

2 研究區(qū)降雨雨型分類

降雨雨型是指研究次降雨過程中降雨量的集中程度，其反應了降雨過程中降雨量隨歷時的分配[7]。選取2012—2014年研究區(qū)逐分鐘降雨過程的自記雨量記錄紙，經過數字化后整理得到次降雨的總降雨歷時(T)、累積降雨歷時(t，0≤t≤T)、總降雨量(P)和累積降雨量(Pt，10≤Pt≤P)。以累積降雨歷時與總歷時的比值作為橫坐標，累積降雨量與總降雨量的比值作為縱坐標，用Origin軟件對得到的次降雨量綱標準化后的數據進行曲線擬合，得到擬合的降雨過程曲線，即標準化降雨歷時—降雨量曲線。在SPSS軟件中運用K均值分類法按照降雨量值之間的距離進行簡單分類，將研究區(qū)51次降雨聚為6類降雨事件，6類降雨事件占總降雨場次的比例分別為：4%，10%，27%，10%，35%，14%。其中3，6類降雨事件占41%，1，2類降雨事件占較小的比例。

對按照降雨特征值分類的6種降雨類型再按照雨型曲線的凹凸特征歸類為4大類，得到：3，6類降雨歸為Ⅰ，4，5類歸為Ⅱ，1，2類歸為Ⅲ，部分5類歸為Ⅳ，4大類降雨事件分別占總降雨次數的41%，27%，14%和18%。將標準化后的總降雨歷時以0.2為間隔分為5等份，繪制圖1。Ⅰ類雨型降雨歷時—降雨量曲線呈現上凸，先陡后緩，這類降雨類型稱為降雨前期集中型。Ⅱ類雨型曲線呈現兩端平緩，中間陡峭的線型，這類降雨稱為降雨中期集中型；Ⅲ類雨型曲線是下凹型，前期上升較緩，后期上升較快，這類降雨稱為降雨后期集中型；Ⅳ類雨型曲線趨于直線，全段降雨量增長均勻平緩，這類降雨稱為降雨均勻分布型。

圖1 貴州省畢節(jié)市小流域4類標準化降雨歷時與降雨量關系

3 徑流曲線數模型(SCS-CN)簡介與評價

3.1 模型介紹

徑流曲線數模型(SCS-CN)假定地面徑流量與潛在徑流量之比等于流域實際入滲量和最大入滲量之比,即：

(3)

式中：F——流域實際入滲量(mm)；S——流域最大入滲量(mm)；Q——地面徑流量(mm)；P——降雨量(mm)；Ia——初損雨量(mm)。

根據水量平衡原理有：

P=Ia+F+Q

(4)

推得：

(5)

Q=0 (P≤Ia)

(6)

方程(5)中有2個變量Ia和S,Ia是一個與植被截留、入滲、前期土壤含水量等有關的變量。大量小流域研究結果表明,Ia和S之間存在著良好的線性關系：

Ia=λS

(7)

式中：λ——初損與流域潛在入滲量的比率,SCS模型在實際應用中取λ=0.2進行計算[8]。大量試驗資料表明λ因地區(qū)不同變化范圍為0.00～0.30[8-10]，本文會重新修正研究區(qū)λ的取值范圍。

S與CN值的經驗轉換關系如下：

(8)

CN值是反映土地利用、土壤類型、前期土壤含水量的一個綜合指標，美國國家工程手冊給出了詳細的CN值。雖然CN可以通過查表得到，但CN值是在美國測定的，不太適合國內的情況[11]，利用查得的CN值推求的徑流量與實測值差別很大。因此本文將逐一運用平均值法、中值法、算術平均值法、S對數頻率分布法和漸近線法[12]對CN值進行反推，確定適合研究區(qū)的CN值的取值范圍。

(9)

參照黃土高原地區(qū)的降雨雨強修正函數[13]，得到貴州地區(qū)的降雨雨強修正函數如下：

(10)

3.2 模型評價參數

利用SCS模型可以計算降雨形成的地表徑流量的模擬值，將之與實測徑流值進行比較，計算得出模型效率系數和相關系數后，可得知SCS模型的徑流模擬效果。其中相關系數由回歸分析得到，Nash和Sutcliffe提出的模型效率系數通過下式得到[14]：

(11)

4 基于SCS模型的坡面產流模擬

坡面產流模擬計算是小流域水土流失研究的重要內容，徑流模擬計算時水文模型的選擇與模型參數的率定十分關鍵。SCS模型具有參數簡單以及計算簡便的優(yōu)點，基于貴州典型小流域實測的降雨與徑流資料，可對SCS模型進行參數修正，主要針對該模型中的初損系數、CN值和降雨等參數。通過參數修正優(yōu)化后的SCS模型可以進行同類地區(qū)坡面產流量的模擬計算。

4.1 SCS模型模擬方法比較

由于坡度、坡長對產流機理有重要的影響，所以有必要按照坡度、坡長分別對SCS模型的應用進行探討。

4.1.1 標準小區(qū) 標準小區(qū)的坡度為5°，坡長為20 m(表2)，其CN值確定方法為平均值法和S對數頻率法，但是得到的最大的模型效率系數以及相關系數不能滿足模擬精度要求，需要進一步進行模型參數的優(yōu)化。

表2 標準小區(qū)CN值反推方法的最優(yōu)化結果

4.1.2 坡長小區(qū) 坡長小區(qū)的坡度都是15°，利用5種常用的反推計算CN值的方法[12]，得到了不同CN值反推方法下的CN值及其相關系數和模型效率系數，將得到的相關系數和模型效率系數較高的方法整理得到了表3。初步判定在坡長小區(qū)上降雨類型分類下反推CN值的方法沒有明顯的一致性，但是可以得知平均值法、中值法和算術平均值法是既方便又有效的方法，而漸近線法只在模擬坡長為20 m小區(qū)的前期型降雨產生的徑流時呈現較好的模擬效果。

4.1.3 坡度小區(qū) 由于坡度對產流具有重要影響，通過模擬分析得到結果(表4)。坡度小區(qū)的坡長都是10 m，坡度小區(qū)與坡長小區(qū)在反推計算CN值的方法上比較相似，平均值法與算術平均值法有較好的模擬效果，但在坡度小區(qū)中中值法的模擬效果較差。后期型降雨與均勻型降雨運用S對數頻率法的徑流模擬效果較好。

綜合分析，CN值的確定方法是多樣的，與坡長、坡度的相關性不明顯，但是總體來說平均值法與算術平均值法是比較好的方法，使用S對數頻率法時的頻率取值一般為10%和50%。

表3 坡長小區(qū)CN值反推方法的最優(yōu)化結果

表4 坡度小區(qū)CN值反推方法的最優(yōu)化結果

4.2 SCS模型的初損系數調整

4.2.1 坡長小區(qū) 初損系數λ一般取值0.2，但是在研究區(qū)2個小流域模擬計算后發(fā)現λ取0.2不能滿足SCS模型徑流模擬的精度要求，初損系數λ是一個對徑流模擬影響較大的參數，必須進一步優(yōu)化。

研究表明初損系數λ是一個區(qū)域經驗值，其值范圍通常在0～0.3[9]，本研究基于此對畢節(jié)小流域進行SCS模型的初損系數λ優(yōu)化，得出畢節(jié)小流域的標準小區(qū)的初損系數如下：前期型降雨、中期型降雨、后期型降雨和均勻型降雨的λ值取0.01，0.01，0.2，0.05。采用上述λ值后最低模型效率系數分別提高了26，10，4，12個百分點。

按照坡長變化得到表3，λ的取值多數是0.01，0.05，其中初損系數取0.01，較經典取值0.2低，說明試驗小流域的坡長小區(qū)在降雨發(fā)生后、地表產流前消耗的雨水較少，很大部分都成為徑流。后期型降雨的相關系數和模型效率系數是最大的，模擬效果較好，前期型和均勻型降雨的徑流模擬效果也較好，與未進行雨型分類前的徑流模擬效果比較可以得知λ的取值與雨型有較大的關聯性。

4.2.2 坡度小區(qū) 由表4得知隨著坡度的變化4類雨型的初損系數λ的取值呈現一定的規(guī)律性，例如前期型降雨和中期型降雨的λ值都隨著坡度的升高變小，這與黃土高原地區(qū)的λ值隨著坡度的增加而逐漸變小的規(guī)律一致[8]。后期型降雨的λ值在25°時最大，中期型降雨和均勻型降雨的λ值隨著坡度的增加變化的不明顯。從數值上看，λ的取值和坡長類似，多數是0.01，0.05，說明坡度小區(qū)與坡長小區(qū)具有相似的徑流初損規(guī)律，進一步說明了λ取0.01是具有代表性的初損值。與黃土高原地區(qū)的研究成果做比較，張鈺嫻得出黃土高原地區(qū)的λ值在緩坡取0.2，在坡度為15°時取0.1，30°時取0.03[8]。由于黃土高原研究區(qū)的坡度較大，所以其λ的取值較小，其取值為0.01[13]。紫色土丘陵區(qū)的λ取0.2或0.3[15]。綜合比較，西南地區(qū)的λ值比黃土高原地區(qū)和紫色土丘陵區(qū)的小，這與區(qū)域地形、植被、氣候因素以及土壤性質有關。

4.3 不同雨型下SCS模型的降雨雨強調整

由上文對畢節(jié)小流域降雨雨強的分析得知坡面產流產沙與每種降雨類型的雨強有極大的相關性，分析SCS模型與降雨雨強的關系將可能進一步提高SCS模型的模擬精度，所以有必要引入雨強修正函數。根據已有的黃土高原地區(qū)的雨強修正公式[13]，以及畢節(jié)小流域坡面產流產沙與I60的關系最強[16]，初步確定適合本流域的雨強修正公式如下：

(12)

4.3.1 標準小區(qū) 由于在未引入雨強修正系數時λ取0.01是該區(qū)的初損值，且通過幾個小區(qū)的分析后得知在λ取0.01后進行雨強的修正優(yōu)化是可行且較有效的，所以通過計算優(yōu)化分別得到雨強修正后的SCS模型模擬結果(表5)。

優(yōu)化后的雨強修正后的SCS模型模擬結果中坡度為5°，坡長為20 m，λ取值0.01。 在經過降雨雨強的修正后，對標準小區(qū)徑流模擬的結果表明：前期型降雨的模型值與徑流實測值的相關系數與未經雨強修正的結果相差不大，但模型效率系數有一定的優(yōu)化，提高了13%左右；中期型降雨的模型相關系數以及模型效率系數都有提高，分別提高了19%和13%左右；但是降雨雨強的修正對后期型降雨以及均勻型降雨的優(yōu)化程度不大，所以該2種類型的降雨不需要進行雨強的優(yōu)化。

表5 雨強修正后的標準小區(qū)CN值反推方法的最優(yōu)化結果

4.3.2 坡長小區(qū) 坡度為15°，λ取值0.01。從表6得到中期型降雨在坡長為5 m小區(qū)的SCS模型經過雨強修正后相關系數以及模型效率系數分別提高了4.9%和19%；中期型降雨在坡長為10 m小區(qū)的SCS模型經過雨強修正后相關系數以及模型效率系數分別提高了14%和25%，后期型和均勻型降雨的相關系數都沒有改變，但模型效率系數分別提高了3.5%和3.3%。中期型降雨在坡長為15 m小區(qū)的SCS模型經過雨強修正后相關系數以及模型效率系數分別提高了0.5%和6.9%；中期型降雨在坡長為20 m小區(qū)的SCS模型經過雨強修正后相關系數以及模型效率系數分別提高了2.4%和7.2%，后期型降雨的模型效率系數提高了0.6%；中期型降雨和均勻型降雨的SCS模型經過雨強修正后相關系數在坡長為25 m小區(qū)有較大的提高，分別提高了39%和28%，其模型效率系數都提高了45%。后期型降雨的相關系數沒有改變，其模型效率系數增加了40%，說明在坡長為25 m小區(qū)雨強修正的效果比較明顯。

表6 雨強修正后的坡長小區(qū)CN值反推方法的最優(yōu)化結果

4.3.3 坡度小區(qū) 如圖7所示，后期型降雨在坡度為5°小區(qū)的SCS模型經過雨強優(yōu)化后相關系數和模型效率系數分別提高了1.6%和14%，均勻型降雨的相關系數提高了0.7%和1.6%；坡長為10 m，λ取值0.01。后期型、均勻型降雨在坡度為10°小區(qū)的SCS模型經過雨強調整后相關系數分別提高了0.6%和0.4%，其模型效率系數分別提高了5.8%和7.0%；前期型降雨在坡度為15°小區(qū)的SCS模型經過雨強調整后相關系數和模型效率系數分別提高了9.7%和0.2%。后期型、均勻型降雨在坡度為15°小區(qū)的模型效率系數分別提高了14%和3.6%；后期型降雨在坡度為20°小區(qū)的SCS模型經過雨強調整后相關系數和模型效率系數分別提高了13%和20%；后期型降雨在坡度為25°小區(qū)的模型效率系數提高了1%，相關系數沒有改變，均勻型降雨的相關系數和模型效率系數提高了0.3%和6.2%。

表7 雨強修正后的坡度小區(qū)CN值反推方法的最優(yōu)化結果

4.4 不同雨型下SCS模型的CN值

分析圖2得知各類型降雨的CN取值與坡度、坡長有一定相關性。隨著坡長的增加前期型降雨、中期型降雨和后期型降雨的CN值呈減少趨勢，均勻型降雨的CN值呈增加趨勢。

分析坡長小區(qū)，得到坡長為5，10，15，20和25 m的小區(qū)各降雨類型的CN平均值為67，64，69，68，61。各坡長間的CN值最大差值為8，表明坡長小區(qū)的雨前土壤綜合狀況只有細微差別。

圖2 坡長小區(qū)各類型降雨的CN值

分析坡度小區(qū)得知，前期型降雨、中期型降雨的CN值也隨著坡度的增加呈減少趨勢，后期型降雨以及均勻型降雨的CN值隨著坡度的增加變化不顯著。

分析坡度小區(qū)，得到5°,10°,15°,20°和25°小區(qū)各降雨類型的CN平均值為51，46，50，35，40。各坡度小區(qū)的CN值最大差值為15，表明坡度對CN值的影響較坡長的大。

圖3 坡度小區(qū)各類型降雨的CN值

5 結 論

(1) 根據降雨隨歷時的分配可將研究區(qū)的降雨雨型分為4類，即前期型(Ⅰ類)、中期型(Ⅱ類)、后期型(Ⅲ類)以及均勻型(Ⅳ類)。

(2) 反推CN值簡單而且效果好的方法為：平均值法、算術平均值法、S對數頻率法和漸近線法。基于SCS模型對研究區(qū)坡面徑流進行模擬時，初損系數λ取值0.01，0.05分別占73%，20%，遠小于λ=0.2的取值，表明試驗小流域的降雨產流過程中徑流損失量少，與典型喀斯特地區(qū)參數λ小、產流條件好的特點一致[17]。

(3) 基于SCS模型得到的模擬徑流量與實測徑流量的相關系數和模型效率系數都較高，具有較好的模擬效果。其中標準小區(qū)的SCS模型參數范圍還需要進一步率定，對λ和β取值范圍做進一步優(yōu)化。對降雨進行調整可在一定程度上改善SCS模型的徑流模擬效果。

(4) 坡長為5，10，15，20和25 m小區(qū)的平均CN值為67，64，69，68，61，隨著坡長的增加，前期型降雨、中間型降雨和后期型降雨的CN值變幅較小，均勻型降雨的CN值呈現增加的趨勢；坡度為5°,10°,15°,20°和25°小區(qū)的平均值CN為51，46，50，35，40，隨著坡度的增加，前期型降雨和中期型降雨的CN值呈現減少的趨勢，后期型降雨和均勻型降雨的CN值增幅不大。

[1] 周赟,胡順.畢節(jié)市喀斯特石漠化的成因及防治措施[J].貴州大學學報:自然科學版,2007,24(4):421-425.

[2] Ponce V M, Hawkins R H. Runoff curve number:Has it reached maturity?[J].Journal of hydrologic engineering, 1996,1(1):11-19.

[3] 周淑梅,雷廷武.黃土丘陵溝壑區(qū)典型小流域SCS-CN方法初損率取值研究[J].中國農業(yè)科學,2011,44(20):4240-4247.

[4] 王紅雷,王秀茹,王希.利用SCS-CN方法估算流域可收集雨水資源量[J].農業(yè)工程學報,2012,28(12):86-91.

[5] 王興鵬,馬軼,張維江,等.SCS模型在黃土丘陵因子徑流場中的應用[J].寧夏工程技術,2005,4(2):157-159.

[6] 王英,黃明斌.徑流曲線法模型參數在黃土地區(qū)的優(yōu)化研究[J].水土保持通報,2008,28(1):54-58.

[7] 殷水清,王楊,謝云,等.中國降雨過程時程分型特征[J].水科學進展,2014,25(5):617-624.

[8] 張鈺嫻,穆興民,王飛,等.徑流曲線數模型(SCS-CN)參數λ在黃土丘陵區(qū)的率定[J].干旱地區(qū)農業(yè)研究,2008,26(5):124-128.

[9] Cazier D J, Hawkins R H. Regional Application of the Curve Number Method[C]∥Water Today and Tomorrow. New York, ASCE, 2011.

[10] Bosznay M. Generalization of SCS curve number method[J]. Journal of Irrigation & Drainage Engineering, 1989,115(1):139-144.

[11] 魏文秋,謝淑寨.遙感資料在SCS模型產流計算中的應用[J].環(huán)境遙感,1992,7(4):243-250.

[12] 符素華,王向亮,王紅葉,等. SCS-CN徑流模型中CN值確定方法研究[J].干旱區(qū)地理,2012,35(3):415-421.

[13] 王英.徑流曲線法(SCS-CN)的改進及其在黃土高原的應用[D].陜西 楊凌:中國科學院教育部水土保持與生態(tài)環(huán)境研究中心,2008.

[14] Nash J E, Sutcliffe J V. River flow forecasting through conceptual models(part I): A discussion of principles [J]. Journal of Hydrology, 1970,10(3):282-290.

[15] 陳正維,劉興年,朱波.基于SCS-CN模型的紫色土坡地徑流預測[J].農業(yè)工程學報,2014,30(7):72-81.

[16] 顧璟冉,張興奇,顧禮彬,等.黔西高原侵蝕性降雨特征分析[J].水土保持研究,2016,23(2):39-43,48.

[17] 鄭長統,梁虹,舒棟才,等.基于GIS和RS的喀斯特流域SCS產流模型應用[J].地理研究,2011,30(1):185-194.

Soil and Water Conservation Implications on Soil Security and Civilizations Decline—Commented by José Luis Rubio, An International Famous Expert of Soil and Water Conservation Research

The mismanagement or abandon of soil and water conservation systems is a crucial factor triggering soil security consequences including the decline or even the collapse of civilizations. This paper is a wide reference review. The objective of this paper is to analyze implications of soil and water conservation on soil security and civilizations decline from the point of view of man’s relation to productive soil in world history. Some conclusions are based on logic rather than in known facts. But we consider that the basic arguments and conclusions are sound and reasonably correct.

1 Threatening Factors, Implications and Achieving Measures of Soil Security

Soil security that can be analyzed from the perspective of state, human, and environmental security refers to the depletion of the soil capacity to provide foods and ecological functions that puts in extreme cases—the survival of affected people at risk.

(1) The five dimensions of soil security include: capability (functionality), condition (soil health), capital (monetary, ecosystem services), connectivity (social dimension of soil) and codification (policy and regulations). Soil security is a multidimensional issue: environmental, economic and social; Soil security is the extra dimension of soil degradation affecting primordial aspects of human life.

(2) Soil security is threatened by: ① the severe degradation of soils and related fertility and biodiversity; ② the loss of soil capacity to regulate & store water; ③ the climate impacts and changes (including drought) and human mismanagement. All of those could lead to bad harvests and crop yield declines. That has triggered severe and extended periods of famine affecting several billion people and causing the death of millions of persons.

(3) Soil security implications include: ① access to primary resources for subsistence; ② access to land, land tenure, technical knowledge and resources; ③ sufficient quantity and quality of food to all people at all times; ④ adequate nutritious diet for a healthy life. In other words, soil security can affects essential human needs.

If lacking the previous implies, the results could be: ① poverty, hunger, famines, social and political instability; ② conflicts, violence and wars for scarce or degraded arable land; ③ forced migrations and displacement; ④ decline or collapse of social and political structures.

(4) Soil security is achieved when efforts succeed in the following areas: ① containing land degradation and applying adequate soil conservation management to conserve soil fertility; ② when the consequences of soil degradation are reduced by improving livelihood and human well-being of the people.

Soil security is the assurance to access to basic soil resources that can provide food, and ecosystem services to meet essential human needs for healthy living.

2 Some Historical Collapsed Antecedents of Ancient Societies and Civilizations

There have been 10～30 different civilizations that collapse. The four interesting and illustrative civilizations are as follows: ① Indus Valley Harappa; ② Mesopotamia and the Fertile Crescent; ③ Mesoamerica (Olmec, Maya, Chacoans, Huorí), North Africa, Mediterranean; ④ Hittite Empire, Anatolia.

A large number of references show that the development and decline of those civilizations were closely related to the development, protection, utilization, loss, damage and destroy of local soil resource.

2.1 Aztecs and Mayan Empires

They developed some techniques of soil and water conservation (SWC) to increase agricultural yields and to adapt to their climatic and topographic conditions. There is no agreement on the reasons of their collapse. It is still a kind of mystery. In 1940s, S. Cook proposed the hypothesis of considering the soil exhaustion to explain the decline. Today, deforestation, erosion, forced population displacement and climatic considerations are gaining acceptance.

First, they used to slash and burn forests, then extensive terracing. More than climatic changes, we have to consider the impact of droughts. The center terraces of the Mexico-Teotihuacan valley lie between 600 BC and 1000 AD. They used the traditional “milpa” system.

In the Mayas Lowlands, there were advanced water conservation systems and many terraces. Sediments shows two different episodes: the first correspond to forest clearing; the second to agriculture intensification, but it is only agricultural methods, including soil and water conservation, could not longer avoid erosion, and not sustain population, either.

Failures of their methods of soil conservation in deforested slopes are as below. Soil and water conservation systems were not adequate (not maintained) to avoid massive erosion and causing serious impact by sediment delivery to the agricultural areas in the valleys. Areas populated most densely in pre-conquest times had the worst exhaustion. Bench terraces is an old soil and water conservation system. It seems Phoenicians made the first bench terraces probably in fifteenth century BC.

2.2 The Inca Empire

The Inca Empire developed an outstanding and elaborated system of keeping productive soil in place in a very complicated and challenging environment. High and varied agricultural yields could be obtained under a complex social communal system. However previous to the Spanish conquest, the empire was in declining owing to civil wars and demographic recession. Atahualpa reached the throne after a devastating civil war against his brother Huascar.

There were many traditional Andenes (which is the traditional Spanish name for bench terraces in South America) system in Perú. The social and political instability trigger the gradual abandon of the use and maintenance of Andenes. The abandon of Andenes, under a complex communal system, was a factor contributing to the fall of the Inca empire. In this case, it was fighting a losing battle for the technical and social effort of heroic measures to save their soil.

2.3 The Indus Valley Harappa Civilization

Chris Sloan, an artist, reconstructed the gateway and drain at the city of Harappa. Lacking of capabilities on flood control for agricultural use, change on climate patterns, violence and disease played a key role in the collapse of the Harappa.

It was a civilization larger than Mesopotamia and Egypt. The reasons of the collapse are not clear and under general discussion. But recently, the theory of changes in the climatic pattern along with difficulties in the control of sediments from Indus river and the maintenance of irrigation systems is gaining adepts. It was good developed in 5200 BC and collapsed 3 900～3 000 years ago.

It lacked of adaptation to a changing environment by adequate water-soil management. Initially flood supported farming in the Indus Valley, and allowed the development of agricultural surplus and the growth of Harappa’s cities (rich soils). They did not develop irrigation capabilities relying on the summer monsoons floods. But the monsoons started kept shifting southward and became more erratic (increase of aridity) for adequate agricultural yields. The residents migrated eastward to the Ganges basin where they established smaller villages and isolated farms. The small surplus produced in these small communities did not allow development of trade, and the cities died out. The place at present is a barren country. Now, the climate is apparently the same as in ancient time, but the land is not.

2.4 The Fertile Crescent and Mesopotamia

The Fertile Crescent was a cradle of civilization, and also a permanent battle ground and one area always invaded and conquered by many civilizations, but it seems they did not collapse by that. It was mainly because the difficulties on the management of increasing sediments and the inadequate maintenance of their irrigation channels. The other crucial problem was the salinization of the soil.

Mesopotamia is a case of sediments and salt of the Fertile Crescent. Ur, ancient sea port is now 241.40 kilometers inland. Mesopotamian cuneiform clay tablets tell of crop damage due to salts and the earth is turning white.

The decline and collapse of the Fertile Crescent could be considered as a case of unrestrained download of sediments and a unrestrained increase of salt in their soils.

3 Discussion: Why Civilizations Collapsed？

There are many viewpoints on the causes of human civilizations development and decline, and scientists in different research areas made different explanations. It should be further explored, studied and verified. In our opinion, the following patterns and most repeated circumstances are their possible causes: ① strict dependence on rich land to rise; ② soil mismanagement; ③ soil resources depletion; ④ decline in agricultural yields; ⑤ instability and conflicts.

It would be better to take lessons from the past: deepen into the root of the problem which basically seems to be linked between soil degradation and food security. There is an enormous amount and variety of theories to explain the collapse of civilizations.

The socio-economic points of view are as below: ① declining in productivity systems and economic activities; ② declining marginal returns; ③ catastrophes (earthquakes, climatic component…); ④ insufficient response to circumstances; ⑤ intruders, conquerors; ⑥ chance concatenation of events; ⑦ social dysfunctions…; ⑧ little consideration to “l(fā)and resources depletion”.

However as for the “declining marginal returns” theory, historians do not agree on the specific reasons of the fall of civilizations.

There are some enlightening references on “erosion of civilizations”: ① Several writers have documented the decline of civilizations throughout history in parallel with the destruction of their soil. For example, Edward Hyams described people as “parasites” of the soil inSoilandCivilization(1952). ② Already in the first AD, Collumela wrote that “soil will retain its fertility indefinitely if properly cared for and frequently manured”. All of them contributed with significant thoughts, reflections and conceptual developments to the historical role of soil conservation in human history and its influence in the rise and fall of civilizations.

The link between “climate changes” and the collapse of civilizations is not well supported; Severe drought affected the Late Bronze Age; Empires seems to have being a factor influencing the decline of Egyptian, Mycenaean, Hittites, Indus Valley and the Crescent Fertile civilizations; They had not the capability of land management to adapt a changing conditions. They lacked the necessary reactions. Sometimes, they acted too little, or too late.

The flourishing of civilization about 10 000～7 000 years ago, critically depended on stabilization of climate conditions. Egyptian, Mesopotamian and Indus civilizations about 4 200 years ago declined due to severe drought (Bronze Age). However more than “climate change”, which basically remained the same, the fundamental factor seems to be the misuse of the land.

Many ancient civilizations mined soil and accelerated soil erosion well beyond the pace of soil formation and soil capacity to sustain life. We can say that man has succeeded into increasing locally and temporarily the productivity of soil, but in general and until now has failed to develop a real sustainable land management. History tells us that management and conservation of the land has given life expectancy of civilizations. As Jared Diamond described in his bookCollapse, societies choose to fail or survive. Such problems are not just ancient history. Recent historical great catastrophes are: Dust Bowl in 1930s and Droughts of Sahel in 1970s. We know now all the facts by which the soil is destroyed, but are modern world going to suffer the same fate?

4 Conclusions

All classes of land worldwide require some type of transformation so as to obtain some adaptations for agriculture use. It can be some type of terracing, leveling, runoff control, irrigation, contour leveling, etc. When the users of these measures abandon their maintenance, a declining trend could start leading to degradation processes and to the declining of productivity.

Scientists do not agree on the reasons of the collapse of civilizations. Little consideration has been taken to the effects of soil mismanagement and land resources depletion. However history reveals soil and water conservation is crucial to the permanence of any civilizations. The management and conservation of the land has shaped human history.

Today we confront a climate change, increasing land degradation and shrinking land available for food production. Considering the dimension and perspectives of present world problems, history tells us that we cannot ignore, not repeat mistakes of the past, including errors on soil and water conservation issues.

To keep and improve healthy living conditions for all mankind requires learning from the bad performances and incorrect assumptions of the past. We are more aware now of the likely consequences of our choices than any society in history. Would not it be embarrassing if we continued to make the wrong ones? Soil and water conservation has become a necessity for maintaining the sustainable development of social civilization. We cannot do little and too late.

ThispaperwaspresentedatTheThirdWorldConferenceofWASWAC,Belgrade,Serbia,August2016.Itisatightsummaryofinitialresultsofaninternationalproject(CIDE-Spain)ontheinteractionsofsoildegradationandthedeclineofcivilizations.Thereferencesareomitted.

Mr.JoséLuisRubioistheDeputyPresidentoftheWorldAssociationofSoilandWaterConservation(WASWAC),ViceChairmanofEuropeanSoilBureauNetwork-ESBN(JRC,EC),ImmediatePastPresidentoftheEuropeanSocietyforSoilConservation(ESSC)theChiefoftheDepartmentofSoilDegradationandSoilConservationatCIDE(CSIC,UniversitatdeValencia,GeneralitatValenciana).

Application of SCS Model to Simulate Runoff in Slope Field at Shiqiao Small Watershed in Bijie City of Guizhou Province

ZHANG Xingqi， XU Pengcheng, GU Jingran

(SchoolofGeographicandOceanographicSciences,NanjingUniversity,Nanjing,Jiangsu210023,China)

[Objective] The rule of runoff formation in karst area was studied to get the parameters of space cooperation system(SCS) model and to test whether it is applicable in the study area. [Methods] Based on field observations of runoff and rainfall characteristics, rainfall events in the study area were classified. Runoff generated on the slopes was simulated by using the SCS model. [Results] (1) Rainfall events in the study area can be divided into four types according to event-based rainfall concentration, namely the pre-type(typeⅠ), medium-type(typeⅡ), back-type(typeⅢ) and even-type(typeⅣ). (2) The value of the initial loss coefficient(λ) was 0.01 or 0.05, much less than 0.2, when simulating runoff on the slopes in the study area by using the SCS model. (3) The curve number(CN) values of typeⅠ, typeⅡ and typeⅢ rainfall showed small changes as slop length increased, while the CN value of typeⅣrainfall tended to increase. The CN values of typeⅠand typeⅡrainfalls tended to reduce with the increase of slope gradient, the CN values of typeⅢ and typeⅣ rainfall did not increased obviously with the increase in slope gradient. [Conclusion] Based on SCS model, the correlation coefficient and model efficiency coefficient of simulated runoff and observed runoff are both high, which indicates the model performed well.

rainfall type; SCS model; runoff simulation; Guizhou Province

2016-07-15

2016-11-24

貴州省水利廳重點科研項目“貴州喀斯特地區(qū)土壤侵蝕機理研究”( KJZD200801) ， “貴州喀斯特地區(qū)坡耕地坡度、坡長與水土流失關系研究”( KT201007) ， “西南喀斯特地區(qū)土壤侵蝕機理及水土流失預測”( 2006200); 國家自然科學基金資助項目“喀斯特地區(qū)人類活動主導下的生態(tài)環(huán)境變化與流域水文循環(huán)響應耦合機理研究”( 41371045); 江蘇高校優(yōu)勢學科建設工程資助項目

張興奇(1964—),男(漢族),貴州省仁懷市人,博士,副教授,主要從事水資源與水土保持研究。E-mail:zxqrh@nju.edu.cn。

B

1000-288X(2017)03-0321-08

S273.1，S157.1