Linking nutrient resorption stoichiometry with plant growth under long-term nitrogen addition

Increased nitrogen(N)input can potentially lead to secondary phosphorus(P)limitation;however,it remains unclear whether differences in the plant's ability to cope with this P deficiency are related to their growth responses.Using a long-term experiment of N addition in a boreal forest,we explored the potential role of plant nutrient resorption efficiency and its stoichiometry in mediating plant growth responses to increased N input.We recorded the cover and measured the concentration and resorption efficiency of leaf N and P as well as the photosynthesis of a grass Deyeuxia angustifolia and a shrub Vaccinium vitis-idaea.The cover of the grass D.angustifolia increased with increasing N addition,while that of the shrub V.vitis-idaea decreased with N addition rate and almost disappeared from the high-level N addition over time.P resorption efficiency(PRE)increased in D.angustifolia but decreased in V.vitis-idaea with increasing leaf N:P which was increased by N addition for both species.In addition,photosynthesis increased linearly with N resorption efficiency(NRE)and PRE but was better explained by NRE:PRE,changing nonlinearly with the ratio in a hump-shaped trend.Furthermore,the variance(CV)of NRE:PRE for V.vitis-idaea(123%)was considerably higher than that for D.angustifolia(29%),indicating a more stable nutrient resorption stoichiometry of the grass.Taken together,these results highlight that efficient P acquisition and use strategy through nutrient resorption processes could be a pivotal underlying mechanism driving plant growth and community composition shifts under N enrichment.

江西常绿阔叶林分类及其分布格局

常绿阔叶林是中国东部的演替顶极植被。江西省位于中国亚热带东部湿润常绿阔叶林区的腹心,拥有该区第二大面积的常绿阔叶林。尽管江西常绿阔叶林已有大量的研究,但其主要类型和分布格局与主要驱动因素仍不清楚。本研究采用植物群落清查法调查了205个常绿阔叶林样方,样方包含了江西常绿阔叶林主要群系类型。采用双向指示种(TWINSPAN)和去趋势对应分析(DCA)进行了数量分类,确定各群落类型的标志种、指示种。使用Kruskal–Wallis非参数检验比较了不同群落类型的地形和气候差异,并采用冗余分析(RDA)和变差分解的方法,区分地形、气候以及样方间距离对不同类型群落分布的影响。研究结果表明,江西常绿阔叶林可分为丘陵常绿阔叶林、沟谷季雨林、山地典型常绿阔叶林和山地云雾常绿阔叶林。这4类常绿阔叶林的分布受地形和气候条件的驱动,其分布格局也反映了这4类常绿阔叶林类型间的生物地理差异。本研究为构建中国东部湿润区常绿阔叶林的高级分类系统提供基础资料和重要参考。同时,本研究还发现丘陵常绿阔叶林易受人为活动干扰,沟谷季雨林和山地典型常绿阔叶林拥有非常丰富的第三纪孑遗植物等,这些特征都暗示江西常绿阔叶林亟需开展更严格的保护实践。

我国草地面积有多大?

草地是不可或缺的自然资源,但关于我国草地面积到底有多大存在很大争议.本文主要利用植被分布与降水之间的相关关系来探讨我国草地的分布和面积.归一化植被指数(NDVI)数据能够很好地反映植被的覆盖状况,与降水之间也存在良好的相关关系,因此可以通过建立已知草原地区的降水与NDVI的关系来反演草地的分布和面积.利用此方法及与遥感数据相匹配的过去30年(1982~2011年)的平均降水量数据,估算得到我国的草地总面积约为293×104 km2.将研究期间的降水数据每5年求其平均,估算得到我国草地面积的年际变化并不显著,变动于290×104~295×104 km2.

Terrestrial carbon sinks in China and around the world and their contribution to carbon neutrality

Enhancing the terrestrial ecosystem carbon sink(referred to as terrestrial C sink) is an important way to slow down the continuous increase in atmospheric carbon dioxide(CO_(2)) concentration and to achieve carbon neutrality target.To better understand the characteristics of terrestrial C sinks and their contribution to carbon neutrality,this review summarizes major progress in terrestrial C budget researches during the past decades,clarifies spatial patterns and drivers of terrestrial C sources and sinks in China and around the world,and examines the role of terrestrial C sinks in achieving carbon neutrality target.According to recent studies,the global terrestrial C sink has been increasing from a source of (-0.2±0.9) Pg C yr^(-1)(1 Pg=1015g)in the 1960s to a sink of (1.9±1.1) Pg C yr^(-1) in the 2010s.By synthesizing the published data,we estimate terrestrial C sink of 0.20–0.25 Pg C yr^(-1) in China during the past decades,and predict it to be 0.15–0.52 Pg C yr^(-1) by 2060.The terrestrial C sinks are mainly located in the mid-and high latitudes of the Northern Hemisphere,while tropical regions act as a weak C sink or source.The C balance differs much among ecosystem types:forest is the major C sink;shrubland,wetland and farmland soil act as C sinks;and whether the grassland functions as C sink or source remains unclear.Desert might be a C sink,but the magnitude and the associated mechanisms are still controversial.Elevated atmospheric CO_(2) concentration,nitrogen deposition,climate change,and land cover change are the main drivers of terrestrial C sinks,while other factors such as fires and aerosols would also affect ecosystem C balance.The driving factors of terrestrial C sink differ among regions.Elevated CO_(2) concentration and climate change are major drivers of the C sinks in North America and Europe,while afforestation and ecological restoration are additionally important forcing factors of terrestrial C sinks in China.For future studies,we recommend the necessity for intensive and long-term ecosystem C monitoring over broad geographic scale to improve terrestrial biosphere models for accurately evaluating terrestrial C budget and its dynamics under various climate change and policy scenarios.

Family-level leaf nitrogen and phosphorus stoichiometry of global terrestrial plants

Leaf nitrogen(N) and phosphorus(P) concentrations are critical for photosynthesis, growth, reproduction and other ecological processes of plants. Previous studies on large-scale biogeographic patterns of leaf N and P stoichiometric relationships were mostly conducted using data pooled across taxa, while family/genus-level analyses are rarely reported. Here, we examined global patterns of family-specific leaf N and P stoichiometry using a global data set of 12,716 paired leaf N and P records which includes 204 families, 1,305 genera, and 3,420 species. After determining the minimum size of samples(i.e., 35 records), we analyzed leaf N and P concentrations, N:P ratios and N^P scaling relationships of plants for 62 families with 11,440 records. The numeric values of leaf N and P stoichiometry varied significantly across families and showed diverse trends along gradients of mean annual temperature(MAT) and mean annual precipitation(MAP). The leaf N and P concentrations and N:P ratios of 62 families ranged from 6.11 to 30.30 mg g–1, 0.27 to 2.17 mg g–1, and 10.20 to 35.40, respectively. Approximately 1/3–1/2 of the families(22–35 of 62) showed a decrease in leaf N and P concentrations and N:P ratios with increasing MAT or MAP, while the remainder either did not show a significant trend or presented the opposite pattern. Family-specific leaf N^P scaling exponents did not converge to a certain empirical value, with a range of 0.307–0.991 for 54 out of 62 families which indicated a significant N^P scaling relationship. Our results for the first time revealed large variation in the family-level leaf N and P stoichiometry of global terrestrial plants and that the stoichiometric relationships for at least one-third of the families were not consistent with the global trends reported previously. The numeric values of the family-specific leaf N and P stoichiometry documented in the current study provide critical synthetic parameters for biogeographic modeling and for further studies on the physiological and ecological mechanisms underlying the nutrient use strategies of plants from different phylogenetic taxa.

Yield and quality properties of silage maize and their influencing factors in China

Silage maize(Zea mays L.) is one of the most important forages in the world, and its yield and quality properties are critical parameters for livestock production and assessment of forage values. However, relationships between its yield and quality properties and the controlling factors are not well documented. In this study, we collected 5,663 observations from 196 publications across the country to identify the relationships between yield and quality properties of silage maize and to assess the impact of management practices and climatic factors on its yield and quality in China. The average dry matter yield of silage maize was(19.98±6.93) Mg ha^(-1), and the average value of crude protein, ether extract, crude ash, crude fiber, acid detergent fiber, neutral detergent fiber, nitrogen-free extract, and relative feed value was 7.86%±1.71%, 2.53%±1.01%, 5.05%±1.66%,23.97%±6.34%, 27.62%±7.12%, 51.60%±9.85%, 59.68%±7.72%, and 131.17±31.49, respectively. In general, its nutritive value decreased as its yield increased. Increasing planting density could increase the yield but inhibit the nutritive values, while increasing fertilization could benefit the nutritive values. Geographically, the yield increased and the nutritive value decreased from warm(south) to cold(north) regions. The length of growth duration was a major controlling factor for the patterns of these properties. Our findings provide insights for police-makers to make strategy for achieving high yield and good quality of silage maize and help local people to implement better management practices.

HTDet:A Clustering Method Using Information Entropy for Hardware Trojan Detection

Hardware Trojans(HTs)have drawn increasing attention in both academia and industry because of their significant potential threat.In this paper,we propose HTDet,a novel HT detection method using information entropybased clustering.To maintain high concealment,HTs are usually inserted in the regions with low controllability and low observability,which will result in that Trojan logics have extremely low transitions during the simulation.This implies that the regions with the low transitions will provide much more abundant and more important information for HT detection.The HTDet applies information theory technology and a density-based clustering algorithm called Density-Based Spatial Clustering of Applications with Noise(DBSCAN)to detect all suspicious Trojan logics in the circuit under detection.The DBSCAN is an unsupervised learning algorithm,that can improve the applicability of HTDet.In addition,we develop a heuristic test pattern generation method using mutual information to increase the transitions of suspicious Trojan logics.Experiments on circuit benchmarks demonstrate the effectiveness of HTDet.

Patterns of plant diversity at high altitudes on the Qinghai-Tibetan Plateau

To describe the biodiversity patterns of plants along an altitudinal gradient on the Qinghai-Tibetan Plateau and to clarify the bias in plant specimen records at high altitude.Methods We conducted a large-scale investigation of vegetation at a wide range of altitudes,focusing on a high-altitudinal range(3200-5200 m)at different locations on the Qinghai-Tibetan Plateau.We then compared the altitudinal distribution of plant species obtained from our field investigation with that in plant specimen records from published sources and an online database.Important Findings Our data provide evidence that altitude plays a large role in regulating species composition on the Qinghai-Tibetan Plateau.We could not,however,detect a clear relationship between altitude and species richness,although a weak monotonically increasing trend of richness was detected with increasing altitude.According to specimen records,most species have been sampled at a wide range of altitudes,and the average range of 145 species is>2000 m.Despite this wide range,more than half of the species we observed were at higher altitudes than the specimen records indicate.High-altitude areas have probably been so poorly sampled that only a small fraction of the resident species has been recorded.This study clearly shows the regional bias of specimen records in the Qinghai-Tibetan Plateau.

Altitudinal variation of ecosystem CO_(2) fluxes in an alpine grassland from 3600 to 4200 m

Aims Recent studies have recognized the alpine grasslands on the Qinghai-Tibetan plateau as a significant sink for atmospheric CO_(2).The carbon-sink strength may differ among grassland ecosystems at various altitudes because of contrasting biotic and physical environments.This study aims(i)to clarify the altitudinal pattern of ecosystem CO_(2) fluxes,including gross primary production(GPP),daytime ecosystem respiration(Redaytime)and net ecosystem production(NEP),during the period with peak above-ground biomass;and(ii)to elucidate the effects of biotic and abiotic factors on the altitudinal variation of ecosystem CO_(2) fluxes.Methods Ecosystem CO_(2) fluxes and abiotic and biotic environmental factors were measured in an alpine grassland at four altitudes from 3600 to 4200 m along a slope of the Qilian Mountains on the northwestern Qinghai-Tibetan Plateau during the growing season of 2007.We used a closed-chamber method combined with shade screens and an opaque cloth to measure several carbon fluxes,GPP,Redaytime and NEP,and factors,light-response curve for GPP and temperature sensitivity of Redaytime.Above-and below-ground biomasses and soil C and N contents at each measurement point were also measured.Important Findings(i)Altitudinal pattern of ecosystem CO_(2) fluxes:The maximum net ecosystem CO_(2) flux(NEPmax),i.e.the potential ecosystem CO_(2) sink strength,was markedly different among the four altitudes.NEPmax was higher at the highest and lowest sites,ap proximately
7.460.9 and
6.760.6 lmol CO_(2) m^(-2)s^(-1)(mean 6 standard error),respectively,but smaller at the intermediate altitude sites(3800 and 4000 m).The altitudinal pattern of maximum gross primary production was similar to that of NEPmax.The Redaytime,however,was significantly higher at the lowest altitude(3.460.3 lmol CO_(2) m^(-2)s^(-1))than at the other three altitudes.(ii)Altitudinal variation of vegetation biomass:The aboveground biomass was higher at the highest altitude(154627 g DW m
2)than at the other altitudes,which we attribute mainly to the large biomass in cushion plants at the highest altitude.The small above-ground biomass at the lower altitudes was probably due to heavy grazing during the growing season.(iii)Features of ecosystem CO_(2) fluxes:Redaytime and GPP were positively correlated with above-ground biomass.The low ratio of Redaytime to GPP at either the measurement point or the site level suggests that CO_(2) uptake efficiency tends to be higher at higher altitudes,which indicates a high potential sink strength for atmospheric CO_(2) despite the low temperature at high altitudes.The results suggest that the effect of grazing intensity on ecosystem carbon dynamics,partly by decreasing vegetation biomass,should be clarified further.

Characterization of CO_(2) flux in three Kobresia meadows differing in dominant species

Aims Kobresia meadows,the dominant species of which differ in different habitats,cover a large area of alpine grassland on the QinghaiTibetan Plateau and act as potential CO_(2) sinks.Kobresia meadows with different dominant species may differ in carbon sink strength.We aimed to test the hypothesis and to clarify the differences in CO_(2) sink strength among three major Kobresia meadows on the plateau and the mechanisms underlying these differences.Methods We measured the net ecosystem exchange flux(NEE),ecosystem respiration flux(ER),aboveground biomass(AGB)and environmental variables in three Kobresia meadows,dominated by K.pygmaea,K.humilis,or K.tibetica,respectively,in Haibei,Qinghai.NEE and ER were measured by a closed-chamber method.Environmental variables,including photosynthetic photon flux density(PPFD),air and soil temperature and air and soil moisture,were monitored during the above flux measurements.Important findings The measured peak AGB increased with soil water content and was 365,402 and 434 g dry weight m
2 for K.pygmaea,K.humilis and K.tibetica meadow,respectively.From the maximum ecosystem photosynthetic rate in relation to PPFD measured during the growing season,we estimated gross ecosystem photosynthetic potential(GEPmax)as 22.2,29.9 and 37.8 lmol CO_(2) m
2 s
1 for K.pygmaea,K.humilis and K.tibetica meadow,respectively.We estimated the respective gross primary production(GPP)values as 799,1063 and 1158 g C m^(-2) year^(-1) and ER as 722,914 and 1011 g C m^(-2) year^(-1).Average net ecosystem production(NEP)was estimated to be 76.9,149.4 and 147.6 g C m^(-2) year^(-1) in K.pygmaea,K.humilis and K.tibetica meadows,respectively.The results indicate that(i)the three meadows were CO_(2) sinks during the study period and(ii)Kobresia meadows dominated by different species can differ considerably in carbon sink strength even under the same climatic conditions,which suggests the importance of characterizing spatial heterogeneity of carbon dynamics in the future.

Species richness and composition of shrub-encroached grasslands in relation to environmental factors in northern China

Aims Shrub encroachment has taken place in many of China’s northern grasslands.This study attempts to answer the following questions:which plant communities are present in these shrub-encroached grasslands(SEGs)?What are the species richness and composition of these communities?Which environmental factors determine the spatial distribution thereof?Methods We investigated the community characteristics of 255 SEG plots with a size of 20 m×20 m at 69 locations across grasslands in northern China.In each plot,paired 1-m^(2) quadrats were estab-lished within shrub patches and the neighbouring grassy matrix to record herb species composition.The quantitative character-istics(abundance,coverage and height)of herbs were meas-ured in 0.25-m2 subquadrats,and soil samples were collected in the quadrats.Two-way indicator species analysis(TWINSPAN),detrended canonical analysis(DCA)and canonical correspond-ence analysis(CCA)were used for community clustering and ordination.Important Findings The DCA and TWINSPAN results suggested that the SEGs in northern China can be divided into six community types:Potentilla fruticosa+Carex atrofusca,Spiraea hypericifolia+Festuca ovina,Caragana acanthophylla+Stipa sareptana,Caragana microphylla+Leymus chinensis,Caragana microphylla+Stipa klemenzii and Caragana tibetica+Cleistogenes songorica.At a regional scale,climate and soil nutrients controlled the spatial patterns of species richness and community composition of the SEGs in northern China.Temperature exerted a negative impact,whereas precipitation and nutrients had positive effects on species richness.Among the environmental factors used,climate was the major controller of the variations in commu-nity structure.These results provide new insights into the community composition of SEGs in China and enrich the global dataset of SEGs.

Leaf photosynthesis and simulated carbon budget of Gentiana straminea from a decade-long warming experiment

Aims Alpine ecosystems may experience larger temperature increases due to global warming as compared with lowland ecosystems.Information on physiological adjustment of alpine plants to temperature changes can provide insights into our understanding how these plants are responding to current and future warming.We tested the hypothesis that alpine plants would exhibit acclimation in photosynthesis and respiration under long-term elevated temperature,and the acclimation may relatively increase leaf carbon gain under warming conditions.Methods Open-top chambers(OTCs)were set up for a period of 11 years to artificially increase the temperature in an alpine meadow ecosystem.We measured leaf photosynthesis and dark respiration under different light,temperature and ambient CO_(2)concentrations for Gentiana straminea,a species widely distributed on the Tibetan Plateau.Maximum rates of the photosynthetic electron transport(J_(max)),RuBP carboxylation(V_(cmax))and temperature sensitivity of respiration Q10 were obtained from the measurements.We further estimated the leaf carbon budget of G.straminea using the physiological parameters and environmental variables obtained in the study.Important findings 1)The OTCs consistently elevated the daily mean air temperature by;1.6℃and soil temperature by;0.5℃during the growing season.2)Despite the small difference in the temperature environment,there was strong tendency in the temperature acclimation of photosynthesis.The estimated temperature optimum of light-saturated photosynthetic CO_(2)uptake(A_(max))shifted;1℃higher from the plants under the ambient regime to those under the OTCs warming regime,and the Amax was significantly lower in the warming-acclimated leaves than the leaves outside the OTCs.3)Temperature acclimation of respiration was large and significant:the dark respiration rates of leaves developed in the warming regime were significantly lower than leaves from the ambient environments.4)The simulated net leaf carbon gain was significantly lower in the in situ leaves under the OTCs warming regime than under the ambient open regime.However,in comparison with the assumed non-acclimation leaves,the in situ warmingacclimated leaves exhibited significantly higher net leaf carbon gain.5)The results suggest that there was a strong and significant temperature acclimation in physiology of G.straminea in response to long-term warming,and the physiological acclimation can reduce the decrease of leaf carbon gain,i.e.increase relatively leaf carbon gain under the warming condition in the alpine species.

Grazing alters warming effects on leaf photosynthesis and respiration in Gentiana straminea,an alpine forb species

Aims Vast grasslands on the Tibetan Plateau are almost all under livestock grazing.It is unclear,however,what is the role that the grazing will play in carbon cycle of the grassland under future climate warming.We found in our previous study that experimental warming can shift the optimum temperature of saturated photosynthetic rate into higher temperature in alpine plants.In this study,we proposed and tested the hypothesis that livestock grazing would alter the warming effect on photosynthetic and respiration through changing physical environments of grassland plants.Methods Experimental warming was carried by using an infrared heating system to increase the air temperature by 1.2 and 1.7℃ during the day and night,respectively.The warming and ambient temperature treatments were crossed over to the two grazing treatments,grazing and un-grazed treatments,respectively.To assess the effects of grazing and warming,we examined photosynthesis,dark respiration,maximum rates of the photosynthetic electron transport(J_(max)),RuBP carboxylation(V_(cmax))and temperature sensitivity of respiration Q_(10) in Gentiana straminea,an alpine species widely distributed on the Tibetan grassland.Leaf morphological and chemical properties were also examined to understand the physiological responses.Important findings 1)Light-saturated photosynthetic rate(A_(max))of G.straminea showed similar temperature optimum at around 16℃ in plants from all experimental conditions.Experimental warming increased A_(max) at all measuring temperatures from 10 to 25℃,but the positive effect of the warming occurred only in plants grown under the un-grazed conditions.Under the same measuring temperature,A_(max) was significantly higher in plants from the grazed than the un-grazed condition.2)There was significant crossing effect of warming and grazing on the temperature sensitivity(Q_(10))of leaf dark respiration.Under the un-grazed condition,plants from the warming treatment showed lower respiration rate but similar Q_(10) in comparison with plants from the ambient temperature treatment.However,under the grazed condition Q_(10) was significantly lower in plants from the warming than the ambient treatment.3)The results indicate that livestock grazing can alter the warming effects on leaf photosynthesis and temperature sensitivity of leaf dark respiration through changing physical environment of the grassland plants.The study suggests for the first time that grazing effects should be taken into account in predicting global warming effects on photosynthesis and respiration of plants in those grasslands with livestock grazing.

Resorption efficiency of leaf nutrients in woody plants on Mt. Dongling of Beijing, North China

Aims To explore resorption efficiency of nitrogen(NrE)and phosphorus(PrE)of woody plants in relation to soil nutrient availability,climate and evolutionary history,in North China.Methods We measured concentrations of nitrogen([N])and phosphorus([P])in both full expanded mature green and senescent leaves of the same individuals for 88 woody species from 10 sites of mt.Dongling,beijing,China.We built a phylogenetic tree for all these species and compared NrE and PrE among life forms(trees,shrubs and woody lianas)and between functional groups(N-fixers and non-N-fixers).We then explored patterns of NrE and PrE along gradients of mean annual temperature(MAT),soil inorganic N and available P,and phylogeny using a general linear model.Important Findingsmass-based NrE(NrEm)and PrE(PrEm)averaged 57.4 and 61.4%,respectively,with no significant difference among life forms or functional groups.Neither NrEm nor PrEm exhibited significant phylogenetic signals,indicating that NrEm and PrEm were not phylogenetically conserved.NrEm was not related to[N]in green leaves;PrEm was positively correlated with[P]in green leaves;however,this relationship disappeared for different groups.NrEm decreased with[N]in senescent leaves,PrEm decreased with[P]in senescent leaves,for all species combined and for trees and shrubs.NrEm decreased with soil inorganic N for all species and for shrubs;PrEm did not exhibit a significant trend with soil available P for all species or for different plant groups.Neither NrEm nor PrEm was significantly related to MAT for overall species and for species of different groups.

Phylogeography of Potentilla fruticosa,an alpine shrub on the Qinghai-Tibetan Plateau

Aims Our objectives were(i)to elucidate the phylogeography of chloroplast DNA(cpDNA)in Potentilla fruticosa in relation to Quaternary climate change and postglacial colonization,(ii)to infer historical population range expansion using mismatch distribution analyses and(iii)to locate the refugia of this alpine species on the Qinghai-Tibetan plateau during glacial-interglacial periods.Methods Potentilla fruticosa is a widespread species distributed on the Qinghai-Tibetan Plateau.We sampled leaves of P.fruticosa from 10 locations along a route of;1300 km from the northeastern plateau(Haibei,Qinghai)to the southern plateau(Dangxiong,Tibet).We examined the cpDNA of 15 haplotypes for 87 individuals from the 10 populations based on the sequence data from;1000 base pairs of the trnS-trnG and rpl20-rps12.Phylogenetic relationship of haplotypes was analyzed using the Phylip software package and the program TCS.The diversity of populations indices was obtained using the program ARLEQUIN.Important Findings With the limited samples,we found that(i)higher nucleotide diversity often occurs in high-altitude populations,(ii)the ancestral haplotypes distribute in the populations with higher nucleotide diversity than recent haplotypes,(iii)the expansion time of population in the high altitudes was estimated to be approximately at 52-25 ka BP(1000 years Before Present,where"Present"is AD 1950)and that in the low altitudes to be;5.1-2.5 ka BPand(iv)the source location of P.fruticosa is at the high altitudes,which might provide refugia for the species during the interglacial warm periods.The species expanded from the high-elevated locations on the Tanggula Mountains during the Holocene.

Achieving data-driven actionability by combining learning and planning

A main focus of machine learning research has been improving the generalization accuracy and efficiency of prediction models. However, what emerges as missing in many applications is actionability, i.e., the ability to turn prediction results into actions. Existing effort in deriving such actionable knowledge is few and limited to simple action models while in many real applications those models are often more complex and harder to extract an optimal solution. In this paper, we propose a novel approach that achieves actionability by combining learning with planning, two core areas of AI. In particular, we propose a framework to extract actionable knowledge from random forest, one of the most widely used and best off-the-shelf classifiers. We formulate the actionability problem to a sub-optimal action planning （SOAP） problem, which is to find a plan to alter certain features of a given input so that the random forest would yield a desirable output, while minimizing the total costs of actions. Technically, the SOAP problem is formulated in the SAS＋ planning formalism, and solved using a Max-SAT based ap- proach. Our experimental results demonstrate the effectiveness and efficiency of the proposed approach on a personal credit dataset and other benchmarks. Our work represents a new application of automated planning on an emerging and challenging machine learning paradigm.