城市景觀設(shè)計與研究之代爾夫特方法
——荷蘭代爾夫特理工大學(xué)斯特芬 奈豪斯博士專訪

采訪：張博雅 校對：熊亮

本文中的代爾夫特方法特指荷蘭代爾夫特理工大學(xué)建筑與建成環(huán)境學(xué)院自1947年以來所發(fā)展的一種用于理解和設(shè)計城市景觀的風(fēng)景園林學(xué)手段（圖1～2）。代爾夫特方法受到典型的荷蘭地理環(huán)境和空間規(guī)劃傳統(tǒng)的影響，將城市景觀看作在不同尺度上的連續(xù)體，以“針對設(shè)計的研究和通過設(shè)計的研究”為主要的教學(xué)和研究策略，以繪圖和圖析（mapping）①作為重要的思考工具。作為荷蘭代爾夫特理工大學(xué)建筑與建成環(huán)境學(xué)院風(fēng)景園林學(xué)科研究帶頭人，奈豪斯博士主要致力于基于景觀的城市學(xué)、GIS的應(yīng)用以及視覺景觀研究。他曾受邀參與在北京林業(yè)大學(xué)舉行的2019世界風(fēng)景園林師高峰講壇?！讹L(fēng)景園林》雜志有幸邀請到他參與此次專訪，向讀者介紹代爾夫特方法的背景和原則，并結(jié)合實踐案例展示其應(yīng)用方法。

LAJ：《風(fēng)景園林》雜志

Nijhuis：斯特芬 奈豪斯

LAJ：非常感謝您參與這次采訪。關(guān)于代爾夫特方法，您曾提到，它的誕生受到了形態(tài)學(xué)研究的影響。那么您能否先為我們簡單介紹一下什么是形態(tài)學(xué)研究，以及它與區(qū)域景觀的聯(lián)系呢？

Nijhuis：在形態(tài)學(xué)研究的視角下，貫穿各尺度的景觀都可被看作一種建筑式的構(gòu)成。當(dāng)我們這樣去看待景觀時，我們或許沒法去重構(gòu)（既有景觀背后的）“設(shè)計師”的思考，但我們可以學(xué)習(xí)其構(gòu)成是如何運行的，以及它具有怎樣的空間品質(zhì)、生態(tài)品質(zhì)和社會品質(zhì)。景觀是由一系列結(jié)構(gòu)和過程組成的系統(tǒng)。景觀的結(jié)構(gòu)，或者說形態(tài)，可以作為研究景觀的（形成）過程和發(fā)展的出發(fā)點。這就是形態(tài)學(xué)研究的課題。

在歐洲，形態(tài)學(xué)研究自20世紀60年代開始發(fā)展。意大利建筑師穆拉托里（Muratori）和移居英國的德國地理學(xué)家康澤恩（Conzen）是早期先驅(qū)，后來又有來自法國思想學(xué)校的卡斯特斯（Castex）等人。在代爾夫特，很多學(xué)者受到他們的影響，并給出了自己的詮釋[1]。早期代表包括帕博（Palmboom）、斯蒂恩伯恩（Steenbergen）、范 德 霍溫（Van der Hoeven）、盧沃（Louwe）和范 沃登（Van Voorden，圖3）。（在風(fēng)景園林領(lǐng)域）早期研究者的經(jīng)驗證明，形態(tài)學(xué)方法對于景觀設(shè)計研究也是有效的手段，這些先驅(qū)包括杰里科（Jellicoe）[2]、弗蘭克（Franck）[3]，卡斯克（Kask）[4]和哈茲赫斯特（Hazlehurst）[5]。代爾夫特風(fēng)景園林教研組基于對人造景觀的精確幾何描繪，發(fā)展了一種專注景觀的建筑學(xué)構(gòu)成的形態(tài)分析方法，其中的代表是斯蒂恩博恩（Steenbergen）及其同僚。這種方法系統(tǒng)分析了景觀的4種建筑學(xué)類型：基本形態(tài)、空間形態(tài)、象征形態(tài)以及程序形態(tài)。基本形態(tài)指的是在設(shè)計底圖上對人工或自然景觀本底的地形進行簡化、合理化和激活的方式（圖4）?？臻g形態(tài)則關(guān)注三維景觀空間的形態(tài)及運行。象征形態(tài)指將景觀的建筑學(xué)結(jié)構(gòu)型形態(tài)與圖像意象、神話象征和自然元素相關(guān)聯(lián)的方式。最后是程序形態(tài)，它產(chǎn)生了功能分區(qū)或與人、物流相關(guān)的空間布局，以及其他運動的功能性形態(tài)。這些類型系統(tǒng)地詮釋了景觀的建筑學(xué)形態(tài)與人的感知之間的復(fù)雜關(guān)系。它不僅僅是對形態(tài)的分析和描述，也是對功能和構(gòu)建方式的解讀。諸如《建筑與景觀》（Architecture and Landscape）、《景觀構(gòu)建》（Composing Landscapes）、《荷蘭圩田地圖集》（The Polder Atlas of Netherlands）等著作均是基于此方法的著名代表。

1 阿姆斯特丹擴張規(guī)劃（1834年）是范 伊斯特恩和范 路豪森的作品，他們也是代爾夫特理工大學(xué)城市系的聯(lián)合創(chuàng)始人，是基于自然景觀和文化景觀進行規(guī)劃的現(xiàn)代主義城市規(guī)劃代表人物The Amsterdam Extension Plan (AUP) 1934 created by Van Eesteren and Van Lohuizen, some of the founding fathers of the Department of Urbanism, is an icon of modernist urban planning and was based on extensive knowledge of the natural and cultural landscape

2 代爾夫特形態(tài)學(xué)方法案例：城市景觀的地形地貌特征分析An example of the Delft morphological approach analysing the territorial and topographical qualities of the urban landscape

3 代爾夫特風(fēng)景園林教研組的典型分析圖解：法國梅倫沃勒維貢特府邸對稱軸線形態(tài)分析A typical drawing from the Delft Landscape Architecture group. Morphological analysis of the axis of symmetry at Vaux-le-Vicomte, Melun (France)

4 這張荷蘭圩田地圖反映人工水利景觀的模式與地表海拔高度緊密相關(guān)。藍色區(qū)域是低于海平面的國土，棕黃色區(qū)域是高于海平面的國土，深藍色的線代表堤壩Polder map of the Netherlands. The pattern of this manmade hydraulic landscape corresponds closely with the elevation. Blue areas are below sea-level, brown areas are above sea-level. The dark blue lines represent dikes

LAJ：能在不同尺度應(yīng)用同一方法還挺有趣的。

Nijhuis：是的，這個方法將景觀視為尺度連續(xù)體。這首先根植于荷蘭的地理環(huán)境，我們有大片低于海平面的人造陸地。由于土壤條件和水文條件都比較苛刻，我們必須根據(jù)水土條件精心規(guī)劃建筑和城市。這樣，建筑、城市和景觀便自然而然地緊密相連。當(dāng)劃定新設(shè)計中的骨架時，我們可以對其所在的更大尺度的景觀本底進行分析，并從中獲得啟示。

5 英國威爾特郡斯托海德園主要游覽路線沿線的海拔高度與空間明暗體驗的對應(yīng)關(guān)系A(chǔ)nalysing the experience of light and shade and the relationship with the height gradient along the main routes at Stourhead landscape garden, Wiltshire (UK)

6 英國威爾特郡斯托海德園特定視點的三維視域分析與實際景物的對應(yīng)關(guān)系Three-dimensional visibility analysis, and corresponding field of view, of a famous view point at Stourhead landscape garden, Wiltshire (UK)

另外，就算是不同區(qū)域景觀是基于相同的原則建造的，它們的形態(tài)表達也受到環(huán)境的影響。以電影為例，在成千上萬的電影中，只有大概7～8種制作電影的方式。我們荷蘭西部和北部的低地圩田景觀也是一樣（圖5）。如果你戴上“X光眼鏡”去看，你會發(fā)現(xiàn)不同圩田治水的原則、為農(nóng)業(yè)生產(chǎn)創(chuàng)造條件的方法，都大同小異，但最終的空間形態(tài)卻千差萬別。但居住在不同地區(qū)的人們擁有不同的社會文化、政治傳統(tǒng)和治理方式。這些環(huán)境要素讓景觀變得不同。比如，世界各地都有圩田景觀，日本北部的八郎潟圩田的建造即受到了荷蘭東北圩田的啟發(fā)。在區(qū)域尺度上，他們乍看起來極其相似。放大看時，你會發(fā)現(xiàn)日本的圩田用于種植水稻，這意味著水會階段性地淹沒土地。而在荷蘭，圩田水管理致力于維持水位穩(wěn)定。這就是（在不同尺度看問題的）有趣之處，（雖然在大尺度上應(yīng)用）同一原則，（但在中小尺度有）不同的文化表達。所以，我們訓(xùn)練學(xué)生跨越不同尺度看問題，這樣他們才能夠發(fā)現(xiàn)不同尺度上的異同。

LAJ：在不同情況下應(yīng)用代爾夫特方法，是否會根據(jù)情況相應(yīng)調(diào)整？

Nijhuis：隨著時間推移，這一方法也在進化。比如，對這一方法的批判主要集中在它缺乏對空間感官體驗的關(guān)注。具體來說，是指客觀物理現(xiàn)實（歐幾里得空間）和其視覺表現(xiàn)（主觀體驗的空間）的不同，以及運動在空間體驗中扮演的角色。此外，僅就物理空間而言，同一空間的空間構(gòu)成如何隨著時間變化，也未經(jīng)挖掘。這些批判可以作為改進分析框架的線索，也可以作為修正和豐富最早的幾種景觀的建筑學(xué)類型的基礎(chǔ)。我和我的同事們在這方面有了一些進展，試圖理解如何更好地解決這些問題。比如說，德 維特（De Wit）的近期研究發(fā)展了這個分析框架，表明了感官體驗可以在景觀構(gòu)成分析中扮演重要角色，這也是代爾夫特方法發(fā)展的代表。GIS等數(shù)字技術(shù)也為分析景觀空間的三維體驗提供了有趣的線索（圖6）。當(dāng)你在地圖上看到類似貝姆斯特圩田的景觀時，你可能會認為你（在空間中）會感受到方格網(wǎng)，但當(dāng)你實地體驗時，你永遠也感受不到方格網(wǎng)，因為你是在水平視角去體驗的。換句話說，測繪到的景觀和人們感知到的景觀是兩回事。有了像GIS一樣的工具，你可以建立相當(dāng)細致的三維模型，把空間分析得很透徹。你可以在水平視角上模擬人的空間感知。這也正是我在研究中去探索的[6]。比如，我對斯托海德園的分析表明，在GIS工具的幫助下，即使是面對老生常談的研究對象，也可得出新的洞見。GIS工具讓我們不僅能在靜態(tài)視角下去研究觀者的體驗，也可研究特定路徑上的動態(tài)體驗（圖7）。我對（斯托海德園的）不同歷史時段進行分析，這要求我對其歷史發(fā)展過程進行精確的數(shù)字化重建。通過比較分析結(jié)果，我們可以了解它的空間構(gòu)成具體是如何隨著時間變化的，這些信息可幫助該園進行管理維護方面的決策。只要你開始使用GIS工具進行分析，你就可以得出有趣的新結(jié)論。比如我所發(fā)現(xiàn)的，游園路徑上的大多數(shù)視點都遵循一個特定規(guī)則，那就是它們的主要構(gòu)圖都在大約30°的水平視角范圍內(nèi)。如果你懂視光學(xué)的話，你會知道這是一個你剛好（不需轉(zhuǎn)頭）能一眼看清視野內(nèi)事物的角度。如果你分析那些根據(jù)克勞德 洛林（Claude Lorrain）等的繪畫而創(chuàng)作的景觀，你會發(fā)現(xiàn)特定視點的景觀的主要內(nèi)容也在30°視角的范圍內(nèi)。這意味著這個花園的確是根據(jù)畫家的觀看方式去設(shè)計的。雖然這樣的說法一直存在，但通過這樣的分析我們才能確信它是真的。

7 通過設(shè)計的研究并不僅僅是設(shè)計，也是對設(shè)計問題的有效解、高效解的系統(tǒng)探索。在這個探索過程中，按照溯因邏輯原則，問題和目標都有可能被修正或是改寫Research through design does not consist purely of design, but also of a systematic search for the most effective and efficient solution to the problem that has been posed. During this search, following principles of abductive logic, both the problem and the objective can be refined or changed

8 珠江三角洲是世界上城市化速度最快的三角洲。紅色的區(qū)域是已經(jīng)城市化的地區(qū)。大多數(shù)城市化發(fā)生在低地區(qū)域（海拔10 m的區(qū)域，圖中藍灰色的范圍），這一區(qū)域更容易遭受洪水侵襲The Pearl River Delta is the fastest urbanizing delta in the world. The red zones indicate urbanization. Most of the urbanization (in red) takes place in the flood prone lowlands (+10 meter zone, in dark blue), the zone with increased flood risk

LAJ：您經(jīng)常強調(diào)將研究和設(shè)計進行結(jié)合，能否深入介紹一下？

Nijhuis：在代爾夫特，我們認為設(shè)計和研究緊密聯(lián)系，而不是互相割裂的。因此，我們常使用“設(shè)計研究”（design research）和“通過設(shè)計進行研究”（research through design）這樣的術(shù)語。前者指的是為進行設(shè)計而獲取知識，通過對項目語境和既有案例的分析得出設(shè)計原則。后者指的是設(shè)計過程本身是一種將空間問題可視化的工具，（可用于）探索可能性并產(chǎn)生新的解決方案，即“通過設(shè)計進行研究”。如需深入了解，可參見我和同事專門發(fā)表的開源論文[7]（圖8）。

LAJ：如果人們想要嘗試應(yīng)用這一方法，您能否推薦一些工具？

Nijhuis：草圖（sketches）、繪畫（drawing）和圖析的確是很有效的思考工具。通過這些工具，設(shè)計師可以分析、評估現(xiàn)狀、設(shè)計方案和既有案例，以及產(chǎn)生新的設(shè)計。圖析的作用尤其強大。如果你想要理解某處景觀，我想地圖可能是繞不開的。問題是，你如何使用地圖？你可以去書店買一本印刷精美的地圖集，但你真的理解你看到的內(nèi)容嗎？這就引入了圖析的概念。

圖析的一個典型操作是“地圖拆解”（map dissection），意思是你只選取地圖上的幾種要素，然后去理解它們的模式和背后的意義。這不是在開關(guān)圖層，而是真的去讀懂這條線和那條線的意味及其之間的關(guān)系。你可以說他是一個選擇和消化的過程。為了理解（地圖），你需要有這個過程。另一個典型操作是比較，比如說比較不同的情境、模式或者歷史時期。你還可以嘗試把空間信息和非空間信息進行并置，我把這種手段稱為“交叉參照”（cross-reference mapping）。

此外，我還想強調(diào)視覺化思考和視覺溝通的區(qū)別。我?？吹饺藗?yōu)榱俗寗e人理解（自己的想法），更關(guān)注用這些工具進行視覺溝通，卻很少把它們用于視覺化思考。一張用于思考的草圖可能非常潦草。在視覺化思考的過程中，你通過和媒介互動獲得新的理解，也就是“見所未見”。當(dāng)你理解了以后，下一步才是如何把你的理解用一種漂亮的方式傳遞給他人。

LAJ：您能否以一個項目為例，介紹區(qū)域景觀的設(shè)計及研究？

9 順德某地區(qū)的設(shè)計研究：新的區(qū)域規(guī)劃整合了水管理系統(tǒng)、基礎(chǔ)設(shè)施、生態(tài)系統(tǒng)、農(nóng)業(yè)和住房建設(shè)的發(fā)展。這些剖面圖展示了水敏地區(qū)區(qū)域景觀發(fā)展的不同階段Research through design for a region near Shunde. A new integrated regional plan that takes the sustainable development of the water system, new infrastructure, ecosystem, agriculture, and housing into consideration. The sections show the different phases of the water-sensitive landscape development

Nijhuis：在基于景觀的區(qū)域設(shè)計或城市規(guī)劃設(shè)計中，設(shè)計和研究密不可分。我們把這一方法用于為政府和私人業(yè)主進行的研究中，項目遍及歐洲、美國和中國。比如我們近期啟動了中國國家自然科學(xué)基金委員會（NSFC）、荷蘭研究理事會（NWO）、英國工程與自然科學(xué)研究理事會（EPSRC）的聯(lián)合研究項目：適應(yīng)性城市更新（Adaptive Urban Transformation, AUT）——珠江三角洲區(qū)域的區(qū)域治理、空間戰(zhàn)略以及城市景觀動態(tài)變化。這一項目由華南理工大學(xué)的孫一民教授、英國謝菲爾德大學(xué)的?？ㄌ?蘭格（Eckart Lange）教授和我牽頭。AUT應(yīng)用了基于景觀的區(qū)域設(shè)計策略作為一種綜合的、多尺度的設(shè)計和規(guī)劃方法（圖9）。這樣做可以整合各部門規(guī)劃，推進城市景觀和鄉(xiāng)村景觀的轉(zhuǎn)型過程，形成協(xié)調(diào)的可持續(xù)發(fā)展?；诰坝^的區(qū)域設(shè)計方法把景觀本底作為建立可持續(xù)的城市化三角洲地區(qū)的基本條件，這被看作是塑造區(qū)域的物質(zhì)空間形態(tài)的重要策略。它也使得通過空間規(guī)劃和空間設(shè)計塑造長期可持續(xù)的城市景觀成為可能?？傊瑓^(qū)域設(shè)計需要跨學(xué)科的合作，這不僅能保證連貫的、可持續(xù)的地區(qū)發(fā)展，也能引導(dǎo)社會經(jīng)濟發(fā)展和環(huán)境變化的方向，并建立基于區(qū)域特質(zhì)的獨特的地區(qū)身份。我在《風(fēng)景園林》雜志發(fā)表的另一篇文章詳述了具體的研究[8]。

以順德為例，?；~塘系統(tǒng)是這里重要的景觀特征。如今，這些結(jié)構(gòu)特征大多被城市化進程所掩蓋，（很多工程）直接填平了場地并在上方進行建設(shè)。這造成了在蓄水功能、地區(qū)生態(tài)和房屋基礎(chǔ)等方面一系列的問題。因此，我們提出了基于景觀的區(qū)域設(shè)計策略，它將水系統(tǒng)和景觀結(jié)構(gòu)視為基礎(chǔ)，產(chǎn)生兼具水敏性和地域性的城市發(fā)展模式，在這種模式中，智慧交通概念占據(jù)著重要位置。

LAJ：非常感謝您接受這次采訪。這些年來，您曾經(jīng)多次來過中國，您也在世界各地參與合作過研究項目。最后想請您談?wù)?，在您看來，中國的區(qū)域景觀研究和設(shè)計有無特殊之處？

Nijhuis：在中國，你們有很好的分析城市、文化景觀與其所在自然環(huán)境的關(guān)系的研究傳統(tǒng)。中國很多高校的知名學(xué)者都有這方面的著作。你們在建設(shè)新城和新的區(qū)域景觀上也展現(xiàn)了極大的雄心。我認為，通過從成功案例中獲得的經(jīng)驗和設(shè)計原則，具有極大的潛力來應(yīng)對目前氣候變化和城市化帶來的挑戰(zhàn)。我想基于景觀的區(qū)域設(shè)計策略也可助力于中國的新城建設(shè)和城市更新。這一策略尊重規(guī)律，將景觀本底視為城市化發(fā)展的基礎(chǔ)。它跨越各尺度、連接多學(xué)科，并且以科學(xué)知識為依據(jù)。這一策略充分利用了規(guī)劃概念和設(shè)計師知識之間的互動，來解決城市景觀的復(fù)雜問題，使其面向未來、繁榮昌盛。非常感謝你們的采訪，我很期待更多的知識交流和互相啟發(fā)。

注釋(Notes)：

① 為方便理解，筆者將“Mapping”譯為圖析。也有學(xué)者將其譯為“地圖術(shù)”。

圖片來源：

圖1為私人收藏；圖2由帕博繪制；圖3由斯蒂恩博恩和雷繪制；圖4～6由斯特芬 奈豪斯繪制；圖7由斯特芬 奈豪斯等繪制；圖8基于SRTM（30 m）和Hydropolis由斯特芬 奈豪斯繪制；圖9由孫傳致繪制。

Research and Design of Urban Landscapes: The Delft Approach — Interview with Dr. Steffen Nijhuis,Delft University of Technology

Interviewer: ZHANG Boya Proofreader: XIONG Liang

In this essay, the Delft Approach refers to a particular landscape architecture approach to understand and design urban landscapes, which is developed in the Faculty of Architecture and Built Environment at Delft University of Technology(the Netherlands) since 1947 (Fig. 1-2). The Delft approach focusses on urban landscapes as a scale continuum, employs design research and research through design as important teaching and research strategies, and regards mapping and drawing as important tools for thinking. The typical Dutch geographic context and spatial planning traditions are the foundation for this approach. As head of research of Landscape Architecture in the Faculty of Architecture and Built Environment at Delft University of Technology (the Netherlands), Dr.Nijhuis has expertise in landscape-based urbanism,GIS applications and visual landscape research. He was invited to participate in the 2019 International Landscape Architecture Symposium held at Beijing Forestry University.Landscape ArchitectureJournal is honored to invite him to this inclusive interview to introduce the background and principles of this approach, and illustrate the application of this method with projects.

LAJ: Landscape Architecture Journal

Nijhuis: Steffen Nijhuis

LAJ：Thank you very much for joining this interview. As you mentioned, the birth of Delft Approach was influenced by morphological studies. So, could you please start with briefing us the term of morphological study,and how it relates to regional landscapes?

Nijhuis:In the perspective of morphological studies, landscapes throughout the scales can be understood as architectural compositions. By interpreting the landscape in such a way you may not be able to reconstruct the thoughts of the ‘designer’,but you can learn how the composition functions and what the spatial, ecological and social qualities are. The landscape is a system that constitutes of structures and processes. The structures or forms can be taken as a starting point to understand the development over time and the involved processes.This is the subject of morphological study.

In the European context morphological studies evolved from 1960s onwards, with Italian architect Muratori and British geographer Conzen as front runners. Later, there was Castex et al.from French schools of thought. In Delft many scholars were influenced by them, and gave it their own interpretation[1]. Early representatives include Palmboom, Steenbergen, Van der Hoeven,Louwe and Van Voorden (Fig. 3). Morphology studies have proven very useful in landscape design research as exemplified by protagonists of this approach such as Shepherd and Jellicoe[2], Franck[3],Kask[4]and Hazlehurst[5]. In Delft the landscape architecture group developed a particular approach that is focused on the formal analysis of landscape architectonic compositions as exemplified by Steenbergen and associates, and is grounded in the notion of precise geometric delineation of designed landscapes. The method consists of the systematic analysis of four landscape architectonic categories:the basic form, the spatial form, the symbolic form and the programmatic form. The basic form is the way in which the topography of the natural landscape or the man-made landscape is reduced,rationalized and activated in the ground plan of the design (Fig. 4). The spatial form is about the form and functioning of three-dimensional landscape space. The symbolic form refers to the way in which iconographic and mythological images and architectonic structural forms are connected with one another and with elements from nature. And finally: the programmatic form leads to a functional zoning or layout in relation to logistics and functional patterns of movement. These categories systematically lay out the relationship between the various aspects of the landscape architectonic form and its perception. It is not only about analysis and description of its form, but also its function and the way it has been made. Publications likeArchitecture and Landscape,Composing LandscapesandThe Polder Atlas of Netherlandsare some well-known examples of this approach.

LAJ: It’s interesting that the same method can be applied through different scales.

Nijhuis:Yes, this approach focuses the landscape as a scale continuum. Firstly it roots in Dutch geographic context that we have vast manmade lands below sea level. Due to the particular soil and hydrological conditions, the buildings and cities had to be carefully arranged according to the conditions. Thus the buildings, cities and the landscape are inherently connected. When develop the structural elements of new designs, lessons can be drawn from analysis of the landscape on a bigger scale.

Also, even when the landscapes are built employing similar principles, the expression of form makes it contextual. It’s like the movies. We have thousands of movies, but there’s probably seven or eight ways on how to make a movie. That’s the same with lowland landscapes or polder landscape we have in the west and north of the Netherlands (Fig.5). So when you look at different polders with “X-ray glasses” you’ll see that we have similar principles on how to deal with water, how to create conditions for agricultural practice, but result in very different spatial expressions. The people living there are very different in social terms, in political terms and in governance terms. That makes landscape particular.For instance: polder landscapes can be found all over the world. The Hachirogata Polder in the north of Japan is inspired by the Noordoostpolder in the Netherlands. At a first glance, at the regional scale,they look quite similar, but when you look more in detail you see different land use like rice paddies with a fundamental different water system focused on temporal flooding. Whereas in the Dutch case the water system is focused on stable water tables. It’s like a kind of cultural twist to the principle, which makes it very interesting. Therefore we train our students to look through scales, being able to see differences and similarities at different scale levels.

LAJ: When we use Delft approach, is there maybe any adaption in different circumstances?

Nijhuis:Well, it’s evolving over time. For instance critique on the approach focused on the lack of attention to sensorial aspects of the composition. More specific, and relevant to formal analysis, is that the role of movement and the fundamental difference between the physical, metric reality (Euclidean space) and its visual appearance (perceived space) is neglected.The morphogenetic aspect, the development of the composition through time, also remains underdeveloped. These critiques can be considered clues for the development of the analytical framework and served as a basis to redefine and enrich the initial landscape architectonic categories for analysis. Over time my colleagues and myself started to get into that, and try to understand how these aspects can be addressed better. For instance,development of the framework is exemplified in the recent work of De Wit, were sensorial aspects play a crucial role in the analysis of landscape architecture compositions. Also the use of digital technology like GIS provides interesting clues to analyze on the three-dimensional experience of landscape (Fig. 6). When you see a landscape like the Beemster polder on a map, you’ll probably think you will experience it as a grid, but when you go there you’ll never experience the grid because it’s from the horizontal point of view. In other words, the measured landscape is very different from the perceived landscape. With tools like GIS,you can build a highly detailed three-dimensional model, and analyze the landscape from inside out. You can mimic the human perception from a horizontal perspective. That’s exactly what I do in my research[6]. For instance my study on Stourhead landscape garden exemplifies how GIS can help to derive new insights from a landscape design that is studied exhaustively. With the use of GIS we were able to uncover fascinating aspects of the three-dimensional composition from an observers point of view, from static viewpoints, but also from routes (Fig. 7). We did this for different time periods, which required precise reconstruction of the park over time. By comparing the analytical results insight was gained in the development of the composition over time, which in turn can inform management decisions for the park. When you start to analyzing these things with GIS, you can draw very interesting new conclusions. Like what I discovered, most views along the route obey a certain law, that their compositions are constructed within an visual horizontal angle of about 30 degrees. If you know optics, 30 degrees is the angle where you can recognize things in a glance. When you analyze these views, inspired by the paintings of e.g. Claude Lorrain, you will see that the most important compositional elements are also within this 30 degrees. So that means this garden was indeed designed by the eye of a painter.It’s often said to be so, and now you can really underpin it by this kind of analytical operations.

LAJ: You often address the importance of connecting research and design, can you elaborate on this a little bit?

Nijhuis:Yes, in Delft we regard research and design to be closely related and not as two separate worlds. Therefore we often use the terms “design research” and “research through design”. The first is focused on gaining knowledge for the design and entails contextual analysis and precedent or case study to find design principles. Next to this form the design process itself is employed as a vehicle to frame spatial problems visually, explore possibilities and to generate solutions; research through design[7](Fig. 8).

LAJ: For people who would like to try this approach, what tools would you recommend for them?

Nijhuis:Sketching, drawing and mapping are really powerful tools for thinking in design. With these tools, designers can analyze and evaluate situations, plans or precedents, and generate designs.Mapping is especially a powerful tool. When you want to understand the landscape, I think maps are inevitable. The question is, how do you use a map?You can go to the bookstore and buy a really nice atlas, but do you understand what you are looking at? Here the idea of mapping comes in.

A typical analytical operation is map dissection, which means you only select a few elements from the map, and then you start to understand what kind of patterns they are, and what their meaning is. It’s not like clicking one layer on and off, but really to understand the meaning of this line and that line, and what is the relationship in between. It’s a procedure of selecting and digestion you can say. So, you need to do it in order to understand it. Another possibility is comparing,like comparing different situations, patterns, or different time periods. You can also do what I called cross-reference mapping, which means juxtapose spatial and non-spatial information.

And I’d like to make a distinction between visual thinking and visual communication. What I often see is that people are focused on visual communication with these tools, in order to get the idea across to others, but they hardly use them for visual thinking. However, a sketch for thinking can be very illegible. In this process you interact with the medium in order to gain new insight: seeing the unseen. After you gain the understanding, the next step is to convey the gained knowledge in a pretty way that you can get it across to others.

LAJ: Could you please show me an integrated research & design project which is about regional landscapes?

Nijhuis:Research and design comes together in landscape-based regional design or landscapebased urbanism. We apply this approach in research we do for governments and other stakeholders in Europe, USA and China. For example in recent years we started a NSFC, NWO, and the EPSRC Joint Research Project: “Adaptive Urban Transformation(AUT) - Territorial governance, spatial strategy and urban landscape dynamics in the Pearl River Delta”,A collaboration between TU Delft (Dr. Steffen Nijhuis), South China University of Technology(Dr. Sun Yimin) and The University of Scheffield(Dr. Eckart Lange). AUT employs landscapebased regional design methods as an integrative and multiscale design and planning approach (Fig. 9).Doing so can steer urban and rural transformative processes through a combination of sector activities aimed towards more coordinated sustainable outcomes. Landscape-based regional design is considered to be an important strategy that shapes the physical form of regions by using landscape as the basic condition to generate sustainable urbanized deltas. It also provides ways for longterm sustainable urban landscape development through both spatial planning and design. In summary, regional design is a transdisciplinary effort that not only safeguards sustainable and coherent development, but also guides and shapes changes that are brought about by socio-economic and environmental processes, while establishing local identity through tangible relationships to a region.More info can be found in “Towards a Landscapebased Regional Design Approach for Adaptive Transformation in Urbanizing Deltas”[8].

For instance in the Shunde district the mulberry-fishpond system is an important feature of the landscape. Nowadays many of these structures are superimposed with urbanization, just filling the site and built on it. As a result problems occur with water storage, ecology and the foundations of the housing. Therefore we propose a landscapebased approach that takes the water system and landscape structures as a basis for water sensitive and contextual urbanization model in which smart mobility concepts play also a crucial role.

LAJ：Thank you very much for joining this interview. Last but not least, you’ve come to China many times these years and you’ve experienced many joint research projects around the world, in your perspective, is there maybe any particularity in the research &design of regional landscapes in China?

Nijhuis:In China you have great traditions in analyzing cities and cultural landscapes and their relationship with the natural environment as exemplified by the work of renowned colleagues from universities across China. You also show great vigor in building new cities and landscapes.

I think by applying the lessons learned, and design principles that can be drawn from the proven practices of before have a great potential to be utilized for the contemporary challenges related to urbanization and climate change. I think also the planning and design of new urban landscapes and transformation of existing ones in Chinese context will benefit from a landscape-based regional design approach. This approach takes the landscape as the basis for urbanization, goes through the scales,is interdisciplinary and knowledge-based. Such an approach exploits the interaction between planning concepts and designerly knowledge as a means to deal with the complexities of creating future proof and thriving urban landscapes. Thank you very much and I am looking forward to learn more and exchange knowledge and inspiration.

Sources of Figures:

Fig. 1 ? map from private collection; Fig. 2 ? Palmboom;Fig. 3 ? Steenbergen & Reh; Fig. 4-6 ? Steffen Nijhuis;Fig. 7 ? Steffen Nijhuis, et al; Fig. 8 ? Steffen Nijhuis,based on SRTM (30 m), Hydropolis; Fig. 9 ? Sun Chuanzhi.